Jump to:

NYSA CTR 1

The Council _[;'Og Tobacco I:_esearch - U.S.a.., Inco Su_ortino _,Iomedical Inv_$'_ioation (212) 42

Date: 22 Dec 1994
Length: 129 pages

Jump To Images
nysa_ctr1 40023605-40023733

Abstract

Application number 3375 entitled as above was considered in April 1992 and funded for a 3 year project. The present continuation application No. 3375A requests 3 more years of support.

Fields

Named Organization
American Heart Association (Voluntary health organization that focuses on cardiac health)
Voluntary health organization that focuses on cardiac health and stroke. AHA occasionally teams with tobacco retailers to engage in promotions/fund-raisers (see http://www.smokefree.net/doc-alert/messages/247136.html and http://www.rawbw.com/~jpk/stand/Pictures.html).
Biogen
Boehringer, Mannheim
Council for Tobacco Research - USA (CTR) (Formerly Tobacco Industry Research Committee (TIRC))
Originally organized as the Tobacco Industry Research Committe(TIRC) in 1954, and renamed Council for Tobacco Research - USA, Inc. (CTR) in 1964.
DuPont
Eastman Kodak Co. (Kodak) (Cigarette filter mfg from 1950s to 1994.)
Manufacturers of quality control equipment for cigarette packaging
Kodak
Lancet
Louisiana State University
McGraw-Hill
National Institutes of Health (NIH)
Plenum Press
Purdue University
Research Council
Rollins College
Scripps Research Institute
Tel Aviv University
United States Public Health Service (Headed by the Surgeon General)
United States Public Health Service is headed by Surgeon General of the United States.
University Medical Center
*University of California (use specific branch)
University of California Los Angeles (UCLA)
University of Kentucky
University of Toronto
Veterans Administration
Washington State University
Named Person
Allen, Susan
Black, Audrey
Browning, Jeffrey
Curtiss, Linda K.
Holt, Von
*Jordan, W. A. (use Jordan, William A.)
Defense
King, Amy
Lee, Winnie
Mayers, Anna
Scripps, Tim
Stefanski, Eva
Swanson, Mark
Van, Dr.
Watkins, Bruce
Williams, C. A. (RJR)
Date Loaded
11 Jan 2006
Box
0017

Document Images

Text Control

Highlight Text:

OCR Text Alignment:

Image Control

Image Rotation:

Image Size:

Page 1: 40023605
THE COUNCIL ~[;'Og TOBACCO I:~ESEARCH - U.S.A.., INCo SU~ORTINO ~,iOMEDICAL INV~$'~IOATION (212) 42 L-S$85 Date: Decembez 22, 1994 Number:. 3375A Discipline: Cardiovascular CONTINUATION RESEARCH APPLICATION for the Spring 1995 SAB Meeting To: Primary R~viewer: Dr. Bowd~n SecondmT Reviewer: Dr. Swain Applicant: Institution: Location: Project Title: Frederick C. de Beer, M.D. University of Kentucky lexington, KY Function of Serum Amyloid A Protein (SAA) HISTORY: Application number 3375 entitled as above was considered in April 1992 and funded for a 3 year project. The present continuation application No. 3375A requests 3 more years of support. REQUEST: Application Number 3375A requests $95,416 for the first year of a 3 year project. Budgot estimates for tho second and third year are $99,232 and $103,201, respectively. DOCUMENTS SUBMITrED: I. Application dated: 11130/94 2. Biographical sketches: Drs. de Beer (2 pgs.), Lusis (2 pgs.) and Banka (2 pgs.) 3. Reprints: :2 - both acknowledge CTR 4. Manuscripts: 3 - all acknowledge CIR. 5. Letters of collaboration and support: 6 lettcrs 6. Other documents: Abstract of Progress Report (1 pg.) Progress l~port Covering Period 711192 - 1130194. (5 pgs.) Institutional Review Board Approval Letter (1 pg.) GAH/mm 40023605
Page 2: 40023606
THE COUNCIL FOR TOBACCO RESEARCH - U.S.A.., INC. ~UPPORTI~O BIOIMED[CAL IlffVESTIOATION (212) 421-3S85 Date: December 22, 1994 Number: 3375A Discipline: Cardiavascutar CONTINUATION RESEARCH APPLICATION for the Spring 1995 SAB Meeting To: Primary Reviewer: Dr. Bowden Secondary Reviewer: Dr. Swain Applicant: Institution: Location: l~jvct Title: Frederick C. de Beer, M.D. University of Kentucky Lexington, Function of Serum Amyloid A Protein (SAA) HISTORY: Application number 3375 entitled as above was considered in April 1992 and funded for a 3 year project. The present continuation application No. 3375A requests 3 more years of support. REQUEST: Application Number 3375A requests $9:5,416 for the first year of a 3 year project. Budget estimates for the second and third year axe $99,232 and $103,201, respectively. DOCUMENTS SUBMITTED: 1. Application dated: 11/30/94 2. Biographical sketches: Drs. de Beer (2 pgs.), Lusis (2 pgs.) and Banka (2 pgs.) 3. Reprints: 2 - both acknowledge CTR 4. Manuscripts: 3 - all acknowledge CTR 5. Letters of collaboration and support: 6 letters 6. Other documexas: Abstract of Progress Report (1 pg.) Progreas Report Covering Period 711192 - 1/30/94 (5 pgs.) Institutional Review Board Approval Letter (1 pg.) GAHlram 40023606
Page 3: 40023607
Competing Renewal Applicatio CURRENT DATE: 11/30/94 Function of Serum Amyloid A.Protein (SAA) KEY WOROS (uM~r To F~V~) HDL, SAA A.k.her.osclerosis, Phospholipase .A~ PROPOSED START DATE 07/01/95 PROPOSED FUPJDS REQUESTED [I~DLUD~NG INDIRECT COSTS| DURATION YEAR 1 YEAR 2 YEAR 3 (YRS) 1,2oR3 TOTALREOUEST [pOWS): ~ 95,416 ~ 99,232 $ 103,201 PRINCIPAL INVESTIGATOR (do not: indi~,~o any co-P.I, on th~s psge) N.MV~ (La,$T, RR~T MJ.) AND DEGREES: De Beer, Frederick C., M.D. ACAO~MIC OR R~OFESaor~L~I]U~: Professor of Medicine OEPARTI~.NT; Internal Medicine iN~TITErtIoN: Universi%y of Kentucky Division of Rheumatology J515, KY Clinic C~TY. STATE [OR COUNTRY) ZIP ~OD~: Lexington, Kentucky 40536-0284 M~IUNG AaDF(S$ IF D~FE~ENT FROM ABOVE: ¢B.EPHON~UM~ER: 606 ) 323-6700 INSTITUTIONAL OFFICIAL |acceptin,~ for th~ insmution) NAME AND POSITION l'lTEE: JacE Supplee, Associate Directo~ 201Kinkead Hall Lexington, KY 40506-0057 1ELE~OHOKE 606) 257-9420 FINANCIAL ADMINISTRATOR (person to contact for budget information) R.M. Slivers ~anaEer ~pon~ored Projects Accounting M~UNG AOOP,£SS: .:~.~iversity of Kentucky 337 Peterson Semite Building Lex~gUon, ~Y 40506-0005 . I~ONE 606-257-3662 ~K~CHECKSPAYA~LET0: University of Kentucky Research Foundation t~[LC~CKSTOI~MEANO~5~TI~RtE): Mr. Robert Slivers, Manager Sponsored Projects Acct. ~I~G A~: University of Kentucky Rm. 337, Peterson Service Bldg Lexington, KY 40506 40023607
Page 4: 40023608
De Beer, Frederick C. University of Kentucky e-.Tttl=ofl~0j¢~: Function of Serum Amyloid A Probein (SAA) d. Ctearly summarize BACKGKOLrND, WORKING H~?OTHESIS and BROAD GOALS of tk.- p:oj,~ct. %Ve hypoth~z% based on preliminary data, that the function of acute-phase SAA is to act as a co- factor for secretory non-pancreatic phospholipase Az (sPLA2). This holds the potential to alter the direction of HDL lipid transport toward sites of inflammation. We suggest that this is in the short- term a bcneficia~proccss (the delivery of~ipids to cells in need during inflammatio-) but when aberranfly induced by oxidized lipids, particularly in the arterial wall, promotes foam cell d~veloprnent and a~hm-ogencsis. High density lipoprotcin (HDL) can be viewed as central in lipopmt~in metabolism. During inflammation HDL undergoes marked ramodcling when acute-phase SAA increases hundreds of fold, displacing apolipoprot~in AI (apo-A/). A concomitant event during inflammation is the induction of another acute.phase protein, narncly, secretory non-pancreatic phospholipase A~ (sPLA~). Th/s ts induced by cytokines and secreted by a variety of mescnchymal cells including endothelial and vascular smooth muscle cells (1). It has not previously been studied in r~lafion ~o lipoprotein metabolism even though substantial evidence exists that phospholipases of this group akcrs HDL- mediated lipid delivery (2,3,4). This is cffected by hydrolysis of surface phospholipids that rcsuhs in a shiR in the equilibrium of fi~cc cholesterol betwccn HDL and the plasma mcmbranc with an increased net delive~ of frcc cholcstcrol io the cell by a surface transfvr process (2,3,4). An additional possibility is that lysophosphatidylcholine so gcncratcd can result in fusion of HDL and cell membranes resulting in more divcrsc lipid transfer cvcn including phospholipids. During inflammation, sPLA~ activity is present at sites of inflammation and the SAA on HDL promotes this activity. Lipid flow would be altered and the net effect would bc dclivcry of lipids to cells at inflammatoW sites (2,3). Why would sucl~ lipid ddivcry b¢ beneficial? We postulate that the answer resides either in damaged membrane repair, lipid nccds for cell division or mcmbran¢ replac¢mcnt particularly inphagocy~cs where large parts of cell membranes are engulfed in phagolysosomcs. The binding characteristics of SAA-bcaring HDL, augmenting its association with macrophagcs and neutrophils, fits our hypothesis that lipid would flow toward sites of inflammation and away from the liver. No~c how similar our proposed fi~nction for SAA is to that of ~e major apolipoprotein of I-IDL namely apo-AI. Apo-~ is also in essence a co-factor for a phospholipase in fl~at it activates Ie~i~in cholesterol acyltransferase (LCAT). The role o~ apo-AI in this a~tivation has been ascribed t~ the abi|ility of its amphipafl~c a-l~:lical regions to insert between r~ polar head groups ofphosphoHpids, mudcriug the ester linkage of the ~wo position fatty acid susceptible to the enz~ne action. ~e propose rh~t SAA dots exactly the same thing for sPLA~. The result'of LCAT a~tivation ultimately is th~ transl~ of quantities of esterifi~l cholesterol to the live. We hypothesize that the rcsuk of sPLA2 activation is th~ deliver7 of lipids to sites o~inflammation. The balance between the two major apolipopro~ius ofHDL, apo-AI and SAA, and two phospholipases thus have the potential to define lipid flow. Thes~ data rclat~ ~o the development of atheros¢lerosis during chronic infl~rnmato~ disease a~l pat~cularly to the inflammato~ events that o~cur in the dev~lopin~ athcrosclcrotic plaque. CONFI D E NTIA L 40023608
Page 5: 40023609
2. Evaluate the impact oft.he SAA/sPLA~ interactio~ on lipid flow from HDL to THP- 1 c~lls. This has reIevanc~ for macrophagc "foam cdl" format/ore 3. Evaluate the impact of the SAA, sPLA~ interaction in the mouse model of atherosclerosis using transgenic approaches. Two established transgenic mouse strains will be utilized namely SAA2 over-expressing and a human sPLA2 over- ex~essing model th~ following will be evaluated: a. How does the over-expression of sPLAz influence mouse lipoprot¢in profiles? Lipid and apolipoprotein detcrminatious will be performed on HDL, LDL and VLDL. b. Establisk whether the over-expression of sPLA2 alone promotes the development of atherosclarosis in the transgenic model. c. Cross the SAA transgenic mouse with the sPLAz transgenic. Establish how the concomitant expression of both enzyme and co-factor alters the lipoprotein profiles. d. Establish whether c,o-oxpression of SAA and sPLAz promotes development of athemsclerosis. 3. SUPPOR.TING DATA, EXPERIMENTAL DESIGNand PROCEDURES. Do not attach more than six (6) additional pages 0a=30. All fig~es, cha~, tables and roferenc, e~ must fit within pages 3 - 3f. Serum amvlo~Ao~ enhances the aetivl ofseeretor non- anereatl~ AII (sPLA~). Teleological considerations support the potential relevance of SAA/sPLA2 intoradtion, Both SAA and sPLA2 are rapidly induced during inflammatory events with mlhNA for both datectable at two hours and the resultant proteins similarly produced with respect to time. Whereas acute-phase SAA's arc practically only produced in the liver, sPLA2 is produced by a variety ofmeseachymal cells that would im present at sites of inflammation. The concentration ofsPLA2 can increase hundreds of fold in inflammatory fluids and in th~ circulation. Generally, there ate three types ofmamaliai PLA~'s based on the number and position ofcysteine residues. O~_gy the secretory group II phospholipases are relevant to this proposal. They arc similar to those ~'the crotalid and viperid snakes. The hydrolytic activity of sPLAa is critically dependent upon the surface pressure packing density of the phospholipid monolayer. As such, normal sell membranes are not a substrat¢ for this enzyme whereas HDL is. The influenc~ of PLA~ activity on HDL is important inunderstanding the function of~his panicle. During homeostasis, lhe influence of PLA2 activity on HDL has best been studied with respect to this activity of hepatic lipase. Given that this activity of hepatic lipase is fixed during homeostasis, the net result is cholesterol clearance from the endothelial surfaces of hepatic sinusoids for excretion, an important part of reverse cholesterol transport. During inflammation, sPLAz activity is present at such sites and the SAA on HDL would promote this activity. Lipid flow would bc altered and the net effect would be delivery of lipids to cells at inflammatory sites, Initial experiments of the SAA/sPLA2 interaction were conducted in collaboration with Drs. W. Pruzanski and P. Vadas, University of Toronto. The data supporting this contention is provided in an appended paper (Pruzanski, ctal, Biochcm. J., submitted). Our data indicate that purified, acute-phase SAA as well as SAA-bearing HDL markedly enhance the hydrolytic activity of sPLAz. Tiffs was initially established by using multi-lamel[ar liposomes composed of PC, PC:LysoPE. In contrast, normal HDL inhibited sPLAz activity. What we view as particularb, important is that the constitutive human SAA, does not enhance sPLAz activity." CA-& PPL.DOC I~ 13~q)) CONFIDENTIAL 40023609
Page 6: 40023610
Our studies ca the SAA sPLA~ interaction are onsoing. We are now ccrafming ourself to using recombiu~t proteins, rhS,~, and rlmPLA~ (gift from J. Bro-,va~ug, B[o~en), to evaluate the phylogcn=tic conservation of~is activation. Wc compared human rhSAA, to mouse SAA (containing equal amounts of a~u~c-phas~ SAA~, SAA~ i.~ot~Te_~) axtdhors~ S,~,~k using th= multil~mellar lipo~mc assay. All ~hxc~ of the.~e SAA's have identical tonsured regions bmveen amino acids 33 and 44. ]'he horse was chosen because of its nonapeptide insezt at position 59 which makes it somcwha~ differont from other acute- phase SAA's. These three SAA's all promoted the activity of human sPLA~ without significant differences detected. As these SAA's only share about 58% amino acid iden~ty outside of the conserved region, the latter ia likely r.he import~t area promoting the activity. One of th~ l~mitations of oar initial experiments is that we did no[ m~ HDL phospholipids as a substrate. Initially we employed the two well-established and widely-a~cepted assays for sPLA~ activity mmmly the use ofradiohbeled E. coli and liposom¢ assays. We arc collaborating with Drs. W. Prazanski and P. Vadas to develop and evaluate new assays either based on the measurement of free fatty acid release from HDL or d¢lipidation and reassembling HDL in the presence ofradiolabeled phospholipid. Initial investigations monitoring the liberation of free fatty acids from BDL and VLDL when exposed to rhsPLA~ with and without respective SAA's revealed valuable information. These expexime~ts am performed in 0.15 MNaCI, 20ram Tris/HCl, pH 7.4, 1% BSA (w/v), 8 rnM CaCh. Oa~ mi~ogram of sPLA: or heat-inactivated sPLA~ was incubated with various concentration of the respective SAA's. The release of free fatty a~ids is quantitated using the WAKO fatty acid determination kit (Biochemical Diagnostics). This determination involves coupling of the liberated fatty acid to acetyl-CoA and then subsequent coupling to a peroxide generating systex~ followed by a pexoxidv detexraination. Data generated this time indicate: 1) that at 90 minutes ~.7 times the amount of TFA is gcncxatcd from 1 mg HDL in the presence of 100 ~tg ofrbSAA~ when compared to controls with SAA omitted. 2) rhSAA~ does not enhance the libexation of FFA fi:om LDL. B) rhSAA~ enhances the ~bility of a PLA2 from Crotalus dufissus to liberaie FFA only by 19%. We are presently pursuing ~hcs~ experiments further. Oux preliminary data indicate that human sPLA~ has a g~eatcr specificity for substrates and is morn dependent on SAA to provide access to the fatty acid chain than the snake venom phospholipasc~. Specific Aim 1: Establish whether spLA~ is produced in human atherosclerotic lesions. Our findings tha~ acute-phase SAA promotes sPLA~ activity may be particularly impot2ant for the events leading to the development of the athems~lcroiic lesions in the vessel walls. The pathophysiologic processes which underlay this development have much in common with inflammation namely monocyte- endothelial adhesion, ~ans-endothelial migration, activation and production of inflammatory cytokines. The presence of acute-phase SAA--~mlceules in atheroselexotic lesions in mouse and man have been reported (5). In sire hybridization and immuno-cytochemical detection indicated the production of SAA in endothelial cells, macrophagc-derived "foaln cells" and vascular smooth muscle cells in such lesions. These are presumably induced by oxidized lipids present in the microenviroment. Vascular smooth muscle cells are a major source of sPLA2 synthesis. More importantly, sPLA2 have been immunochemically detected in athexosclexotic plaques. The enhancement ofsPLA2 activity by SAA at such sites would lead to lipid delivery by HDL and conceivably promom "foam cell" formation, a key event in athexogenesis. This would he in direct contrast to the suggestions by others, based on the ability of HDL to remove cholesterol from "£oam" cells, that HDL plays a beneficial role in atherosclcrofic plaques. Numerous proteins have been detected in atherosclvrotic lesions. In situ hybridizations is important to differentiam local production from plasma migraticm. The in situ hybridization studi~s we propose will precisely detect which cells produce sPLA2 and correlate this with SAA-producing cells. Co-production 40023610
Page 7: 40023611
ofsPLA2 and SAA in athcrosclcrotic lesions could lc~.,d to a re-definition of the role of HDL at ~uch site~. Methc~lolo~¢: I~NA prob~ will b~ Wanscribed from pGEM-7 plasmid (Promega) that contains a 120-bp s~quetm~ of the huma~ sPLA~ and SAA cDNAs. Tho place,aid will b~ lin~a~iz~d, and an~x~e and sense RNA v~I1 be transcn~bed with T7 or SP6 RNA polym~-ase 0'romcga) in the pres~ce of [~H]OTP (New England Nuct~xr). Probe will be added at 1 x 10~ epm to 100/~1 of hybridization buffer per slide and incubated ovcrn/ghL Slides will be exposed to Hype~na~ fix (Amersham) for 1-3 days and thc~ coated with NTB-2 liquid en~ulsio~ 0~astman Kodak) and exposed in the dark for 2-3 weeks. Slides will be developed in D-19, treated with Rapid Fix (Eastma~ Kodak), counterstnined with hcmatoxylin and cosin, and mounted with Histoclad. Sections of each ~issue ~ be kybddizcd in 2 or 3 separate cxper~ents. For combined immunocytochcm~stry and in situ hybr/dization, aH buffers will be prepared in water ~rcated with 0.1% DEPC. Sections will be reacted with primary rabbit antisera against sPLA~ and SAP, followed by the av/din-biotin peroxidase method (Vectastain, Vector Labs). Slides will be incubated with 0.05% 3-3' dian~obenzimide (DAB) to visualize the reaction product. After treatment with DAB, ~o sections will be washed in water and u'cated with proteinasc K for 10 rain, then washed/n DEPC-treated water and subjected to in situ hybridizations. Sections will be mounted and photographed. Specific Aim 2: Study implications of SAA/sPLAT~fnteraction on lipid exchange between Y~I)L and THP-1 eeH?. THP-1 cells were chosen because they h~ve been established as a good model for "foam ceil" formation. It overcomes the effects of donor variability that arises when human monoc3rtcs arc used. We Imve substantial experience in wolking with these cells. Under the conditions that we iutend to employ, Northern blot analyses established that these cells neither produce SAA or sPLA2. It is well- established that the phospholipid loss from the surface of HDL brought about by enzymatic hydrolysis result in the formation of a cheanical gradient with cholesterol partitioning from the lipoprotcin surface to the cell membrane. Wc intend to study the amplifying influence of SAA on this process. Because of differences between the fatty acid composition of normal and SAA-beming HDL that we are pres6ntly exploring in collaboration with Dr. Bruce Watkins of Purdue University, we will only use normal HDL~ in these studies. Increasing concentrations of recombinant human sPLA2 (flaPLA2) and recombinant hkman SAAt (rhSAAt) Hill be studied. Controls will omit rhSAAt. Studios ofcelMlar lipids will be done in collaboration with Dr. C.L. B~ka, S~ripps Research Institute. Total THP-1 cellular cholestea'ol and cholesterol cater will be measured after exposure to the respective HDL's. Phospholipid content of the cells Hill be monitored. This is important because ofsugge~tions that the lysophosphatidylcholino generated by tim aOion of phospholipases promotes a fusion of I~DL and veil meanbranes resulting in a transfer of both free and esterified cholesterol plus phospholipid. Total celhflar phospholipid will be measured and sp~ific phospholipid species Hill lm idmitified by thin-iayvr chromatography. To bring mine sensitivity and vvrsatility to our experiments, we will additionally use HDL where the free (unesterified cholesterol) will tm labeled with [4-t ~C] ~holesterol. Phospholipids will resp~tively be labeled on h~ui groups or fatty acid chains. In parallel experiments, t2~l-labeled HDL~ will be used to assess cellular association and lipoprotvin degradation. The data so generated would be valuable for more in-depth assessment of lipid flow. This is beyond the scope of the present proposal but include rates of cholesterol esterification, accumulation of cholest~ol ester mass, cholesterol synthesis and seeretion of almlipoprotcin E. Methodology: THP-I cell culture will be performed as in the appcndcd paper (Banka ct al, I. Lipid Rcs., in press). Cells will not bc used beyond 20 passages. Human recombinant SAAt (rhSAA~) is prepared as described in progress report. Human recombinant sPLA,_ is obtained from Dr. ~. Browuing (Biogcn). 40023611
Page 8: 40023612
I.~belinE of Ep.~ds and apopmteh~s. To ]abd ~ ~ (~=st¢~e~ [FC]) cholesterol oraL, [~C] chole~ol (60 mCffmmol, New ~d Nu:l~) u~l E, ~y~ ~ TLC ~d re-p~ ifreq~. ~L WIll b~ m~ohb~l~d ~ ~¢ chol¢~ol ~ d~bed (0- ~ys~ of~c ~L cholesterol ~II be by ~n-lwcr c~mato~a~hy. Pho~holfp£d ~L) on ~L~ ~ b~ labeled ~cnfi~ly ~ d~s~ (7). PL labcl~ on b~d ~oups ~-3-P~dyl ~-m~ylJ4C] c~I~c, 1,2~p~koyl) or fa~, acid, L- ~-I in~ba~on ~ pr~cssin~ of ~Hs ~ m~ N~ ~ w~l b~ ~ubated ~ ~- 1 cells ~ v~ous con~en~o~. In~ ~nts ofr~x, ~z or heat-activated sP~ ~ b~ c~ ~ubat~ for v~o~ ~c po~ up ~ 4 ho~ M 37°C. M~ ~]1 bc rcmov~ ~ cc~s, washc~ ~c~ ~ ~s~ ~sh~ ~ so~cat~ ~uots ~ ~ ~ for ~c fore,g: 0 ~ DNA ~ays: ~ c~le~ol p~ct~s ~ b~ ~~ to cc~ DNA c~t~t to acco~t for v~ab~. CclI~= DNA ~ ~ msas~ us~g a colo~e~c ~my. 1) to~l cell ~o~at~ ~cfi~; S) ~uti~ ofla~l bc~n ~ ~d c~ chol~t~ol ~ ~c cc~; %) ~ll~ chol~tcml ~ss; 0 ~ phosphol~id phosp~. A~0alytical Techniques: Cellular lipids will bc extracted by the procedure of Bligh and Dyer. Lipopmtein lipids will be extracted by the procedure ofFolclL Estcdfied and free cholcstcrol will bc scparatcd by thin-lay~ chromatography on siliczt gel 60A plates (Whatmau). Bands will be scraped and counted. Phospholipid species will be separated by thin-layer chromatography on silica gel G plates developed in clflorofonn-methano195:5 followed by a second development in chloroform-methanol-acetic acid-water 50:30:8:4. Phospholipid phosphoras will be determined by the method of Sokoloff and Rothblat. Cholesterol and cholesterol ester mass will.be measured by fluoromclxic enzymatic methods. Results will bc corrected for recovery based on appropriate internal radiolabeled standaxds. Statistical significancc of all data will be determined by students' indcpcndcnt t-test. Specific Aim 3: Evaluate the impact of the SAA sPLA~interaction in the mouse model of atheroselerosis using transgenie at~uroacb~o HDL is a patti~darly complex pa~dcle to study given that its individual apolipoprotein and lipid components do not turn over as a single integral. When one considers that HDL can bc viewed as central to lipoprotein metabolism in that it continually exchanges components with other lipoproteins and cells, it is important that it should be studied in its total biological context, confirming and defining mechanistic work. The transgenio mouse model offers a good opportunity to achieve these aims. No model for atherosclcrosis is perfect and the mouse model presents some problems. Strains need very high-fat diets and only relatively small lesions are produced at 15 weeks. At this time, lesions often do not.show the proliferative characteristics of the human atherosclemtic lesion. However, recent studies (A.L Lusis, B. Paigvn, personal communication) indicate that mice indeed develop mo~ prolifexativ¢ lesions if the high-fat diet is nmintaincd over a period of 6 months. The real value of the mouse model lies in 1"ccent advances of routine genetics and molecular biology anct the technology to transfer genes as well as to inactivate them. This has brought the mouse to a new prondnencc as it allmvs the study of individual unit processes involved in the dynamic complexity characteristic of HDL metabolism. Wc have developed a successful SAA2 over-expressing transgcnic mouse model. Funded by anothexpmposal, we arc developing transgenic mice over-expressing each of the SAA family members. The development and promoter selections for these transgenics was described in the Progress Report. 40023612
Page 9: 40023613
Irttmuno~hemieal Sta~in~ of~e Proa~ct ofthe SAA~ TraasKen~ Immunoblot of HDL ~om transgenic_and_uoa-tr0ns~enic mic_e SAA isoforms of olectrofocused mousy HDL were identified with rabbit anti-mouse AA antibody as primary antibody (bottom) and rabbit anti-human SA.A (aa 89-104) antibody (top). SAAa was not identified due to the selective use of antibodies. Lane 1, 100 itg HDL from Zn-fed transgenie mice, showing selective induction of SAA2; lane 2, 100 ~g HDL from control transgenic mice (no Zn); lane 3, 100 ~g HDL fi'om LPS-injected mice; showing the expected induction of SAAI, SAA2 and also SAA~. lane 4, non-transgcnic Zn-fed mice, lane 5, control mice (no Zn), lane 6, mouse acute-phase HDL standard showing SAA~, SAA2 and SAKe. We have received from Dr. Mark Swanson of DNX Corporation, Princeton, NI human sPLAz over- expressing transgcnic mice. We have also obtained non-transgenic litter-mates. The transgenics as well as their litter-mates have been bred back to a C57BL/6 background. These mice were generated by it~ecting numerous copies of the human sPLA~ gcne without modifying the promoter region, sPLA2 pxxluction in these mice is approximately 10-fold elevated. Mice experience substantial hair loss ~s a result of this sPLA2 over-expression but this does not provide any problems. In preliminary experiments, we found that mouse SAA (acute-phase SAAt and SAA2) can promote the activity of human sPLA2 similar to human SAAI. All our experiments on the influence of sPLA~ over-expression with/without SA.A on the development of athcrosclcrosis Will bc conducted in collaboration with AJ. Lusis, UCLA. Experimental protocols are cons~,ucted in consultation with him due to his substantial experience in this area. All studies relating to the scoring of the athcrosclcrotic lesions will be carried out in his laboratory. This specialized work is preferably carried out in an experienced laboratory. Mice, or more likely prepared rna!erials (depending on the logistics at that time), will be sent to him for examination. Aim 3 (a) presents no problems as both the transgenics and their litter-mates are availableas controls. Lipoprotein profilos in these sPLAz over-expressing wansgenies will be established on chow and a high- fat atherogenic diet. sPLAz will be measured in the mouse plasmas and con-elated with lipoprotein changes. To study the influence of sPLA~ over-expression on atherogenesis, 4 groups of at least 15 mice each will be used. Fifteen sPLA~ transgenics will be given a high-fat atherogenic diet and 15 normal chow. Fifteen controls will receive a high-fat atherogenic diet and 15 normal chow. We will initially only use male mice due ~o logistical consideration. Later we will repeat with female mice. Diets will be continued for 15 weeks at which time the mice will be sacrificed. Plasma will then be obtained for lipoprotein analyses and atherosclerotic lesions scored. We will cross our SAA transgenic mouse with human sPLA~ transgenics. We will establish how concomitant expression of both SAA and sPLA~ in these mice alters lipoprotein profiles when compared to the parent trausgenic strains. To establish whether co-expression of SAA and sPLA2 promote atherosclerosis, at least 15 of the SAA/sPLAz cross on ZnSO~ over-expressing SAA and sPLA: wqll be compared to transgenics withou~ zinc where only 40023613
Page 10: 40023614
sPLA2 will b~ ov~r-~x-pr¢~scd. Wlxen this expmim~nt L~ performed, w~- should ~ve subsmufi~ ~m on d~veI~pm~ ofa~oscl~. D~g on ~e r~ul~, ~ffo~ ~n~o~ ~1 b~ iuclud~ p~cul~ly ~ p~ent s~ on ~ off~O~. Meth~olo~: P~a li~ote~ me~7~k~. Mic~ ~ b~ f~ted for 12 ho~s p~or to bte~g. Pla~a ~gly~des ~c ~~ by a ~lo~e~ m~ m~ Fre~ ~d ~ed chol~st~ol ~ d~~ ~ d~n~t (g). HDL L~ i~ed fo~ p1~ b), preeipi~fion of~DL ~d LDL h~ ~d ~g~ cM~d~ ~mb~ed ~DL-C pl~ LDL~ conc~fio~ ~¢ ~at~ by s~~ ~o ~L-C ~lue ~m ~ ~ pl~ chol~t~ol v~uo. cm~ of SDS-PAGE ~d ~oblo~g ae~r~g to ~afio~ of~es d~b~d p~ously (9). ApoB is sepmt~ ~g 5% ~d ~ ~o~p~rote~ ~o s~mt~ ~ing 15% poly~do g~. Plea s~l~ ~o ~ 1:25 in ~plo b~ ~d 20 ~1 of ~oh staple ~ prot~ ~ ~sfo~ed to ni~ol~lose ~g a so~ bloR~ ~Rad T~blot SD). A~B is ~sf~cd m ~¢ellulos~ us~g a w~ blo~ Oi~d Trashier C~). ~oh b~ds ~ ~d u~ a ~hmilmh~s~o pm~m (~~ ~.) ~d ~fi~t~ by d~itom~ (SaiSo~ 5000, U.S. Bi~hmi~l ~.). ~ol ~afioa ¢~oma~phy. To s~p~ ~ li~prot~ ~ ~ plea s~l~s ob~ ~om ~, ~aofio~n by a FPLC ~stcm ~ing ~o S~rose 6 col~s ~ series ~1 be mploy~. ~. All valu~ ~o ~v~ ~ m~n ~ S.E. ~alysis ofv~o, Smdcm's test, ~d Fishers l~t si~t ~ff~¢ test w~ be us~ to t~st for diffcmaees ~ ~pid ~d apoHpoproto~ vflues b~ ~gcni~ ~d n~-~g¢~¢ ~co by m~s of~ S~tficw ~ (Aba~s Concepts) comput~ pro~. V~ ~o ~id~ s~fi~io~y si~ifio~t ~ p <0.05. M~se ~. No~a[ mous~ chow ~ ~a Br~'s C~w ~iet 5015). ~-/at a~geni~ di~ts ~sist of I pm ~om~ Hmero~ diet (ToE~d Test Diets, Madison, ~) m~ed ~ 3 pros P~a Brooder's ~ow. IMu~on of a~os~l~o~ ~ ~sg~ ~ce. At ~e ~nd of ~ 15-w~k p~o~ ~ ~ ~11 be s~fio~d ~d o~ for a~osolemfio lesion. Ph~o~ic scoring ~il be peffo~ ~ d~bed (10); pl~ ~e~t~ ~om sa~fi~ ~ls ~ be ~cd ~ d~t~e ~ ~prote~ proffflo of~ ~ls ~ w¢~ ~ for ~L pr~fiom Mice hems ~d the ~n~ng ao~s ~ b~ pla~ed in PBS pH 7.2 for 5-10 ~utos. For lesion ~lys~ ~d ~os~ng, ho~ ~ b~ ~b~d~ ~ O~ ¢~po~d, ~z~ on ~ i~o ~d st~ at -70~C ~fil ~o~cfio~ng. S~pmont at ~s s~go is fe~ibl~. S~fi~fl ~y~s w~ b~ by one-way ~ ~y ~ysis of~g~ ~ ~ T~ m~fiplo- g~p~s ~oo~m¢. Data ~ be ~v~ ~ me~ ~ st~d ~or (SE), a~ ~¢r~ees ~ll be ~o~d~ si~fi~t if P <0.05. R~E~:NCZS 1. Pruz~ski W, Vadas P, Bro~g J. 1993, L L~pid ~d~o~ 8:161-~67. 2. Bmb~ger~ Lund-~ S, P~fiips MC, ~lat GH. 1985, Bioch~i~ 24:3693-3701. 3. B~b~g~ M, G~ck ~, Ro~blat GH. 1983, J. Lipid R~. 24: 869-876. 4. Collet ~ P~et BP, S~ G, Vieu C, ~te-Bl~ L. 1990, Biooh~. Biophys. Ac~ 1043:301- 310. ~. M~k ~ Ufi~lbShov~ ~ ~d Bou~ EP. 1994, P~. ~atl. A~ g~i. U8A 91:3186-3190. 6. Avi~ L 1959, ~. Biol. Gh~. 234:~87-790. 7. ~rdlior P, ~g~i~ D, Clav~ V, F~h~ ~, ~ 8~wl L 1991, L C~ll Biol. 112:267-2~. 8. ~bl~ ~, Vau~n ~, ~u~ H8 ~d A~ L 1978, L Li~i~ R~. 19:1068-1070. D~li~o ~, ~ ~ufRG, W~n CH, B~o LM ~d L~ M. 1990 3. Biol. Ch~, 265:16380- 16388. 10. Me~ab~ M, D~m~ LL and L~is AJ. 1991 A~s~lerosis. 68:231-240. 40023614
Page 11: 40023615
SPACE,~..d FACILITIES ~y fa~lifi~ tha~ Th~ P.I. hRB 1100 sq. ft. laboratory sp~ a~,-aHablu at th~ University of Kentucky M~di~al Center. The, laboratode~ am fully equipped to ex~e~,t~ all ~h~ proposed expe~.ments, A new fully equippe, d and approved animal fa~iIity at the V.A. Med~c'al Cen~r will b~ ulHiz~ for th~ animal 5. BUDGET JUSTIFICATION- Use this space to explain specific needs for items described on budget pages. The P.L is p~sonally involved in exI~dmental design and execution. Amy King is an experienced individual that will work full-time,in this project. Transgenio mouse costs am the majoritcm. As small a number of animals as possible will be • housed at any one time. An additional broad range of supplics will be needed to execute this program as it involves molecular biology,.apolipoprotein and lipid analyses as well as transgenic approaches. Molecular biological reagents wilt be a major factor. The chemic~2 kits for DNA labeling, restrlcdon endonucleases, glassware, media and reagents for growth of ceils are essential. Substantial radiolabel will be required. Large numbers of lipid and apolipoprotcin analyses val~da~ tl~ n~qu~t. 6. APPENDIX: Place the appendix materials after the origln~l and each copy of the application form as indicated in the Instructions for Coml~dng Renewal Applications. a. Biographical Sketches of the professional imrsoanel to be associated with the project. Each sketch should be NO MORE THAN TWO (2) PAGES. The NIH format is acceptable. The P.L should include and indicate by an asterisk the FIVE (5} most significant publications whether or not they relate dhrectly to this application. b. Supporting material (such as letters of collaboration). o. Copies ofnot more than FIVE 0) of ~e applicant's publimtions or manuscripts that ~e peRiaent to the proje~. 7. ABSTRACTS ofPUBLICATIONS : Only one xe~ is required. Submit ONE PHOTOCOPY of the abswaot page of each "pertinent publication" included in the appendix (6.0.) above; For each manuscript, submit atingle composite page dlat includes authors, title.joum,-d, abstract and publication status (for example, "submitted for publicatimf'). COHFIDENTI AL 40023615
Page 12: 40023616
benefits F.C. de Beer, M.D. 20 25657 Tcc/mical Support Amy King I00 24,860 B. Cotmmmbl~pp~bymajor~tego~) Lipid Assays Enzymes Tissue Culture Rad~olabeled compounds Chemlca~ and glassware c. ore= Ex~.s~s O~mizO A. Salaries Subtotal $50,517 3,500 2,300 2,900 B. Co.sumablcs Sub~J ~ 888 D. INDII~CT COSTS (15% ofA+B + C). E. Permanent F.xluipment (itemize) C. Othex Exp¢nses SubtoL~. A+B+C Subtotal $65,217 D. I'NDMECT COSTS $ 9,783 Pertain-rant Equipment Subtotal F ............................................................................................ TOTAL REQUEST $75, ooo Budget Summades Theaz buget amounts should reflect only modest changes to ,our current award. BUDGET PBPJOD Salari~, Supp11~s a~d Permanent In~ TOTAL Oth~x cxpcasc.s Equipment " Costs Y~ar2 if applicabl~ $68,478 --- $ i0,272 $78,750 Year3 if applimblc $71,902 $10,785 $82,687 *You may not use CYK fimds to p/~q~ permanent equipment in the terminal gram P.I. si~atum -'- ~ ~s~gna~m. ~ 40023616
Page 13: 40023617
7. FIRST YEAK'$ BUDGET: ,t $~/ad~. C~w % time P.I. Na.m~De Beer~ Freaev'~rq¢ (2, (Salary and fringe Professional Personn,l inclufl~ng P~l Inve~iga:cr F.C. de Beer, M.eD. 20 $24,190 benefit. Amy King - Researc~ Technician General laboratory ~ssis~,~n~. B. Consum~le ~pplies ~y m~or c~o~) L i pi d a s s a ys Enzymes ~issue oult~re H~diolabele~ Compounds Chemicals and Glassware C.O~erBx~esOtemiz0 Animal costs 200 mice/year a~ $O.15/day x 364 ~ays D. 1NDIRECT COSTS (15% ofA+ B+C). Permanent Equipment (itemize) None "'I00 $24,860 ~ 50 9,000 A.S~sdcsSubto~ $58,050 $3,500 2,000 2,500 3,000 3,000 B. Consmn~MsSuS~o~l$14,000 10,920 C. Other Expenses Subtomll 0,920 ' A~-B+C Subto~ $82,970 . ~ • D. Indkect C0s~/_l 2,446 E, Permanent Equipment ~ Indicate here attd on Page I. F .......................................................... ; ................................... ~. ............ TOTAL REQUESTS 95,.416 \ Indicate here and on Page I. PROJECTED BUDGET AMOUNTS:,. ~ BUDGET PERIOD Ycax 2 S~ries, Suppli~and-. Oth,r expenses 86,289 89,740 -, - Permanent Equipment - 0"- Indirect Cost~ 12,943 13,461 TOTAL 99,23~ 103,201 Year3 ifapplicablo *You may not u~ CTR fimds to pttrchas¢ perm~.,nt equipment in the terminal grant year. CONFIDENTIAL 40023617
Page 14: 40023618
CURRENTLY ACT~rE GRANTS, CONTRACT:?) m~tl OTHER SOURCES of FUNDS L~st firmndal supEort (direct coo-'~, onl.v) from ~! sources, inclu~.~ o%~ imtRmmn. Title ofl~oj:~t SAA: Role in Atherogenesis Structural prerequisites for amylold flbrillogenesis (co-Px) FuRc~ion of SAA (renewal pending) Function of Serum ~myloid A Protein (SAA)(This propo- sal) .~OIASI0886 ROIAG]2981 ROIAR403791 3375 Id~mify and d~ribe~ auy ov~iap of this applic~ion To~1%'alus (direct $605,764 (five $150,000 Share to Co-PI) $352,747 $131,144 Available $111,090 $ $125,000 $ 45,500 with ~h= above Dat~ of Termiaatlon of Grant o6/30/99 06/30/97 o6/30/95 06/30/95 I. None. Only deals with ~eneratlng tz-ansgenic mice over-expressing SAA iso- types. Does not involve sPLA2 at all. 2. None - deals with omyloid. 3. Presently no overlap. Renewal will overlap (See pending). '~ndieate th~ total al~ual ftmds available to you this year for all research' proi~C,S under your supervision. Note: $175,000 up for renewal 1995 $331,500 PENDING OK PLANNED Title of Project Sources Total Value Avg. Annual .... Total Duration (givo grant of Grm~t Amount (give inclusive numbers) (direct costs) Available to dates) ~O1AR40379 $569,229 $140,000 i. Function of SAA Renewal Pending 2. Am¥1oid Fibrillogenesis: Role o~ apo E and SAP VA $225,000 $ 75,000 07/01/95 to ]6/3o/oo 04/01/95 to 03/31/98 ldemi~m, ddesedibeamyoverlap'~fi~is'applieationwi~theaboveprOe~. I. Major overlap with this proposal. If both funded, budget would need sub- s~antial re-negotiation. 2. No overlap whatsoever. CONFIDENTIAL 40023618
Page 15: 40023619
CTR PROGRESS rC_EPORT (for Competing Renewal Application) Grant number 33 75 Progress Period from 07/'01/9 2 Name of Investigator: Frederick C. de Beer Title ofOriginalGrant: Function sf Serum Anyloid A Protein to: 11/30/94 The m~n a/m of the proposal was to elucidate the function of the SAA family based on the hypotheais that it is related to high density lipoprotein (HDL) function. F'trstly, we aimed to characteri~ the constitutive SAA family members which wc had recently discovered and relate them functionally to nonnal HDL during homeostasis. Secondly, we aimed to study the influence of acute-phase SAA on the abifity of HDL to mediate choleslerol egress from cells. We proposed that acute-phase SAA could impair this process and so promote atherogensis. List all publications (author, title and journal) resulting from the current CTR award that contain a printe4 ,.~cknowledgment of CTR su_uuort Do Beer MC, de Beer FC, McCubbin WD, Kay CM, Kindy MS. Structural p~requisites for serum amyloid A fibril formation. 1993, J. Biol. Chem. 268:20602- 20612. Liao F, Andalibi A, de Beer, FC, Fogclman AM, Lusis AJ. Genetic control of inflammatory gone induction and NK-kb-like transcription factor activation in response to an atherogenic diet in mice. 1993, J. Clin. Invest. 91:2572-2579. Sipe JD, Carreras I., Gonnerman WA, Cathcart ES, de Beer MS, de Beer FC. Characterization of the inbred CE/J mouse strain as amyloid resistant. 1993, Am. J. PathoL 143:1480-1485. De Beer MC, Kindy MS, Lane WS, de Beer FC. Mouse serum amyloid A protein (SAA~): Structure and Expw, ssion. 1994, J. Biol. Chem. 269:4661-4667. DeOliveira RN, Sipe JD, de Beer FC, Loose LD, Baxtel LM, Cecil D, Franzblau C. • R~tpid sensitive enzyme-linked immunosorbent assays (ELISA) for serum amyloid A (apoSAA) in human plasma and tissue culture fluids. 1994, Amyloid. 1:23-29. Liao F, Lusis AI, Berliner JA, Fogelman AM, Kindy MS, de Beer MC, de Beer FC. Serum amyloid A protein family: diffenmtial induction by oxidized lipids in mouse strains. 1994, Arterioslcer. Thromb. 14:1475-1479. De Beer, IV[C, Yuan T, Kindy MS, Asztalos BF, Roheim PS, de Beer FC. Characterization of constitutive human serum amyloi~d A protein (SAA,) as an apolipoprotein. 1994, Accepted for publication by"J. Lipid Res. -~ Banka CL, Yuan T, de Beer MC, Kindy M, Curtiss LK, and de Beer FC. Serum Amyloid A (SAA): Influence on HDL-mediated cellular cholesterol efflux. 1994, Accepted for publication by J. Lipid Res. Pruzanski W, de Beer FC, de Beer MC, St~fanski E. and Vadns P. Serum amyloid A protein enhances the activity of secretory non-pancreatic phospholipas~ A2. 1994. Submitted to Biochem. J. P~blieations 1 and 3: The subject of these were not defined in the original specific aims. They constitute adjunct findings made in exploring the diversity of the SAA family. Because of their significance, funds from this proposal were used for a limited exploration. 40023619
Page 16: 40023620
ABSTRACT OF PROGRESS REPORT In l~urstting tmdersta~ding of how the SAA family impact on HDL ftmet/on, we separately focus on the ¢onstitatb,,~ sabfmmily (human constitutive SAA4 and mouse SAAs) a~d the acute- phase subfamily members. The latter can even become the major apolipopmteins of I'IDL. We propose that the constitutive SAA molecules on normal HDL contribute to its normal physiological role and that the dramatic induction of the inflammatory SAA subfamily equips this particle for an altered, yet probably related, functional role appropriate to the inflammatory state. A. Constitutive SAA: The major findings relating to the characterization of this subfamily were published in two papers appended (de Beer MC, etal; J. Biol. Chem. 1994; 269:461-467 and de Beer MC, et al, J. Lipid ges. 1994; in press). We have cloned the genes and defined the structure ofthese mole0ules. Their plasma concentration was measured and their distribution amongst lipoprotein classes were defined as restricted to HDL and V'LDL. Using two- dimensional electrophoresis and phosphorimaging. Constitutive SAA4 was found to be associated w/th a very specific sub-population of only three HDL particles. These HDL particles are not those involved in the initial cholesterol transfer from cells. We presently hypothesize that the funotion of the constitutive SAA4 molecules is likely re/ated to inter-lipoprotein-lipid exchange. B. The aware-phase SAA: In collaborative studies (Drs. A.M. Fogelman, A.L Lusis, UCLA; J. Rotter, Cedars-Sinai) we showed that increased acute-phase SAA levels correlate with the development ofatheroselerosis in man and mouse. Clinical studies focused on patients after cardiac transplantation, when atherosderosis is a rapid complication, as well spontaneous coronary artery disease. Our studies on the influence of acute-phase SAA and cholesterol efflux provide evidence against our original suggestion that these SAA's could promote atherogenesis by impairing reverse cholesterol transport. Our data indicate that like apo-AI, Apo-AII and ape- C, aeuto-#ase ,qAA can also promote cholesterol efflux thou# somewhat less efficiently than Apo-AI. These studies led to the, in our view, major discovery that acute-phase SAA acts as a co-factor for secretory non-pancreatic phospholipase AII (sPLA2). The result of the hydrolysis of suffaoe'phospholipids of HDL by this enzyme will be the delivery of tipids to cells during inflammation. The reason for this probably involves repair of damaged membranes, lipid needs for cell division, membrane replacement particularly in phagocytes or even in metabolic reasons related to the liberation of fatty acids. Whereas beneficial in the short-term, this same process could promote athemgenesis when aberrantly induced by oxidized fipids particularly in the arterial wall We propose that the balance between the two major apolipoproteins ofHDL, apo- AI and SAA and two phospholipases have the potential to define lipid flow. We slacngthened our infrastructure to explore the implications of the SAA/sPLA2 interaction. Firstly, a successful traasgenic mouse over-expressing mouse SAA2 was developed. Secondly, a system was established to produce large amounts of pure recombinant human SAAt. 40023620
Page 17: 40023621
(2) REPORT OF PROGRESS The SAA family comprises two distinct ~bfamilies. The constitutive SAA~ is postulated to hc involved in normal HDL function during homeostasis. The acute-phase SAA subfamily (SAAt and SA.A2) is likely involved in attca'cd HDL function during inflammation. a) Constitutive SAA: The major findings relating to the characterization of this subfamily were published in two patmrs appendcd (do Beer MC, et al; J. Biol. Chron. 199_4, 269:461-467 and de Bcvr MC, ct al; L Lipid Res. 1994, in press). We have cloned the genes and defined the si~axcture of the constitutive SAAs now known as mouse SAA5 and constitutive human SAA,. We showed that they exist as minor apolipoproteins on all HDL density classes and VLDL but absent f~om LDL. Their disn-ibution is similar to that ot the apo-C proteins and we hypothesize that, like ale-C, this subfamily could bc co-factors for enzymes. Our data indicated that in man and mouse, systems operate for the dominance of either the constitutive or the acute-phase SAA molecules on HDL but not for both at the same time. Whereas human constitutive SA~ is not induced by eytokines, mouse SAAs is weakly inducible. However, its presence on HDL during the peak of the acute- phase respon~ is prevented by either translational interference or displacement and rapid clearance. Using two-dimensional eleetrophorcsis and phosphorimaging constitutive SAA~ was found to be associated with a very specifio sub-population of only three HDL particles. These ItDL particles are not those involved in the initial cholesterol transfer from cells. This allowed us to hypothesize that the function of the constitutive SAA~ is more likely related to inter-lipoprotein lipid exchange. Two major avenues of investigations with respect to the constitutive SAAs are still being pursued. The first relates to the fact that immunoprecipitation indicates that SA~M-earrying particles do not have an apolipoprotein composition distinct from other HDL3 particles. Variations in apolipoprotein composition is thus not responst"ole for their specific charge. Given that pttospholipids have recently been identified as important factors in imparting charge to HDL particles, it is likely these particles carry a phospholipid component distinct from other HDL partioles. This could be of considerable interest given that phospholipid exchange between lipoprotein particles remains ill-defined though obviously important. We are in the process of immunoprecipitating substantial mounts of constitutive SAA~-earrying HDL particles for accurate lipid analyses including comparative quantitative phospholipid analyses. The second avenue of investigation involves analysis ofa SAA~ genomie clone. This is important for the future generation oftransgeaic mice. We have obtained such a clone and have already performed extensive sequence an~,ses. We are presently foensing on the 5' control regions of the mouse SAA~ gene. A paper on this data will be s/~bmitted within the next 2 months. In addition to the major findings that were published in the 2 papers discussed, the human constitutive SAA~ gene was also mapped in collaboration with Dr. A.S. Whitehead (Steele, D.M. et al. Genomics 1993, 164:477-454). This is important for future human molecular genetic studies involving the SAA family. At this moment in time, our main strategy to define the function of the constitutive SA.A subfamily involves transgenic approaches. These will no longer b.e pursued as part of this proposal. b) Acute-phase SAA: i) Increaseq acute-phase SAA levels correlate with development of atherosderosis in man anO mouse: In human studies, we focused on 2 groups of patients; i) after cardiac lxansplantation where transplant atheroselcrosis is a well-known and rapid complication, it) Patients with spontaneous coronary artery disease were also studied and compared to family members withont disease. 40023621
Page 18: 40023622
Serum Amyloid A after Cardiac Transplantation: These studies were cond~ctc~i in collaboration with Drs. A.M. Fogclmau and AJ. Lusts at UCLA. Sea-am amylo~d A concentrations were analyzed in 76 sera from 36 pafisnts taken at a mngc of 12-2379 days after cardiac transplantation. The average SAA concentratim~ 173 _+ 142 ug/ml (range 23-934 ug/ml) was 4-fold higher than SAA lc,;'el,~ taken from 261 normal healthy volunteers (39 -+ 36 u~ml, p<0001). Although the. inflammatcrl nature of acute cellular rejection was predicted to be a major contributor to plasma SAA levels, there was no correlation between SAA levels and the concurrent cellular reje~on status of the patients. SAA levels were elevated in 19 patients with documented transplant corona-D' disease compared to the 17 patients without coronary disea.~ (206 + 160 vs 138 + 104 ug/ml, p-0.02). Se~-u.m amyloid A concentrations were an independent predigtor ofnmrtality after cardiac wansplantation. In the 9 patients who died, SA.A levels averaged 291 __. 154 ug/ml compared to 78 +_ 52 ug/ml in those still living (p,0.002). Eight of the 9 deaths were from transplant coronary disease. It may lm assumed that elevated SAA levels aRer cardiac transplantation may reflect an underlying selection bias due to transplantation of patients with spontaneous arthroselerosis. SAA levels in the 24 patients with a pr¢-transplant diagnosis of ischemi¢ cardiomyopathy was 198 _4- 145 ug/ml and in tho~e with non-ischemic cardiomyopathy (1 valvular, 1 Kawasaki's, 10 idiopathic dilated cardiomyopathy) was 139 + 133 ug/ml (p=0.05). This data independently suggests that SAA levels are associated wi~h the underlying athsrosde~otic predisposition. There was no relationship between corticosteroid dose and SAA. SAA in Spontaneous coronary Artery Disease: Conducted in collaboration with Drs. A.M. Fogehnan and A.J. Lusts, UCLA; J. Rotter, Cedars-Sinai: We measured levels in 85 patients with documented coronary artery disease and 261 family members, taken from the Cedars-Sinai family study. The study sample consisted of 30 multiple multi-generational pedi~cs ascertained through a proband with documented (surgically or angiographically) corona~ artery disease. Family inclusion criteria require at least one other blood relative to be affected with coronary artery disease. Serum amyloid A concentrations in plasma were significantly elevated in patients with coronmy disease when compared to maaffectc~ family members 49 + 31 vs 39 _+ 36 ug/ml, p=0.02). There was no age effect seen in females but males tended 1o have increased SAA levels after age 36 (data not shown). Elevated S,S~ |~vels corr.~lat~ w|th the development to atheroselerusls in mlee on an atherogenie diet: In collaboration with Dr. AJ. Lusts, Division of Cardiology, UCLA, we performed genetic analysis using • recombinant inbred (RI) strain made be.tween the atherosclerosis susceptible strain C57BI/6 and atherosclerosis-resistant C3H strain. The RI strain were obtained from lac -ksoa Laboratories and 10 were fed a high-fat atherogenic diet and lO normal chow. At 15"~eeks, the mice were sacrificed and atheroselerotic lesions were scored by serial se0tioning of the ao~c root. We performed Northern Blot analyses oft the livers o~'these mice probing for inflammatory SAA as well as lipoprotein A-IV (ape-A-IV). The latter was chosen as, similar to SAA, the levels of ape-A-IV mRNA are regulated by dietary lipids. Apo-Al levels were identical in both parent stains. The Northern Blots were quantitated using donsitometric scanning. A number of correlations were performed. The key data indicated that at 15 weeks, a significant correlation (r=0.79, I>=0.002) existed between the extend of lesions and the levels of SAA whereas no such correlation was found with ape-A-IV. 00 Oxidized, Llplds Induce Isflammatory SAA: Findings on the induction of the inflammatory members of the SAA family by oxidized lipids arc published in two papers appended (Liao, F. ct al, J. Clin. Invest. 1993, 91:2572-2579 and Line, F. et al, Arterioscler. Thromb., 1994, 14:1475-1479). Three major f'mdings emerged from these data namely: (I) induction of the inflammatory SAA isotypes is the result of the activation of NF-~cJ~-like transcription factors by oxidative stress. (0 The induction of SAA by oxidized 40023622
Page 19: 40023623
lipids is specific for the inflammatory merab~s of the SAA family with the constitutive SAA~ not induced by this stimulus. This pro,~ide~ ~er suppor~ for functional difference bctw-een these two subfamilies. Perhaps th~ most imporlant fir.d~ug is that the induction of SAA (and other inflammatory molecules} is ~ specific with C57BL16 mice the most susceptible to this indu~.tion and also the most susccpi~ole to the devclopmcnt ofatherosclerosis. Large pro~p~cti~'e studies on health)" human lcopuIat~ons shgw that classical or inflammatory markers such as crythmcytcs scdh'nentation rate, white blo~d call count, plasma viscosh'y and plasma h-'brino~cn arc all in~reazed in ~ndi~duals at r/sk for myo:ardial infarction. In preliminary experiments, we show that family me.tubers with athcrosderosis in the Cedar Sinai study also have significantly elevated SAA levels compared to normals. The possibility that oxidized iipids could be contributing/responsible for this low-grade inflammation and that the SAA present on the HDL of these individuals could have negative metabolic sequelae n~riis consideration. We propose that the low-grade systemic inflammation d~tectable in atherosclerotic paticms and the inflammation at sites of arthrosclerotic plaqu~ d~velopm~nt should bc viewed in a more integrated way. ~ SAA: Influenc.¢ on Cholesterol Effiux: This work was done in collaboration with Dr. C.L. Banka, Scripps Reseamh Institute. The data generated by these studies is appended ('Banka ct al, J. Lipid Res. 1994, in press). These data provide evidence against our original suggestion that SAA could promote athvrogenesis by impairing revcrse cholesterol transport. Our data indicate that like apo-AI, apo-AlI and ape-C, SAA can also promote cholesterol cffltLx though somewhat less efficiently than apo-AI. Wc found that the THP-I cells used in these studies bound an SAA-enrichcd, apo-AI depleted subpopulation of particles that confirmed similar curlier findings. Although significantly more SAA-cnriched HDL~ apoprotein bound to these cells in comparison to normal HDL~, the SAA-snriched particles wcrc substantially more dcnsc than the normal HDL. Th~ number of particles bound may thus haw bccn very similar de.spite the increase in protein bound. With respect to cholesterol cfflux, wc show that particles very enriched with SAA that containcd as little as 4% apo-AI promoted cholesterol efflux only 30% less efficiently than normal I-]DL~. Bccaus~ the molecular mass ofapo-AI is 2.3 times that of SAA, the molar ratio of SAA to apo-Al in the~se particlcs were 50: !. Thcretbre, ottr data suggests that SAA can promote cholesterol efflux though somewhat less vfficiently than apo-AI. Our data support the contention ~at a number ofapoproteins share the capacity to associate with cells and to mediate cholesterol efflux. This sharing of function is most likely based on lhc amphlpathiv nature of these proteins. Wc propos~ thai whereas raany HDL apoproteins share certain characteristics with respect to cell binding and cholesterol efflux, their spv¢ific functions is cffcctcd when they act as co-factors for enzymes. ~) Exnressi~n System for in.vitro Synthesis .of Acute-ph~f S,.AA .Proteins: The human SAA~ eDNA was cloned into an expression vector for production of protein. The vector u~cd ~s pGWIHG which contains the human cytomegalovims (CMV) promoter r~gulatory region, a polylinkcr for cloning in the g~ne of interest and the human growth hormone 3' unlranslated region. This expressiun vector contains the GPT gene as a sdectablc marker to increase the production ofprotein. CHO cells were clcctropormed with the pOWIHG vector containing the human SAA~ eDNA. The cells were expanded and maintained in selection medium. Recombinant human acute-phase SAA~ (rhSAA0 was purified from the supematants at final yield of20lxg rhSAA~, per I ml of medium. ~) Creation of a SAA_~ Over-expressln~ Transgenic Mouse: We have genexated transgenic mice hat over-express the acute-phase BALB/c SAA~ gene to study the effects on lipoprotcin metabolism. It has to be recognized f~om the outset that it was not possible to simply inject numerous copies of the SAA gene with intact control regions into raouse oocytes. This rather straightforward approach would not result in SAA expression as the control regions of these proteins are not "turned on:' in the normal state. Concomitant cytokine or cndotoxin administration would bo needed to activate transcription and these per 40023623
Page 20: 40023624
would hav~ major ~luenc~s on lipoprotein m~mbolism. This would obscure and complicam dm~ing ~ ~olc of~ ~ific S~ iso~s ~ i~lat~ ~i~. We have ovc~mc ~s p~bl~m by removing c~o] r~gions of~e S~ g~c ~d ~g a n~w con~ct ~ a ~affcalIy inducible d~vativc o~ themo~eme~o~ion~ (~)promotor. ~ promotor ~ ~cp~uct ofex~sive development by ~s. P~t~ ~d Bdn~er to ~ow for in~ucibl~ e~sion. ~ vector is u~que du~ to hyp~scnsiffvc si~ ~om the ~a].l ~d la~e mctal.l~thJouein MT gene. In the uns~utated we ~c gcn~m~, ~s MT promotor ~ not '~1~"" ~ S.~A s~es£~ is pint,tally ~detec~ble. Upon ~ea~t ~ ~no ~lfate, there is a more ~ 20-fold incre~c in S~ e~r~sion. ~c ~ount of S~ gm~tcd ~ ~e ~ce is eq~valent m the l~'cl pr~uc~ ~ mice ~jected wi~ LPS. ~ ad~fion, the ~sg~c mice o~y pro~c~ ~c S~z ~o~ when expo~d to ~s s~ulus. We have studied ~c - ~flu~ of z~ su~atc on ~c lipo~otc~ pro~cs o£ C5~6 ~cc ~ comp~n to con~ol ~als. ~ f~ of 20 ~ ZnS0, ~ ~ ~g water of~c ~c~ brings about no ~ li~o~ ~ofilcs. rl) SAA Eql~anee~ She AetiviW of sPLAt: These studies were conducted in collaboration with Drs. W. Pmzanski and P. Vadas, University of Toronto: The data supporting this contention is provided in an appended paper (Prummski ctal, Biochem. J. 1994, submitted). Our data indicate that purified, acute- phase SAA as well SAA-bcafing HDL markedly enhance th~ hydrolytic activity ofsPLA2. This holds major implications for I-IDL metabolism during inflammation when both these acute-phase proteins arc concomitantly induced. This was initially established by using mnlti-lamelIar liposomcs composed of PC, PC:LysoPE. In contrast, normal HDL inhibited sPLA2 activity. What we view as particularly important is that the constitutive human SAA~ does not enhancc sPLA, activity. Our studies on the SAA, sPLA2 interaction are ongoing. We are now confining ourselves to using the rhSAA~ and rhsPLA~ (gift Dr. J. Browning, Biogen) to evaluate the phylogcnctic conscrx, ation of this activation. Wc compared human rhSAA~ to mouse (containing equal amounts of acute-phase SAA~ and SAA2 isotypes) and horse SAA using the mulfilamellar liposome assay. All three of these SAA's have identical conserved regions between amino acid 33 and 44. The horse was chosen because of its nonapcptidc insert at position 69 which makes it somewhat different from other acute-phase SAA's. These three SAA's all promoted the activity of human sPLA2 without significant diffcrcnces dctcctcd. As these SAA's only share approximately 58% amino acid identity outside of the conserved region, the latter is likely the impomant area promoting the activity. More recently, we have initiated studies monitoring the liberation of free fatty acids from HDL and VLDL when exposed to rhsPLA2 with and without respective SAA's. We additionally are comparing the activity ofrhsPLA2 Io another group l] phospholipase derived from crot~lid snake venom. Key data include the following: (1) at 90 minutes, 3.7 times the amount of FFA is generated form I mg of HDL in the presence of 100 pg rhsPLAz than when compared to controls where SAA is omitted. (It) rhSAA~ does not enhance the h'beration of FFA from LDL. (il) rhSAA~ enhances the ability ofa phospholipasc A2 from Crotalus durissus to liberate FFA only by 19%. 40023624
Page 21: 40023625
C(1) BIOGRAPHICAL SKETCH De Beer, Frederick C. G:ve the fo|l~v.~g ~n.~orma~n for the key pe~nn~l and co~u:~an~ ~d co~]~omto~. ~n w~lh the ~nva~tor/prog~m d~r~tor. Phot~oW th~ page for each N~ POSIT~N TITL~ Fr~ck C de ~, ~D. Prof~or of M~ic~e EDU~ (B~n ~ b~a~ur~t~ o~ other i~.~fe~io~t e~a~onr s~ as nu~nR~ ~ ~ud~ p~tdmta~ training. ~ ' ~~D ~ . D~REE ~~ Rm n OFS~Y U~v. o~ ~ ~ ~I, ~u~ ~ ~ 19~ Professional 19~-19~ ~ ~[es~r o~ Me~cine, ~~t ~ M~i~e, U~v~i~ of S~~ M~c~ ~l, To~ ~u~ 19~-19~ ~s~iate Prof~sor, ~p~t of M~i~e, Universi~ of St~s~ M~cal ~ool, To~ ~u~ ~ 1~9-19~ R~ear~ ~eHow, ~y~ £os~ua~ ~c~ ~I, ~s~, ~ndo~ U~ ~dom 197~19~ Reddent in Medi~ne, U~v~ of ~eWda ~ic~ ~l, Prcmda, ~u~ I. C, odenir, NI; Jeenah, MS; Coetzee, CA; Van S~d~n of ~e qu~on o~ 19~, ~:217-~. ." 2. She, d, BG; V~ Held~, PD; S~au~, M; Bo~ L; de B e~, FC. F~c~o~ eff~ o~ ~ b~g to nu~ei. ~o~ 19~, ~:~94. A~n~i~g h~ ~ d~i~ ltpopm~ 3~1~, size ~d apo~prote~ compo~o~ J Biol Chem 1~, ~:9~1. 4. ~d~.~; de Be~, FC; de ~fl~ ~ de~da~n of no~! ~d a~ p~ ~ d~i~ U~pro~ 3. Bi~ J 1987, 2~(~):91~. 5. S~a~ ~; de Beer, ~V~ d~ W~u~ D~ C~, GA. Id~fi~on of ~ of h~ se~m amyloid A pm~. BI~ J 1~, ~1)~-2~. 6. ~p~ ~; de B~, FC; Von Holt, C; ~p~, ~. ~ ~ of s~suc~d~l 2-(p- ~ldos~y~do)-l~di~iopmpio~ as ~ crossing reagent to id~fi~ cell surface re~p~. ~ly~ B~ 1988, 1~13. ~el, AE; de ~r, MC; ~phard, EG; S~ac~, ~; V~d~pl~, ML; de Beer, FC Phosp~owlafion of hu~ s~ ~yloid A pm~ by pm~ ~e C. Bi~m ~ 19~, ~5(1):29-~. ~phard, ~; ~er~n, ~ ~r, SM; J~ v~ R~bm~ CE; de Beer, FC. ~eu~p~ ly~so~l d~a~on ofh~ ~~v~ ~p~d~ ~ula~ neu~p~ f~cffon. CI~ E~ ~ol 19~, ~:139. S~' ~; S~p~rd, se~m ~yloId A pmte~ ~vior ~ ~u~ ~d mea~n~g ~lufio~ ~d ~fi~y pr~ucfion. Bi~hem J 1989, 2~(2)~-370. ~ephard, EG; ~, SM; ~e~on, R; S~n, ~; Nel, AE; de Beer~ FC ~ne~tion of biolo~caHy ~tive C-~c~ve prote~ pep~des by a neural protege on ~e memb~e of phorbol myristate acetate sfimulat~ neuUophi]s. J l~uno13989, 143:~74-298]. 40023625
Page 22: 40023626
Biographical Sketch Frcdcrick C. dc Beer, M.D. Page 2 11. 12. 13. ' 14. 15. 16. 17. 18. 19. 20. 21. 27. 28. S~acha,-~, AF; Brandt, WF; Woo, P; Van d~r Westhu),zeo, DR~ Coetz~e, GA; de Beer, ME:; Shel)ard, EG; de Beer, I~C. Evidence fl~at the six major iso~orrr~ of human serum amyloid A prote2n are coded for by three genes at two genetic loci. J Biol Chem 1989, 264:183~18373. Zahedi, K; Gormennan, WA; de Bee~ I~C, de Beer, MC; Steel, DM; Sipe, JD; Whitehead, AS. Major acute phase x~ac~mt synthesis during chro/flc inflammation in amyloid-susceph'ble and resistant mouse strains. Inflammaflo~ 15;1:I-13. Shepha~, ]~G; SmiLh, PJ; Coel:r, ee, $; 5t~achan, .a~; de Beer, I~C. Pentraxin binding to ~at Iive~ nuclei. Biochem J 1991,279-.257-262. De Be~, MC; Beach, CM; Shedlofsk-y, 5[; de Beer, I:C. Identification oir a novel secure amyloid A protein ($AA) i~ BALB/c mice. Biochem J 1991, 280:.~5-59. Beach, (~ de Beer, MC; $il~e, JD; Loose, LD; de Beer, I~C. Human sei-um amyloid A protein: complete amino acid sequence of a new variant. Biochem J 1992, 282:615-620. De Beer, MC; de Beer, FC; Beach, C2~; Ca~reras, I; $ipe, JD. Mouse serum amyloid A protein: complete amino acid sequence and mR~A ana~eis of a new iso~orm. Biochem 1 1992, 283:673-678. Whi .tel~.ad, AS; de Beer, MC; S~eel, DM; Ritz, M; Lelias, TM; Lane, WS; de Bee~, I~C. Identification of novel ~embers o[r the serum amyloid A protein (SAA) supedamfly as constitutive apo]ipol~roteins of high densib~ Ill~oprotein. J Biol Chem 1992, 267"~862-~867. Ying, 1~(~; $hepha~d, P-G; de Beer, I~C~ Sigel, SW; Harris, D; Gewu~z, BE; Friedkin, M; Gew~z, H. Locell~tion of sequence-detemdned ~oepitopes and neutrophil digestion fragraent~ ofC-~eactive protein utlizi~g mo~oclo~al antibodies and sy~tl~etic peptide~. Molec Immunology 1992, 29"~:677-687. Steel, DM~ Sellar, GC~ Simon, S; de Beer, I:C; Whitehead, ,a~. A conslitutively expressed serum amyloid A protein (SAA4) gene is closely linked to and 8hares etructural similarities with an acute phase serum amyloid A protein (SAA2)Genomics 1993,16:447-454. Liao, 1~ ~ndalibi, A; de Beer~ I~C; Fogelman, AM; Lusis, AJ. Genetic control of inflammatory gene induction and NF-kB-like ~ranscripLion factor activation in response to an atherogenic diet in mice. I Clln L~vest 1993; 91:2572-2579. Steel, DM; Rogers, J; de Beer, MC; de Beer, FC; Whitehead, AS. Biosynthesis of human acute phase sen~m amyloid A protein (A-SAA) in vitro: "Ihe role~ of mRNA accumulation, poly (A) ~il shortening and translaIJ(mal efficiency. BJochem J 1993; 291:701-707. De Beelr, MC~ de Beer, FC~ Beach, C.M~ Gonne~man, WA; Carreras, 1; Sipe, JD. Syrian and Armenian hamste~ differ in serum amyloid A (~AA) gene expression: Identification of novel Syrian hamster SAA subtype~.~ Immuno! 1993, 150~.5361-5370. De Beer, MC; de Bee~, FC; Mc~ubbin, WE; Kay, CM; Kindy, MS. Structural prerequisiles for serum amyloid A fibril/o~nal~: A novel mouse model J Biol Chem 1993, 268"~.0602-20612. $1pe, ~ Car~e~as, l; Gonnerman, W.a,; Cathcart, ES; de Beer, MC; de Beer, I~C. Cha~acte~zation of ti~e inbred C~/J mouse strain as amytoid p~sistant. Am J Patho11993,143:1480-1485. De Bee~, MC; Kindy, MS; Lane, WS; de Beer, I~C. Mouse serum amylold A protein (SAAs): Structure and exp~ion. J Bio! Chem 1994, 269.~61-4667. De Oli,¢ea~, Rbl; Sipe, JD~ de Bee~, FC., Loose, LD; Ba~tel, LM; Cecil, D; l:ranzblau, C. Rapid, sensitive enzyme[inked immunosorbent as~ay~a (]~LISA) ior 8erum amy]oid A (apoSAA) in human plasma and ~issue culture fluids. Amyioid 1994, 1:23-29. Liao, l~ Aldons, J; Lusis, 5; Befliner, JA~ Fogelman, AM; Kindy, M; de Beer, MC; deBeer, FC. ,Serum amyloid A protein family: differential indu¢|io~ by oxidized liplds in mouse strains. Artedoscle~ Thromb 1994, 14:1475-1479. De Beer, M(~; Yuan, T; Kiru:ly, MS; ~szbalos, BF; Roheim, PS; de Beer, FC. Characterization of constitutive human ~rum amy]old A protein (SAA4) as an apoHpoprotein. J Lipid Res 1994, In press. 40023626
Page 23: 40023627
BIOGRAPHICAL SKETCH LUSIS, AIdons J" ! l:h-ofessor cf Microbiology and Medicine INSTITUTION AND LOCATION Washington State University, Pullman, WA 3mgon State University, Corvallis, OR .Iosw¢ll Park Memorial Institute., Baffalo, NY DEGREE Ph.D. Post. Doc. YEAR .CONFERRED 1969 1973 1974-5 FIELD OF STUDY Chemistry Bioch¢mistry Genetics SEARCH AND PROFESSIONAE EXPERIENCE: Goncl~¢~;~ with ~sent ~dl~ isl, ~ ~ro~gl~l o~r, pre~= em#oy~nh e~=~. aM ~, Key ~rso~el t~e IM ~d~l ~ves~igator ~ any other ~als who pa~ate ~n the ~t~c ~veto~ent ~ ex~ of the ~. , ~rso~el tyroly w=l i~ al ~M~s w~ d~al ~ ~hor pml~s~l &gr~s. ~t In some prints ~11 I~e l~Iv;dual= at the mtsle~ ¢r ~u~eate ~11 F~ they ~l~e ~ a s~slan~ way to t~ ~mir¢ d~el~pment or ex~i~n of Ihe pr~ I~e ~esent ~er~ on Federal G~emment ~ ~ ¢om~,ee. LbL ~ ¢~1 ~der, the t;¢~, all aulho~s, and ~mplele ~eleren~s to all ~b,cations ~r~g the t three years and Io represeN=~ ea~;er ~B~at~ ~t to t~ a~on. ~ NOT EXCEED ~O PAGES. ~SEARCH A~ PRO~SIONAL EXPER~NCE 1974-76 [976-78 i977 t979-83 t9~ "9 t98., ~0 1986-90 t987-90 1989-pres. t989-90 .990-pres. National institutes of Health Postdoctoral Fellow (Roswell Pk. Mem Inst.) Research Scientist, Dept. of Molec. Biol., Roswell Pk. Mere. Inst., Buffalo, NY Visiting ].,¢cturer, Dept. of Genetics and Ecology, Univ. of Aarhus, Denmark Assistant Pmfes~r of Microbiology and Mexticinc, UCLA . Associate Professor of Microbiology and Medicine, UCLA Established Investigator of the American He.an Association Associate Editor, Journal of Lipid Research Member, NIH Study Section on Mammali~/n Genetics Professor of Microbiology and Mexiicine, UCLA Membex, Committee on Research, American Heart Association Associate Editor, Arteriosclerosis and Thrombosis ;ELECTED BIBLIOGRAPHY (since 1987) ~/io~ I<J.,, Kirchge.ssne, r TG, Lusis A J, Scho|z MC and Lawn RM. Human lipoprotein lipase complementary DNA sequen~cc. Science 235:1638-164 l, 1987. ~ajavashisth'T, Eng R; Shadduck RK, Wahced I~ Ben-A~,ram CM, Shively JE and Lusts AJ. Cloning and dsstm-Sl~Cific expression of mouse macmphage colony stimnlating factor mRNA. Proc Nail Acad ScHdSA 84:1157-1161, 1987, .as'is A], Taylor BA, Quon D, Zollman S and LeBocuf RC. Genetic factors controlling gtructure and expression ofapolipoprot¢ins B and E in mice. I Riol Chem 262:7594-7604, 1987. 'aigen B, Mitchell D, R¢ue K, Morrow A, Lusts AJ and LeBoeuf RC. Ath-..__.~l, a gent determining atherosclerosis susceptibility and high density lipopmtein levels in mice. Proc Nat Acad Set USA 84: 3763-3767,1987. ',ajavashisth T, Dawson PA, Williams DL, Shakclford JE, Lebherz H and Lusts AJ. Structure, evolution and expression of chicken apolipopmtein A-L J Biol Chem 26~:7058-7065, 1987. loire C, Kirchgessner TG, Svenson KL, Fredrickson G, Nilsson S, Miller CG, Shively JE, Hcinzmann C, Sparkes RS, Mohandas T, Lusts A J, B¢lfrage P and Schotz MC. Hom~one sensitive lipase: sequence, expression and chromosomal localization. ~ 241:!503-1506, 19gg. :irchgessner TG, L~Boeuf RC, Langncr CA, Zollman S, Chang CH, Taylor BA, Schotz MC, Gordon JI, Lusts AJ Genetic and developmental regulation of ~he lilx~prutein lipase gene: Lx~ci both distal and pm×imal to thc- L ;tructural gene control enzyme expression..~ Biol Chem 264:1473-1482,1989. .ajavashisth TB, Taylor A, Andalibi A, Svenson KL and Lnsis AJ. Ideutification of a zinc finger protein thai binds to the stcrol regulatory element. ~cienE.q 245:640-643,1989. ~irchgessner TG, Chant JC, Heinzmann C, Etienne J, Guilhot S, Svenson K. Amcis D. Pilon Co d'Aurio l., Andalibi A,Schotz M, Galibcn F. Lusts A'J. Organizmion of the human hpoFrotcm lipasc g~ne and evolution of the lipase ge,m family. Prtyc Nad A¢~KI Sci HSA 86:9647-9651.1989. 40023627
Page 24: 40023628
illiams SC, Grant 8(3, Reue K, Carrasquillo B, Lusis AJ and Kinniburgh AL Cis-acting determinants of basal and lipid-reg~llalcd apolipoprotein A-IV expression in mice. l Biol Chem 264: i 90~39-19016. 1989. W~rden CH, Langner C, Gordon JI, Taylor BA, Mc~,e.an lW~,~usis AJ. Tissue specific expression, developmr, ntal reg~datit}n and chr(}mosom:d mappingof the l,,cithin-choIesterol acyl tmnsferase gene: Evidence for rxpr¢ssion in brain and testes ~ well as liver. J Biol Chem 264:21573-21581,1989. Rajavashislh TB, Andalibi A, Territo MC, Bediner JA, Navab M, Fogelman AM and Lusis AJ. Modified low density lipopro~ins induce en.dothelial cell expression of gmnulo=yte and maerophage colony stimulating factors. Nature 394:254-257,1990. Johnson DF, Nterl|ch DP and Lusis AJ. Use of the la_..q repressOr in creating sequential deletions and a new sequencing vector. Gene 94:9-14, 1990. Doolittle MH, Lel}oeuf RC, Warden CH, Bee LM and Lusis AJ. A polymorphism affecting _apo_ !ippprotein A'II translational efficiency detemllnes high density lipopmtein size and composition. J "~ 265~16380-16388, 1990. Heinzmann C,, Kirchgessner T; Kwiterovich PO, Ladias JA, Derby C, Antonazakis S and Lusis A J. DNA ,p.olymo~ht.sm h.apl~ypes_of the h:u.man lipoprotein lipase gone: possible association with high density zlpoprotem Jevels. Hum tiepgt 86.578-584, 1991. Liao F, Bediner JA, Mehrabian M, Navab M, Demer L, Lusis AJ and Fogelman AM. Minimally modified low d.ensity lip0pmt¢in is biologically active in vivo in mice. J~lin Invest 87:2253-2257, 1991. Mehrabzan M, Demer L and Lusis AI. Differential accumulation of intimal monoeyte-macrophages relative to lipoproteins and lipofusein corresponds to hemodynamie forces on cardiac valves in mice. ArteriQ and Thromt7 11~. "9.47-957, 1991. Hwa J.J., Zoliman.S., Warden C., Taylor B.A., Edwards P.E., Fogelman A.M., Lusis A.J. Genetic and dietary interactions in the regulation of HMG-CoA reductase gene expression. ~ 33:711- 725, 1992. Lusis A.J., Rott~ J.l., Sparkes R.S. (Editors) Molecular Genetics of Coronary Artery Disease: Candidate ~enes and Processes in Atherosclerosis Karger, Basel, Swit~rland, 1992. Mehrabian M., Qiao J.-H., H~,man R., Ru~ldle D., Laughton C., Lusis A.J. Influence of the apoA-II gene locus on ~DL levels and fa ~ty streak developmen ! i n mice. Aterio~l. Thmmb. 13:1- ! 0,1993. Rrue K., Purcell-Huynh D.A., Leete J.H., Doolittle M.H., Durslenfeld A., Lusis A.J. Genetic variation in mouse apolipoprotein A-IV expression is detem~ined pre- and post-transcriptionally..I. Lipid.Res. 34:893-903, 1993. " Fisler J.$., Warden C.H.. Pace MJ., Lusis A.J. BSB: A new mouse model of multigenie obesity. Obesity Res. I: 271-280, 1993. Line F., AndalibiA., deBeer F.C.o Fogelman A.M., Lusis AJ. Genetic control of inflammatory gene induction and NF-KB-Iike transcription factor activation in response to an atherogenie diet in Clin. lnvrsligo 91:2572-2579, 1993. Qiao J-H., Castell~ni L.W., Fishbein M.C., Lusis AJ. Immune-complex mediat~ vaseulitis increases coronary artexyJipid accumulation in autoimmune-prone MRL mice. Arteriosei, Thr0mb. 13:932-943, 1993. Wardea C.H., Fisler LS., Pace M3., Svenson K.L., Lusis A.L Coincidence of genetic loci for plasma cholcs~er01 levels and ob~si~/in a multifactoria] mouse model. Z_.Cl~l,_lu.~t,~i~ In press. 1993. Hedrick C.C., Castellani L.W.0 Warden C.H., Puppione D.L., Lusis AJ. Influence of mouse apolipoprotcin A;II on plasma iipoprotcins in t~ansgcnic mice. ,L Biol. Chem. In press, 1993. Lin S-C., Lin C.R:, Gukovsky [., Lusis AJ., Sawchcnko P.E., Rosenfcld M.G. A defective GRF receptor in the I~¢t/~ mouse reveals spatially distinct growth factor rcquircmc.nts for proliferation of pituitary soma~0~rophs. ~ature 364:208-213, 1993. Warden C.H., Hetlrick C.C., Qiao J.-H., Castellani LW., Lusis A.£ Athcrosclcrosis in transgcnic mice overcxprcssing ap0lipoprotein A-l[. ~ 261:469-472¢ 1993. Qiao J.-H., Welch C.L., Xie P.-Z., Fishbein M.C., Lusis A.L Involvement of the tyrosinasc gcnc in deposition of cardiac lipofuscin in mice: Association with aortic fa~ty sneak development. L Clin. .Invcsti~. in press, 1993. Warden C.W., Dalviski A., Bu X., PurceiI-H~ynh D.A., DcMecstcr C., Shieh B.-H., Puppoine D.L., Gray R.M.. Rcavcn G.M.. Chen ¥. D. I.. Rotter LI. Lusis A.J. The apal{poprotcin AII gcac determines plasm:~ apolip¢)pr~tei,-All !cvcls and free f:~tly acid levels in t'~:::h h,.:m~m,~ :::':i mire. F'r¢~c. Na.':~St~'i._I,LS..'Az~. in prc.~,L It,)93. -'S 39~ (Re,.,. 40023628
Page 25: 40023629
BIOGRAPHICAL SKETCH CAROLE L. BANKA, Ph.D. [ Assistant Member IN sTrI'UTION .t~D LODATION Rollins College, Winter Park, FL" Ad¢lphi University, Garden City, NY University of California. Smt ,Francisco, CA DEGREE B.S. M.S. PkD. YEAR CONFERRED 1966 1974 1982 FIEt.D OF STUDY Pm-Mcdicino Biology Anatomy. RESEARCH AND PROFES~OI4AL EXPERIENCE: .Co~du¢~ w~th present pos~on, I~sl, in ~r~ol~l o~. ~s eminent, experience, and ~r=. Key~er~nn~ ind~ ~ ~l~al I~e~at~ a~ any ~her i~Md~ ~o padldpale ~ the ~en~c d~ or ~xe~t~ oI the K~ ge~ I~ ~1 ~ all I~vidu~s with d~r~ or o~ ~essional d~rees, but ~ so~ ~ ~ ~e ln~v~uals at the masters ~ureale I~el pmv~ Ihw ~te ~ a subsidize ~y ~ ~ sclenlil~ ~el~ent or ex~ulMn of ~e ~l~ I~e prese~ membership any Fede r~ Gover~ enl ~ adv~ ~mm]ltee. UsL M ~~ order, the lilies, a~ aurora. ~d ~e refore~ to ag ~bt~eaUo ns during ~t ~ee yems and to repr~tbe ea~er ~l~io~ pe~t to ~s applaud. DO NOT EXCEED ~O PAGES. APPOI~MENTS,: 1982-[984 Postdoctoral R~eard~ Fellow, Department of Reproductive Medicine. University of California, La Jolla, CA. R~ipi¢nt of Giannini Foundation Fellowship 1987-1992 Postdoctoral R~arch Fellow, Departmont of hnmunology. The Scripps Rcscat'c'h imstitu[¢, L~ Jolla, CA ""}2-1993 Appoin~d as Flomn~ Seibcrt Follow of tho Atncrica~ Association of University Women 1992-1993 Senior R=e~ch Ass~ia~, DepartmcAt of Immunology. Tim Scripps Research Institute, La Jolla, CA 1993-Pr~ Assistant Mem~r, Depa~nent of hnmunology, ~ Scripps Research ~ Jolla, CA PUBLICATIOHS: Article.s: I. Banka, C.L., aud Erickson, G.F. 3. 4. o 'IS 3~3 |llev Gonad.otropin-rclcasing hormone induces classical meiotic maturation in subpopulatioiks of almtic preantral fo.0JJicles. Endocrinology 1 [7:1500-1507, 1985. Banka, C.L., and Calarco-Gillam, P.G. A new eanbryonic antigen, p67, present during mou~ preimplmttat|on development. Gamete Res. 13:19-27, 1986. Banka, C.L., Unger, M.W., Dulbecco. R., and Erickson, G.F. A rat oocyte differentiation antigen (OA-l) dt~tccttxl t)y a mom~lonal antibody. Gamct~ Rcs. t3:29-38, 1986. Banka, C.L., and Calarco, P.G. The immunoh)giciH approach to the study of pmlmphmtation mammalian dcvclol)mcnt. In: Manil)ulation of Mammali;m Dcvclopmcn! (.R.B.L. GwatkitL cd.1. Plenum Press, Now York, N¥, pp. 353-381, 1986. CuHiss, L.K., Dyer. C.A., Banka, C.L., m~d Black, A.S. Platclct-tncdiamd foam cull fromarion admrosclerosis. Clin. Invest. Mud., 13:325-335, 199D. Curtiss, L.K., Banka, C.L, and Dyer, C.A. Functionally important cpitop~s on apolipoptotcin In: Disordcm of H.D.I,., ~m A. Cadson, cd., Smith-Gordon and Company Limited, London. 4[-49. Banka, C.L, Black, A.S., Dyer, C.A. and Curtis% L.K. TIIP-I ceils form ft)an) cells in rcspottsc tO ¢(~'uhurc wilh lipoprotcin.~ but nol platclcl~. J. Lipid Rcs.. 32([):~-43. l~:mka. C.I... Btmnct, D.J., Black, A.S., ~ntiih. R ~, :rod Gut'ti~. I..K. [.vcalifatiot~ t)f ;tit I)l'tlleili A- I ¢l)ilt)l)c critical 1~)1" aclivatiot~ of ]t:cilhhi;~holextcrt)l ;Icylll'al:sl~'ragc. J. llit)l. Uh,'m . 26f):2.1XS6-23892, 199 I. 4OO23629
Page 26: 40023630
Tcrkeltaubo R., Banka, C.L., Solaal, J., Santoro, D., Brand, K...and Curtiss, L.K. Oxidized LDL induces mont~cylic cell exprc~sitm of itlterleu~:in-g, a chemo~:ine with T-lymphocyte chemotactic ~.:tiuity. Art~rit~.~'ler. "lT~remzb., 14:47-53, 1994. Bank~ CL., Bhv t:, A.S., and Curtiss, ~ ~eali~tion of~n apoliFoprotein A-I epitope critical for lipoprot~in-mediat~ cholesterol effl~ from mon~ytie cells. & B~oL Ch~m., 269:10288- 10297, 1994~ B~nd, K., B~, C~, Maekman, N., T~eltaub, R.A., Fro), S-T., and C~aiss, ~ Oxidi~d low density lipopmtdn ~han~ li~lys~adde-induced tissue fa~or exp~ion in human adherent moemc~s. A~erioscler. ~womb. 14:7~797, 1994. Bank~ C.~, ~u~, T., debar, ~C., ~ndy, M., ~udips. I.K., ~d debt, RC. Serum amyloid A (S~): ~fluen~on HD~medi~cd ~]lularcholesterol ¢ffiux. J. OpidR~, submitt~, 1994. .Abstracts: 1. Banka, C.L., Black, A.S., and Cudiss, L.K. Platele~s are potent mediators of macrophage, but not THP-I cell cholesteryl ester aeemnulation. FASEB J., 2:A1172, 1988. 2. Dyer, C.A., Banka, C.I., and Curtiss, L.K. Platelets stimulate THP-I cell apolipoprotei,t production, FASEIJ J., 2:A1597, 1988. 3. Dyer, C.A., Banka, C.L., and Curtiss, LK. TI-IP-I cell ape E production is not del.mndcnt on cellular eht)lesteryi ester accumulation. Circuldtton, 7851-47, ! 988. 4 Banka, C.L.,~ Dyer, C.A.,. and Cuttiss, L.K. Can THP-I cells form fdam cells? ./. Cell 107:573a, 1988. 5. Banka, C.L.,Smith, R.S., Bonnet, D.J., and Curtiss, L.K. Localization of an apolipoprotein A-I epitope critical for LCAT activation. Circulation, 82(4):111-330, 1990. " 6. Brand, K., Banka, C.L., Mackman, N., and Curtiss, L.K. Induction of tissue factor in THP- |cells stimulated by modified low density lipoprotein. Circttlatim~, 82(4):II[-207, 1990. 7. Banka, C.L., Black, A.S., and Cuaiss, L.K. Inhibition of apolipoprotein A-I-mediated eholeste,'ol efflux from mmmeytic cells by monoclonai mdibodi~s. 9th h~t'l. ~ytttposhtm el! Atherosclerosis, p. 235, 1991. 8. Banka, C.L., Cutliss, L.K., Yuan, T., and De Beer, F.C. Acute phase HI)I., promotes cellular cholesterol cfflux. FASEB J., 8(4):A454, 1994. 9. Banka, C.L. and Curtiss, L.K. Estradiol seleettvely inhibits oxidation of high density lipoprotcin. Xm International Symposium on Athemselerosis. Montreal, Oct. 9-14, 1994, in. pre'~, 1994. I 0. Banka, C.L., Guo, Y., Riehards, K.B., Laranjo, S.A. a,id Curtiss, L.K. Detection in plasma of vitro oxidized high density lipoprolein (HDL~. Circuh~tion, in press, 1994. 40023630
Page 27: 40023631
i.475 Serum Amyloid A Protein Family Differential Induction by Oxidized Lipids in Mouse Strains Feng Liao, Aldons J. Lusis, Iudith A. Berliner, Alan M. Fog¢lman, Mark Kindy, Maria C. de Beer, Frederick C. de Beer A/~r~t' During inflammation, scs~-n ~m~old A (S~) protein incr~s up to l~-fold and ~ be~c a ~r ~n~t of hi~ty ti~prote~ ~L). We t~d a n~w apoli~prot~ mollie (~) ~ s ~tm~ ~bcr of ~ S~ ~mi~. It differs ~om the ~t~s (SAAs. S~, and SAA~) ~t ~ ~ stature a~ for mo~c th~ ~ of total ~ ~c~ mcmbc~ of the SAA f~l~, ~du~s SA~. ~ to cndoto~n sdmi~st~6on, SAA~ ~ntmt~ ~ ~ i~ rcsist~n~ to induction eider ~ z ~st, high- cholesterol athcrogcnic diet or ~e in]~ ofmi~ o~di~d ~L (~-LDL). ~c data p~i~ ~cr c~d~ t~t the ~du~ion of in~mmaloff molcc~ ~ o~dhed fipids is selective. In athemsclero~is-susccptible C57BL~6 mice and athcmsclerosis-re~'tant C3H/HeJ mice, the inflammatory SAA L~t~pes (SA&. SAA~, and SAAj) responded in a straln-speclfic manner to oxidized lipids either generated with feeding of the ath~rogenlc diet or introduced by MM-LDL injection. We h~pothesize that the cansthutive SAJq mole- cules on normal HDL may contribute to its am-real physlaing- ic~l role and that the dramatic induction of the inflammatory SAA subfamily equips r.he particle for an altered yet related functional rote appropriate m the inliammatory state. (Artzn'o.~/o" T/trond,. 1994;14:1475-1479.) Xey Wocd¢ * acute phase * athcrog.encsis * serum amylold A s~bfsmilies We have recently idcntificd that members of the serum amyloid A (SAA) fatally arc among the inflammatory genes induced when mlcc are fed a high-fat, high-cholesterol athcrogenic dict.t This induction could have implications bcymtd localizcd atherosclerotic plaque development as SAA. has the ability to associate with preformed high-denslty llpo- protein (HDL) with displacement o~f apolipoprotein A-I (apoA-l)2 This rcsu|ts in remodeling of HDL, yielding larger and denser particles.~ The functional implication of SAA association with HDL remains to be elucidated, but a number of possibilities merit consideration. When one considers that the afl]nlty of apolipopeoteins for HDL ~s apoA-ll>SAA>apoA-I,3 this displacement wottld shift the ratio of apoA-lI to apoA-I toward apoA-1l. The influence that this might have on fipopro- rein metabolism is conceivably impm'tant, given recent studies in tran~enic mice showing that overexpression of apoA-ll promotes the development of athemscicro- sis.* HDL may protect against the development of athcr~clero~is by removing excess cholesterol f~om peripheral tissues during the procest of rcvct~e choles- terol transport as well as preventing low-density lipo- protein (LDL) from oxidation. It is conceivable that the Recclved January 21, 1994; revision accepted June 16, 1994. From the Department of Medicine, Div~ot~ of University of California Los Angeles (UCLA) School of Mcdlcinc (F,L., AJ.L., J.A.B., A.M.F.); the Department of Mlcroblo~o~, and M~ccular Gcnctlc~ and Molecule" B~o~ogy Institute, UCLA (AJ.L.~, the Department of Bi0chcm~tt3' and Medicine, Univer- sity of Kentucky Col|ego of Medicine, Lexlagtcn (M.K., M.C. de B.): and the V~q, Mcdlca! Center, LcxinSIon. Ky (F.C. de B.). G~ran Bonders kis~dly acted as Guest Editor for this article. Co~rc~pcndcncc to F.C.. de Beer, M D0 Department of Medicine., Ro~m M~619, University of Ken~ck'y Mcdi~l Center, ~ Rc~c St, Lcx]nglon, presence of S'AA on HDL might influence these pro- cessna. An addition~l factor that needs to be considered is that during inflammatory state~, such as when an athemgenic diet is administered to mice, SAA is pres- ent on the HDL particle when Wtal HDL cholesterol is significantly decreased, which in itself is. generally con- sidered to be a risk factor in athcmgenc~is? The mo~e SAA family can be divided into subfamilies. The first constitutes the c~asslc inflamma- tory acute-phase SAAs (SAAs and SAA2) that arc dramatically elevated by cytokinc-drlven hcpatic~ synthe- sis to even become the major apolipopmtein spec/es of HDL.~,s The SAAt and SAAz genes arc expressed mainly in the liver, and the rcspe~ivc ro.17,1qA can be dctccted as em4y as 2 hours after endotoxin administxa- tlon.~.~ Pb~-aa levels for these molecules peak at 20 hours after the stimulus,s Vexy low levels of SAAs and SAA~ mRlqA can be identified in the k/deny, whereas only SAAs mRIqA, is evident in the ileunkt The second subfzmily (SAA~) is, in additk3n t~ hepatic production, produced at a variety of cxtrahcpatic sites and is only present as a minor apolipoprutcin on acute-phase HDL.re.st Macrophages and adipocytes arc the main sources of extrahepatic SAA~, and it has been postu- lated that SAA~ produced at sites ofinflammution could locally associate with lipids generated by tissue necrosis to promote tissue repair,s: A third subfamily (SAA~) h~s recently been identified.~.tz We report here that in contrast to the inflammatory SAA. family members, these molecules arc constitotively e.~prcssed and arc the major form of SAA on normal HDL where they consti- tute more than 9D% of the total SAA on the particle as a minor apolipoprotein species. In this study wc exam- ined the induction of the various members of the SAA family hy the injection cf mitdly oxidized LDL (MM- LDL) compared with consumption of an atherogcnlc 40023631
Page 28: 40023632
diet in atherosclerosis-~usceptible (C57BL~6) an0 -resi~ tant (~H~) mo~ strMns. Th~ I~vcls of both h,~ati~ mRNA a~d plasma protein for th~ -r~is~t C3~H~ s~in after mic~ were cith~ ather~enic diet or ~jmed ~th MM-~ ~he ~mtD ~tively expre~d SAA~ mollie, al~ou# in~lc ~ en~to~n a~inisUa6on, is r~ismnt to indu~on by MM-~L and the a~et~eniu di~ ~ ~1 ,t~m. f~ ~at MM-LDL ~Bcd to indu~ infl~matory SA~ in ~H~ mi~ su~ ~t ~e r~ to d~ed lipi~ ~ t~ ~n t~id~ et ~e ~lleler level in addition to the ~ redu~d abBity of H~ ~ to generate o~d~ed ~pid spe~ in ~e l~er ~ zes~nse to an a~cz~enic dieU Methods ~ce and Materials Ferule ~J-Ip#l~t (b~lpr) and (M~n) m~ were obt~n~ from the Ja~n ~tory. F~e ~&6, ~ and ~Hd ~ w~ p~- chad ~m eider ~e J~n ~mory or H~aa Dawley~ I~ ~ ~ us~ for e~e~ ~e 3 to 6 m~ths old. ~e ~ntrol diet wm ParMa ~ow (R~ton-~6~ ~n~i~ 4~ f~L ~e a~crogenic dieL obtained fr~ T~lad (~ ~21), ~nta~ed 15.75% fat, 1.~ ~fol, and 0~ ~dium chelate. ~pol~a~ (~) p~p~ed frm ~htr~cMa ¢d~ Olll:~ was ~tchmd f~m Io#~ ~bonmrie~ Inc. ~poproteins M~se ~L ~ ~olated frm plume by d~ty m 1.~ #mL ~ mild KBr and ~n~ h~n at ~2 ~ in a ~i80 rotor (~ Insmen~) at I~C. The ~fmnatan~ ¢oataiMng HDL wine mlle~, the d~ty w~ ~adjusted to 1.21 ~mL wi~ $olld KBr, and the infranstant~ were r~entrifuged for 9A ~u~ at 242 the ~i80 rotor at 1~ ~e ~pernatan~ ~t~ng HDL were emnsi~ d~ a~mt a ~lut~n of 0.~ m~ Na~ ~d 0.01~ (~1) ~T~ pH 7.4#° Hm~ LDL a~d ~ta~ of MM-~L ~ ~ld ~t~ge and ~n o~- t~a w~ p~d aS pr~m~ d~sos ~ ~S in ~-LDL p~am~ ~s I~ than ~ ~ dct~ed ~ ~ ~romge~c amy J* Mo~ ~ w~ m~m~ ~ng a n~it antoine am#o~ A anti~ ~ift h~ Dr J.D. Si~ ~ton Uni~nity Medical and S~ Rabbit anti-h~an S~ (amen ~ t~idu~ 95 1~) (~fft f~ Pmf ~R. Steime~ U~er~ty of Gemny) rea¢~ wi~ SA~ on~. ~it anti-mou~e SAA~ ~ a g~t fr~ Dr ~ Mee~ U~nhy of W~in~o~ S~ttle. Ele~ofoming ~i~ (~ ~) of mu~ HDL were frem~tled and dellpldat~ with 03 mL ch~mfo~meth~ ~e ~lipid~ted p~eim were resus~d in a ~xture (w~ol) ~i ~dlum sulfate {~tm~ ~d~ ~), 7 urn, and 5~ (~J~l) 2-merm~t~l and eMctmf~d on ulttatMn a~yla~de ge~ ~ntainlng 2~ (vol~l) amphw line pH 3-10, 40~ (vol/vol) amp~llne pH a-63. and 4~ (v~l) amp~tiae pH 7-9 (Pharaoh LKB ~te~nol~y) ~ ~S~i~tt~ Ex~acti~ of ~A ~roln ~u~ and Blot ~bHdi~tion Ltve~ were coll~cted, frozen M liquid nitrogen, and stored at -7~ "IMtal RN~ ~s ~t~ m~ing the acid guan~din- -- -SAA a I 2 Rs 1. Is~lectd~ focusing al1~ysis of Coomassie-stskmd Ngh- densRy Ipoprottln {HDI~ from nolmal B.q.B/c and MRUIpr mlce (~LE phenotyl~). Lane I, 20O ~d'not~nal 8AU~/o HOL: l~ne ~ 104) U@ l-tDL f~- Md~IX mlce wl~ In~nrnato~/.senun am~old A |SAA) lsotype~ ~A/q, ~, and ~ p~esent. lure th~>c,j~nate--phenot-chlotoform methodJt RNA samples (20 ~g) were denatured and clcctrophoresed through a 1% forma]dehyd¢ agaros¢ g¢l followed by blotting onto-a nylon filter and ultravlo~et cross-linking.t" Northern bio~s were hy- bridlzed with SAA=, S~, and .~AAs eDNA probe~ as well with end-labeled 18-base.lo~ ollgonucleotldes sp'.cifie to SA.A3 and SAAb re~ecti~Jy. Specific 18-met oligonuc~otide~ were derived from exert 4 ¢5 reported previoudy.~* Mntine .SAA~ cDlqA eJojlc (do~e pRS48) wa~ a gift from Dg B. Taylor, the • Jeck~m Laboratory. ThL5 483-bp probe does not cro~:~eact with SAAs~mder ~ecoudldons des~ibed. The cDN,~clo~e for SA.A~ has rcccndy been e~tabllshed.~2 The co~ditions for hy- bridlzadon were de.u:dbed el.fewhere.~ The autoradio~rams of Northeen blo~ probed with radiolab¢led SAA~ eDNA were measured by densitometric scanning (Ultro~an XL FcJ~hanccr La.~r Demimmeter, Pharmacta LKB) and the vllue~ analyzed for statistic=d slgnlllcance. Results SAA~ Is Constitutively Present i~,Mouse HDL Isoelectric focusing analysis of normal and acutc- pha~e I-I~T- f~om BALB/c mice indicated that practi~ cally only SAAs is detectable on normal mouse I-ID~ (Fig 1, lane 1). Durlng an inflammatory stimulus (200 ng IV LPS per mouse), dramatic induction of SAAt and SAA= could be seen (Fig 2). SAA~ was detected as a minor apolipoprotein species on the HDL. The mouse SAA family is thus similar to the human, in which a constitutive SAA was recently identified.2a A similar situation exists in C57BIJ6 and C3I-UHeJ mice, with the only cxceptlon being MRIJlpr mice. MRL mice thee carry the lpr mutant gane (MRIJlpr) suffer from a phenotype characteristic of human systemic lupus ewthematosus (SLE)~; inflammatory SAA isotypes (SAAj, SAA:, and SAA~) were cleerly the dominant forms of SAA on the HDI.. of these unstimu|ated mice (Fig 1, lane 2). Induction of SAA Family Members by LPS By using a cDNA probe for SAA~ ~nd specific oligonuclcotide probes for SAAt and SAAz, wc con- 40023632
Page 29: 40023633
aAA2 ~ firmed the significant cxtrahcpatlc expression of SAA3 in re.sponse to LP$ in BALB/c mice. SAAs and SAA, wcrc concomitantly induced but practically only in the i~vcr. A small amount of SAAI mRNA was also detect- able in the kldncy. SAA~ was the only SAA family mamber whose ml~A could be constitutively detected, and it it,.cteased in response to LPS. The mRNA for SAAswas practically l~esent otdy ilx the ]Der, with trace amounts present in the heart (Fig 2). Induction of SAA Family Members by an Atherogen|¢ Diet in Mouse Strains Fig 3A clearly indicates a sigtfhqcant induction of mR/~.A, for SAAt a~d SA~ in rcspoose to the atheso* genlc diet in C57BIJ6 mice at 5 weeks. At the same dine, C3H/HeJ mice showed a .less striking increase in F~ 3. B~0te show d'~fet, engM Inductkm o/serum amylo~d A (SAA) Isotyp~t In C57BW6 an¢~ (~H/HeJ mice on chalenge with an atherogentc dlal MIc~ wore fed eight, ch~,~ or the eahem- ~c diet for S weeks (A) or 15 weeks (B), and thetr hepatlc mRNA leve)a Ior d~fferent SAk5 were Sll~lied I~ Nodhem anaIy- ~ In A. Lines 1.2, and 3 me 057taU6 ~ on the ChOW diet; lanes 4 and S, C578U6 adce o(1 the athero~en|c diet; (ane$ 6.7. and 8. C3H/HeJ rn/ce on the d3ow ¢l/et; lanes 9 and I0. C3H/HeJ miCe oa the a~herogenic o~et. In B, lanes 1,2, and 3 am C57BU6 mice en the chow died; fanes 4, 5. and 6. C57B~6 m;c~) on the a~hemge~k: dio~; tar~s 7, 0. and 9. C3H/HeJ m~ce on ~ Chow d/e(; I,~ne; tO, II, and 12. G3H/HeJ mTce ~n t~ a~J~ro~er.~c d~eL L2.=~ et a! SAA Ind,,ctt:n l~y Oxidized Llplds 147"/ th-~se acute-phase SAAs. In contrast, the diet did not influence SAA~ mRNA I=veIs in both strains. When athe.rog~nic die.t was continued ~or 15 weeks, aAA~ still remained un:hang:d, bat hhc cI:ar induction of inflam- matory SAA (SAA: in this in*tar, cO mRNA was longed in C57BU6 n-dec and remained at much low¢r levels in C3H/HeJ mic," (Fig 3B). The induction in the C3H/He$ strain again was less than that in C57BL/6 mice (Fig 3B). Using an immunoradlomttric a.~ay for SAA, and SAA], we confirmed that these increased mRNA. levels ~esulted in i/~Jcase.d SAA on" HDL particles. SAA, and SAA~ measured 7±3 #g/mL in C.57BU6 mice and 9±4 Fg/mL in C3H/HeJ mice on the chow diet. After mi~ had been $ weeks ~n the athero- gcnlc diet, the SAA levels in C57BIJ6 mice increased fivefold, to 36-4-11 #g/mL The correspouding incrcas~ in C3HJHeJ mice was much le~ doub|ing in concen- tration to 19"-8/~/m.L. MRIJIpr mice on the chow diet had SAA~ and SAA~ levels of 59±13 pglmL. For BALB/c mice, after ~ administration (25/zg IP) the levels for SAAt and SAA, incre.a~ed to 287__.43 ]ndu¢llon of SAA Family Membea's After II~Jectinn of MM-LDL Feeding mice with the atherogenic diet resulted in inflammatory gene induction in tissues that acc~melate llpids, such as the liver. Although the hepatic total lipid levels arc similar in the atheroselerosls-susceptibie C.qTBIJ6 mice and the -re, brant C3H/He_T mice, the former has a greater capacity to oxidize these lipids as indicated by significantly higher levels, of con}ugated dienes in their livet~.t The response to MM-LDL was investigated in C3HJHeJ and C57BIJ6 mlc¢. The remalts presented in Fig 4 are representative of several experi- ments. Aliquots of 50 or 200 #g MM-LDL were injected intraperitoneally and the animals killed 4 hours latcL In both C57B]J6 and C3H/FIeJ mice, control phosphate- buffe~ed saline (PBS) gave no induction 0f the mRNA fo~ acute-phase SAAs (Fig 4A tnd 4B). in atheroscle- msis-snsceptibl¢ C57BL/6 mice., 50 and 200 ~ cold- storage--prepared MM-LDL caused a dramatic induc- tion of SAA~ mRNA (Fig 4A). C3H/He.T mice were resistant. Fig 4B compares tim induction of SAA family membe~ in C57BIJ6. C3H/HeJ, and ]~L1]/c mice after admilziaUMion of MM-LDL prepared by iron oxidatioth sho~ing the dramatic induction of the inflam- matory SAA~ in C57BL/6 and BAI.B[c but not in C3H/HeJ mice. SAA, was also dramatically induced in athe~os¢lerosis-suscep6ble mice (C57BIJ6) in response to MM-LDL adminiatration (data nee shown). As shown in these Northern blots, SAA, is conatitutively pressed, and injection o~ MM-LDL does not ~ignifi- cantly alter its expression. To account for the variability in SAAs expression among individual mice (F~g 4B), multiple Northern blots were measured by dansitomet- tic scanning, yielding a value of 4.664-0.87 (SD) U for four anhrtals bisected with PBS or native LDL and a value o~ 6.05"*-2.3"/(SD) U for ci~hl animals in, coted with MM-LDL (~=.30). Discussion Our studies on the SAA family indicate Ihat oxidized iipids h~ve the potcutial to induce inflammatory genes in the liver with systemic inflammatory sequelae. Among these genes are tho~c [or c~rta], members of 40023633
Page 30: 40023634
I SAA3 B the SAA family (SAA0, SAA2, and SAA~) that have the capacity to associate w~th and rad/cally alter HDL.~ The SAA proteins are likely to affect both lipoprotein metabolism and inflammatory processes, particularly when one view~ recent data of altered HDL binding to macrophages and othex cell-" when SAA is present on the HDL partlcle.~ Inflammator~ gene induction in the liver thus amid have a W~temic influence on lipopmtein metabol~m, amplifying atherogenic potential. It wa~ recently demot~trated that several members of the SA+4. family can accumulate at site~ of vasculitis and lipid accumulation in MRI31pr mice.~ These included SAAt, SAA=, and SAA+. A~ SAAt and SAAz are not produced by cells at these sites, it h presumed that they were: traflsported there as part of a systemic inflamma- tory re~pot~, whereas SAA3 is probably largely pro- duped locally. We suggest that systemic and local inflam- mation at site~ of athcrm~t~otic plaque development should bc viewed in a more integrated way. Large prospective studie~ on heaRhy human p~pulatinn~ that classic inflammatory mmker~ r~ch as er~hrt~'~e sedimentation rate,~s white bl~od cell c~nt, plasma viscosity, and plasma fibzinogen are all inmca~ed in indivldual~ at rbk f~r myocardial infarction.~s The billty that oxidized l~ids could be c~ntrlbufing to or be resl~nsible for this I0w-gradc inflammatinn and that SAA could be pre~ent on the HDL of such indi'eidual~ in incrca~d amounts metlts con~der~tion. Until tmw, the v~¢~v was prevalent that the SAA family m~mhers were present on normal HDL in an 40023634
Page 31: 40023635
~ USA. ~ £ I~I~-4Z ~ a~ ~re ~ R~ 19~;t4:~-~. II. ~k ~ ~a N, ~R EP. M~ac ~ am~ A~ b a 12. Dt Beer M~ Ki~ l& S~n~,DeB~r F~V~derW~u~ D~ 14. ~ner l~T~t9 tiepin sfimuttt~ ~n~c ~th~ial ~ll i~ Cl~ 15. ~ FJ, ~rU~r F~elman A~ Minlma~ m~l~d L~o e! a! SAA Indt:clbn by OxiSizcd L~plds 1479 19. ~ P& ~ of deaa~ed ~ ~d ~ DNA d~ntmoum t~J~pMt£ 19~;1~-~7. ~ ~ ~ ~ I~m~fi~ of n~cl mem~ Of the p~otel~ of ~;gh d~s~y li~r~eln, l Bi:! ~ I~7: M~-~7. ~-m~ ~t~ in~ mro~ ~ lipid mubfi~ ~ au~un~pm~ MRL mi~. ~l~ ~. ~n ~ ~t~er ~ ~dt P~ ~ fa~ for ~ ~ ~c S~Im Pmsp~'StudF Ac~a Med S~ ~tckc~ PI. ~w~ P~ ~b~e~ ~, ~d whltc b~ ~ ~nt ar: nm~r ~ f~o~ for ~emic k~ d~ ~y ~d S~dw~l ~IIaMntM Hea~ D~e~ S~ic~ ~. 40023635
Page 32: 40023636
. .ouse Serum Amyloid A Protein (SAAs) Structure and Expression* Marl• C. de Beer:[:, Mark S. Kindy$, Will/am S. Lano|. and Frederick C. de Beer||'* A novel member of the mo~se serum ~nyloid A pro- t~ln family, SAA~, has been identified as a normal • I~olil~protein component of non-acute-pha~e high density li~protet~ (HI)L), T~e derived from & clone ~ fl~om a no/~aal Balb/¢ ]~ver eDNA I[braey. The clone predicts a pre-SAA~ mclseule of 130 r~m ~m w~ ~ 18-~ue ~pfldo ~ ~g ~m ~ition ~0 to ~. d~ly, in a~p~ S~ mcl~ of a n~r of o~ ~1~. Thee k 48~ ~o a~d Iden~ty b~een apo-S~ ~d the o~er mouse S~ pro~ ~d 67% identity ~n the h~ C-S~ ~ ap~ 8~s. ~e S~ ~A ii t~ t~es l~er th~ pr.vlo~ly ident~ed S~ ~N~. ~h S~s k mmituflve~ e~r~ ~ ~e fiver, it h~ a ra~d ~t mu~d r~ ~ i~ma~ ~muli. Z-~r~e of S~ ~A ~ dne ~ ~cr~ t~ip- s rather ~ ~A s~tlo~ P~ S~ ~ovek d~l~ ~e ~ p~ ~e biph~c, ~er ~- ~e of tr~~ con~ol or ~lac~t ~om ~L ~d rapid c~e~ We p~ ~ ~tutive 8~ (~) ~ ao~ ~L ~n~um m i~ no~ p~slolo~c~ role, where~ ~e ~ati~y ~du~ble f~y m~b~s (S~s, S~ S~) equip ~ p~ti- ~e for ~ ~red ~nctio~ role d~ng i~mation. The acute-pha~e response encompasses a variety of meta- bolic changes (1). Chamcts~tstic is an increase in the concen- t~ation of cert~n plea prote/n~ ]~wn ~s a~ute-phase I~ctants (2), Serum smylold A pmteh~ (SAAs),t a family of ~ homologous proteins with alarge del~ee ofintsnpedee relstedneu (3), are some of the meet dramatic acute-phase reactant=. Pla~ma levels can increase within 20 h from trace amounts to ~ much e~ I mg/nd through cytokine-driven hepatic synthesis (2, 4). All SAA~ ar~ apop~otelns of high density lipoprotein (HDL) (4-6) with which they a~ociate by displacement of apolipoprotein AI (apoA1) yielding larger "'I~i~ work wa~ ,~pported in part by Grants ~J56 and 3375 fxom s~.¢.i] for ~ObaCCO Re~earch. Veteranl Admin~trstion. medical ~-!~ ~I~, and United Stat~ Public Health Service Grant AR 40379 (to F. C~ de B.). Th, costs ef public4t[o~ ofth~ ar~iclo were ~efrayed in pare by the payment of l~S ¢h~es. Th~ erect, must therefore be hereby marked "¢u~r~m~r,Z" in sccozda~¢e with 18 U.S.C. Section 17~t ~lely The ~ _se~_ ,ence(~) reported b~ th~p~p~r has been.~d~mitted "" ~he GenBcqkrM/EMBL °" To whom ¢on'e~pondence Iho~l be address~:L Tel.: 606-233- t "F~I ab~revi~ti~r~ u.e.ed are: SAA, ~erum ~mylo~d A pr<~in; HDL, high density ~p~rctcin~; LPS, |i1~ly~a~h~e; HPLC, h~gh ~r- /~r~ancc li~id chr~mzltcgra~hy. particles with • higlar hydrat~l density and relatively less apoA1 (7}. The ~on~ evolutionary e0naervation of the SAA pz~,ein~ ia segt, eJtive of an in~oxC~nt bi~lo~kal role that zemains ~ yet unidentified, but th~ premnce of apo-SAA~ on HDL suggests that it is involved in the biological function that th~ perfide ful~l~. HDL is a p~r~cu]ady complex partlch to study glean th~ fact that its components do not turn over -- • single internal (8); apo-SAA with a t~ of 75-80 min turns over mo~t rapidly (8). HDL plays a central role in lipid metaboliam by continuoualy exchanging components with other lipoproteh~ and celb (9). The role that respective mambtm of the SAA family could play in these precedes merits careful consideration. SAA (the genes, it~ expression, and the pmteine) has been well characterized in the moule {10-12). This animal is an excellent model in which to investicate the mistlonship be- tween HDL and apo-SAA.The interspeeie~ relatsdne~ would allow conclusions to be extrapolated to the function of human apo-SAA. Until ~cently it was at~med that the mouse hns three actively t~s~cribcd gene~ (10). The ©ozre~pondlng pro- teins, ~po-SAA~, apo-SAA~, end apo-SAA~, are th~ cla~ic acuto-ph~e ~e~efants synthesized in the liver. Apo-SAA~ aloneis ahoproduced at avari~ty ofextnhepafic sites, mainly by ma=ophages and adipucytet (13). W'~h the recent di~cg~- cry of apo-SAA~, a novel Balb/e SAA molecule, it .became evident tlmt the SAA family i~ more complex in genomic structure, and probably function, than previously ~urm~ed (14). The presence of the claesic acute-phase reactant~ (apo- SAAffi and apo-SAA~) on HDL sugcests that the~e proteins are involved in altezod HDL function during inflammation. It has bran postulatsd that apo-SAA~ produced by macro- pha~es at site~ of inflamnmtion, could locally bind to HDL to effect functional changes (15). We here ~eport that apo-SAA~ is the main SAA family msmber present on normal HDL and suggeat that it might be involved in the normal metaboliam of thi~ partiole. SA.~ haa acut#-phase cheractaristics, and a moderate induction occurs with endotoxln and casein admin. ist~ation. The complete structure of~po-SA~t is deduced from a eDNA clone, and unique differences with respect to other SAPs are identified. Comparative inductive kinetic~ between the varicus SAA family members indicate that apo-SAJ~ dlf[ers di~tlnctly and that its synthesis is both tmn~crlption- ally and translationslly controlled. MATERIALS AND METHODS ArJm~--M.t~ and fema]~ Ba]b/c m~ce, 2 months old, were ob- tsined t'r,~n H,uhn Spragu--Dswl*y Lab0~stodes (Ind]sne~olis. IN). An ecu~-phnse r~pon~e w~s ellclt~l by intrsped~oneal in~sction of 25 ~g of lipopolysacchexlde (LPS) (Esc~.A~. ¢~i 0111:B4. Difco), or I~ scb:utaneous in]cctlon of 0.3 ml of 10% azocasein (U. S. B[o~.em~ce! Co~.). C~ntro! an|m-~s fete/veal m, LPS ~r casein EDTA-a~zic~al~ated b~oDd and c~gans were obta;ned /rcm me:o- f~e-ave~:hetizedonlmals a~ tl~e t~e~ i~ic~*.Pd. O.'[.~ w~e 40023636
Page 33: 40023637
4662 2,fcuse Serum Am:,'b~ A~ Expression Pep~|~. pe~k~ from the ~ ~ endoproteln~e Ly~.C dlze~ti.. ~,,k~m -,,si~ocd ~ each ~el~.& ~eak as detennlned b~ U~ anc~ rat~ ebaraot~i~tic for tho~ two amino ~ (19). p~kPo~don ~, Tq~T~" S~i~-- K5 1-56 666~ 4 6 l)~¢tSlqqtl~vqGTm3L'~ll~~.. K1 67-~ 804 IISTSRK K4 64-89 3132 3 yp~LLmtrt, t~im~a~L~-s~TqK K2 90-98 1019 1 AI~'~I~SOK K3 99-112 1682 NPt~IFRPEGLPEKF E/~cteo/'oc~/n~--In or&r to ezpandtha basic pH ~ of~ (v/v) ~, pH 7~ ~ ~ Bb~, (14). ~ p~ (~ ~) ~ ~ ~ ~ ~ v/v) (18). ~ &~i~ p~J~ ~n ~ ~ (w/v) ~ ma, ~d ~5 Iv/v) 2-~~d ~d ,~~ u ~ (S~kkher ~d ~h~]l) o~z~ at ~m ~a~. The m~br~e w~ wet ~ ~ ~ ~-HC~ pH ~3, 192 ~ 15% (v/v) meal. A~r ~u~ b~, ~e m~e w~ ~er~ ~ con~ g~ (wD) ~ ~ ~" (14). ~en~g for S~ ~o~o~s ~ p~o~ ~ one of~e foUo~¢ ~ ~ti~s: rabbit ~-m~e ~ ~ ~ Dr. S~me~. Univc~ky of Mar~r~ L~, ~Y). An for sik~ir~ ph~phs~e, 5.b~o-4.chlor~3-ind~lyl p~ ~lui~ne ~lt and nl~roblue ~r~ol~um ~Ior~de, we~ c~ to th~ manufacturer's [n;~:t~n~ (~fe TIc~no~c~e~ ]~.), In S~tu I'ro~,:~r ('l,'c~.~ ~[ S~--Fcr L~ ~hu p~tease cleava~. the" ~A.~. fr~.. IO .,g o~ aC~l)h~e Balb/c HDL w¢~ el~trob!otted fl~) c,mc, mlr~w.'lh,],.~(- trod x~cte~ ~ cnzymc~c d:gr~ation described (18). Brlef~. the pro~zin bindi~ sites were blocked with po~vvlnyl~yr'ml~&~c (M, - 4o,ooo) t,e~ed with endaproteisa~ Ly,-O (B~ehs~n~er Mannhetm) estlmt~d #g of ,u~t~e) in 'the p~ence of X-100. I0 u~ tcet~nit,.-iIe, 1o0 nod TdJ-HC4 pH 8~0. Re~pA~s~ HPLC S~n and HPLC ~ z V~ ~ ZI- x I~ ~-p~ ~I~ on Hswk~P~ I~ ~ ~ B~ 4~A P~ ~*~ ~g a ~~* ~d ~e~ ~t~m~ for a ~ ~ ~ ~ ~ (19). eM ~ ~ ~p~ B~ 3~ DNA t~ at ~e M~* ~. O~n~d~w~ ~d ~ ~ (20). R~ ~ P~--Ol~onu~ti~ p~s we~ ~ola~[e~ ~ ~ a ~mp~ ~ p~ w~e ~ ~ ~ o~nucl~ w~ s~ ~ ~e 5" e~ of ~ wu t~nampl~ ~gh the ~l}~er~ c~n ~action ~g94 "C ~or de~tu~o~ 40 "C for ~nesllng, and ~2 "C Cot e~ (Qon- ~h ~ra~d~). A l~-n~cle~de DNA f~ent w~ [~s~ and ~on~ ~o pGEh~ (P~mega) using the ~oRI ~rlc~on site. lba~ I~EM3 contai~d a ps~i~l cl)~JA clone f;r SAA~. aa-d fr~cment wa~ ,~,1 t- ~ roe,, a Bntb/c liver eDNA I~ra~ (MI. 40023637
Page 34: 40023638
Meuse 3~ram Arny~id A~ Expre.~i~n S~5 GCC ACA AGC ~C ~ C~ CGA tAG ¢~ CT& ~C ~G G~ ~ A~ ~ A~ A~ ~G ~G CcA GAG A~ C~ AGT ~ CAC ¢~ I14 ~ A~ c 103~, Clont~¢h L~o~etod.) to olxa~ • ~-len~ ~ne. ~ Ho~ ~). ~he ~I~ ~re b~ ~ va~ For 45 ~ ~t DNA for 16 htt 42 "C ~ 6 X ~, 5 X Den~'a ~u~n, 0.~ ~ p~h~p~, I~ F~ ~mon ~ DNA, 0~ S~. Th= ~ w~ su~y w~ for 20 e~h ~ 2 x ~C, 0A% SDS st ~m ~ra~, ~ea ~ t~ ~uti~ at ~ "C, ~d ~hen in 0.2 x ~C, 0.1~ SDS a~ ~ "C. ~lOa ~0 (U. S. B~hemic~ ~.) via ~ ~y ¢~n ~rmina- ~ m~ of Sa~er m~ c~wozke~ (~). M m~fi~ ~ the A~c~,~+--'F~I IINA was cx~aclcd from t~s+~ by the lithium No~hur:~ hb, t }sybrid~tJ~n aaal~is as ~b~ {~)- No~;ern prolx~l wi~h the radlolabelcd DNA ~'alment wer~ h~ ovep night at 42 "C in a sol~ion ~n~g ~% ~o~, 5 x D~- DN~ 0.1~ ~S (26). S~enUy, ~e b~ w~ w~ for ~ ~ 2 x $~, 0.1% SDS at ~m ~ ~ ~ ~e ~e • o~ st ~ "C. ~ f~y for 20 ~ ~ ~ "C ~ 1 x ~C, O.t~ r~ola~ exon 4-sp~c $~ l~m~ o~o~1~ (13) hybd~ at 42 *C ~ = ~lucton of 6 X SSC, 5 X ~t's 0,1% SDS. I~ #~ml s~mon s~ DN~ 0.05~ ~um ~, and 2~ mM ~F~. The~ blo~ wer~ w~b~ t~ Urn. for ~ ea~ ~ 6 x ~C, 0.1~ SDS at 4S "C ~fom ~[nC e~s~ ~ ~ Run.o~ Tr~c~p~n A~a~--Nu~ei were [~d d~r~d ill) from the Et'ers of no~l B~b/c rake and from live~ of IS'#~ of{ :- "PII'TI~ :- ,at. mCi msL tCN} Ibr ~Omiu. l~l~d 40023638
Page 35: 40023639
A B pl~mid DNA (pGEI~, Promeo), Sb~cenddehyde~-ph~ph~ d~- ,wer~ ~coves~L Since ~0% of a~ bad been ~equenced hydn~n~e (sift from J. D~rne]l), SA~ cDNA, ~nd SA~ On .mnlc_.. DNA immobiltr~i on .~ton membrane (Duralon U~, Stations). l~m~ioudy, the molecular weisht~nd aromatic cont~n~ of the .p~teoly~i¢ ~-e~nents could be pr~iict~d (Table D- The peaks ~,~JLTS AND D|SC~SSION The mo~e SA~ f~ily ~n be ~v~ into ~hr~ flee. ~e F~t ~nsci~s ~e ~s~ic ~-ph~ S~ (apo- S~s, a~-S~,), whkh ~e &~s~cs~y elevs~d acu~-pha~ zespon~ due ~ ~k~e-~ ~pa~ic ~- s~ W be~ mawr apo~p~Jn cons~en~ of HDL (5). ~e ~cond subfamily (a~-S~) is n~ ~cl~[vely ~ th~ Sv;r bu~ abo at e~hepa~c si~; ~nZ infl~on ~d ~ p~sent ~ s minor a~lipopro~ on am~ph~ ~L (13, 15). The third su~famBy ~mp~ ~le~[~ t~ ~i~vdy e~d (S~) and ~e p~nt as the mawr 3~b/c a~-SAAs, a S~ mole~le ~vi~y/denti~ed and ~u~hlly ~equenced by ua (H) be]on~ ~ t~ ~up of ~n~- tu~vely exposed S~s. To further chsr~tefize this novel m~ S~ ~ecul~, w~ have cloned and ~cnced the cD~IA encoding th~ p~tein and inve~a~d ~ ti~e di~ri- hutlon a~ well ~s ~ inductive "H~e high n~leotide humoJo~ b~twcen the rnm~e SAA ~nes (12) ~d the ralativo low abundance of SA~ mP,~A made conventional ©loning techniques ush~ oflgodeo~ucle- otides or SAA cDNAs as probes unsucce~fuL In order to clone SA~, the amino acid sequence of the unique octapep- tide, inser~d relative to the other routine SAAs ~tween amino ~cld~ 69 ~nd V0, w~ determined. This ~gment of S~s - had not pr~vlously been sequenced by us (14) due t~. the extreme hydrophoblclty of the represent~ive peptld? gen'- erated by in s~n~ pmtease d~estion. With the aid of hydrogenated Triton X-IO0 to enhance the recovery of hydrophob|c or long peptld~s f~oro membranes (IS), the peptldes s~narated by the/~ tlon of apo.SAA, electroblott~d onto nlt~caflulo~ w~e sep- s~atedby HPLC (FI~ 1). In comp~konwith stte~ptslacklng the dete~ent, two previously u~obeerved resolyed b~" I~LC using multiple wavelength detection were sub~c~:~ to laser desorption mass spectroa:etry to identi~y the endo-Lys C-generat~t pep~ide iff each. By m~s analysis and • UV ainorbance ratio characte~istlc of tr~pcophan (19). peak K5 was determined to be the amino-terminal peptide 1-56. The reroslninc peak, K4, with • roolecular welsht of' -3100 and absorhance indicative of tyroslne, was identified as the putative peptide containing the 8.amlno acid ircsert and was subkcted to amino ,chi sequence analyai~ It repre- sents amino acids 64-89 (Table I, Fi~. ~) and includes the erpected octapeptide. The octapep~ide o~ Balb/c SA~ NRYYFGIR shares three identities with the octapeptide DYYLFGNS o~ human C-SAA pressed human SAA molecule (28). It L~ in the saroe position as the C-SAA oc~apeptide and shares a double tyro~ine fea- ture. P~r~doxically. oc~peptldes (and a r~nopep~ide in the cow) nre f~und ]z~ th~ ~ame p~i~ion in the ~te.ph~e (A. SAA~) of" t~ d~ f~). cr,~ C-SAA vc~apep~de are all hound by 40023639
Page 36: 40023640
~creen~:l w~th a polyrne~ chain z~ac~/~n-gen~ra~d fr~- pGEM~ ~AA2 A B 185 4~5 c~" A pa~hl ~NA clone, ~M5, ~s kvla~d ~d ~yzed ~i~. 2). It ~ 1415 n~ in t~ ~d ~s of c.>~n~ r~r,, ~e s~p cc~on, ~d 874 refidu~ un~ re~ ~s ~ s~uen~ pm~c~ a prepro- ~ of 13~. Ths le~ ~e is lg ~no ~i~ lon~ which ~ one ~ a~id le~ ~m ~e l~r ~e~ of~-S~x, i~ ~ ~ h~ ~ (~). ~ ~ con- (~. 2). ~on 3 o~ ~ ~ ~ ~ ~ ~ lm ~ exon 2 ~ ~ 4, ~ ~ ~ ~e ~o~ ~ve~. The mo~e S~ ~ ~ ~ek m~me ~ the s~n~ a~~ ~ ~th ~ ~ ~ ~u~ ~ w~ch they ~ ~r~e~ ~ the ~ of ~e ~nse ~d (13), ~C~ ~t ~ ~om ~ ~s S~ ~t ~ re~ ~ the s~ o~ e~e~n pr~ ~ly ~ the ~ver ~d ~e ~ve ~ lc~b ~ at spprox~ly ~ h po~ (2). Ve~ tow leve~ of apo-S~x ~d a~-S~ ~A ~ ~ ~ ~enti- ~ ~ ~ ~ ~r ~S ~~ ~e o~ ~-" • ~s ~NA ~ e~dent ~ ~ fl~ (~). ~n ~n~ ~o- . ~ ~A ~ ~c~t~ e~re~ not o~y ~ ~ ~ver, z0ph~ ~ a~p~ ~ ~n~ ~ ~S (13). elki~ a ~e~nt ~s~ flora ~ ~ ~ it only ~es We ~ N~ blot hyb~n ~ ~ ~o~ ~ for ~e e~n of S~ ~ ~ ~ ~S ~d ~ ~d co~ it ~ ~t ~. ~ S~ ~A t~ hyb~on co,flora ~t o~ ~r "Ma~ Ms~ Un~ th~ co~o~ ~e S~ cDNA ~ not i~n~ wi~ a ra~l~l~ S~ cDNA ~ne p~48 (34). only ~ ~e ~ver of BA~/c mice a~ ~pe~n~ FI~. ~. SA.~. ~otype~ in pl-ama at A tlm~d Intervals after ~ ~atlon. ~ ~t~s ~ p~a ~les ~m BALB/e m~, ~c- f~d at ~ times i~cat~ ~r ad- rian of 7 pg of LPS. ~..~ w~ ide~fied wi~ a rabbit an~- bum~ S~ (05-J~) anti--y, and in B with ra~t anti-m~:us~ AA Oh lh 4h 8h 12h 18h 2-~1h 30h APO-SAA 5 APO'SAA1 APO-$AA2 40023640
Page 37: 40023641
4~66 ~e r~e was ~d wh~ comped ~ ~ #~ of ~. ~e ~ ~o AUG ~1~ ~ p~mt ~ of ~n of S~ co~ ~ ~ p~, ~ ~ifi~ --3 ~ not ~ in ~e ~r but ~ ~nt ~ (1~). The ~a~o~ 12 h ~r ~e ~~n of ~S, at ~eh e~ gene is U~ ~ p~ov~ ~m~ly one-~d of ~e ~ S~ m~A {11). Ho~, w~ ~e le~ of ~ LPS ~is~tlon, the level of S~ m~A ~re rapi~y (11). ~cs of ~ e~ton ~ ~at of S~: (F~ 4, ~e liver of c~tm] ~s, but it ~ be i~nt~ 4 h ~r ~2 ~ It k m~ at p~ ~ up ~ ~ h ~ wh~ it ~du~ly de~es ~ m~ levek ~ h ~r ~k~a~on (Fi~ ~). S~ on ~ o~ hand h ~n ~ et~ula~on ~ ~A b mp~ i~ It ~ ~t ~12 h a~ ~ich l~el it k m~d for ~pm~- ~A level~ ~e ~ hight t~ ~in mn~! ~ ~). ~ ~i~ w~r the ~Me in ~ ~ '~el ~e ~ of ~cre~ ~c~p~on or ~ ~e ~on, w~ ~ffomed n~t~r ~~ (3~ ~ti~ of en~no~ ~NA ~l~er~ ~ ~, ~p~on of s~ci~e gene s~nc~ ~ e[vo (~9). In our DNA ~GEM3), ~e DNA f~m a ~n~e~-ph~e ~ S~ genomic DNA. The ~r would hyb~ ~ ~fie ~NA due ~ ~e nucles~de ~ homolo~e~ ~ ~o~ ~ F~, ~, we find no ~nce ~ ~ of ~e no~-acu~ ph~ ~n~ ~[~e~dehyde-B~ha~ hy~o~en~e in cont~! and ~mul~d ~]ma~. ~e • en~ ~e ~ot tr~scribed in tonal ~i~ls but ~ .n]~d with ~ #g of LP~; the ~zc~pfion of the acu~- ph~ SA~ gen~ iu~asez a~matdy ~-fold ~tion i~ d~ffi~ult du~ to the I~w levd of h~rld~tion in cvntrvl an~m~ L~we]l ez ~]. (II) re~o~d a ~-fo]~ in~ in SAA gent transcr]ptivn after J.))S ~dmlnistmfion). As expected, transcrlp~ion of the SAA~ k~ne d~s ~ur in control in~e~ app~ma~ly 3-fv]~ p=o'v~ng ~er ev~enc~ for ~ c~r~ti~v~ e~m~on of The S~ mRNA p~:~ in ~e ~a~ of m~ ~d~zg~s a z~on ;n ~, ~ t~e due ~ p~Iy{A) ~ mho~g (~, ~). ~ ~o ~ ~ ~W ~m for S~ We ~p~ ~e ~ of ~e ~dS~Aat 4~d~ ~ea~ ~ ~ ~S a~s~fion ~d s~j~ it ~ ~o- foHo~ ~e ~e ps~ ~ t~ ~on o~ ~s ~ve ~I ~ ~he~ ~ a ~a~ ~ ~ ~e ~o~t of S~ ~r ~S a~afion ~g ~ concen~a- ~ns at a~ut 18 h ~r ~S ad~afion ~d ~ently ~ ~e p~ ~ ~ ~en~ in ~e pl~ of ~n~I ~ Apo-S~ sho~ a ~ K The l~ ~t in apo~ o~ at ~2-20 h. T~ eu~e$~ either that ~ by ~-S~x ~d apwS~ ~m ~L ~d ~pi~y ~A ~ ae ~latively d ~ ~NA ~ the v~ ~ ~A n~s ~ ~ c~d~ for ~ ~m of ei~ p~ ~ mol~ on g~n~ {3). It ~ th~ not ~du~ duri~ ~a~n ~l~g A-S~s ~ ~na~. In the mou~ • e ~iH~ ~ ~pond, albeit we~. ~t i~p~c~ on ~L ~ ~er ~ff~n~ or ~placement ~d mp~ ci~. We h~th- ~ ~at apo~ is ~volv~ in ~L ~ ~ revemed ~ ~e p~ of A-~A on the t~ ~ht ~ the ~zation of HDL flow ~w~ si~s of ~fl~a~on eff~g aRe~ations in Hpid flow (42). BEFERENC~ ~i ~." .~ A. 76. I~ZI~ -lbl~ 40023641
Page 38: 40023642
40023642
Page 39: 40023643
J. Lipid Res 1994 IN PRESS SERUM AMYLOID A (SAA): INFLUENCE ON HDL-MEDIATED CELLULAR CHOLESTEROL EFFLUX* CAROLE L. BANKA,§:I: TI YUAN, II MARIA C. de BEER,¶ MARK KINDY,¶ LINDA K. CURTISS,~: AND FREDERICK C. de BEERI** :I:DEPARTMENT OF IMMUNOLOGY THE SCRIPPS RESEARCH INSTITUTE • 10666 NORTH TORREY PINES ROAD LA JOLLA, CA 92037 TELEPHONE: (619) 554-8235, FAX: (619) 554-6146 DEPARTMENTS OF IMEDICINE AND ¶BIOCHEMISTRY UNIVERSITY OF KENTUCKY MEDICAL CENTER LEXINGTON, KY 40536 • *VETERANS ADMINISTRATION MEDICAL CENTER LEXINGTON, KY 40536 Running title: Serum Amyloid A and Cellular Cholesterol Efflux " "Banka, C.L., Curtiss, L.K., Yuan, T., and De Beer, F.C. Acute phase HDL promotes cellular cholesterol efflux. FASEB J 8:A454. §To whom correspondence should be addressed 40023643
Page 40: 40023644
ABSTRACT Normal high density I[poprotein (N-HDL) is remodeled during acute phase (AP) reactions by the association of serum amyloid A (SAA) and the depletion of apoprotein (apo) A-I. To determine the impact of this remodeling on HDL function, the capacities of N-HDL and AP-HDL to associate with and promote cholesterol efflux from human monocytic THP-1 cells were compared. N-HDL and AP-HDL promoted cholesterol efflux from THP-1 cells equally efficiently and in a dose-dependent manner. However, THP-1 cells preferentially bound AP-HDL compared with N-HDL. Examination of the AP-HDL particles bound to THP-1 cells revealed a disproportionate association of an apo SAA- enriched, apo A-I-depleted subpopulation compared with the composition of the starting material. When N-HDL ~ras experimentally remodeled with apo SAA to achieve an apoprotein composition similar to that of the preferentially bound particles, cellular cholesterol efflux was reduced by 30%. The data suggest that apo SAA is bound preferentially to THP-1 cells and that preferential binding of AP-HDL results in near-normal promotion of cholesterol efflux. Key words: Acute phase proteins, apolipoproteins, reverse cholesterol transport 40023644
Page 41: 40023645
INTRODUCTION Threats to homeostasis including injury, infection and inflammation elicit hepatic synthesis of a vadety of acute phase proteins in mammals (for review, see 1). Serum amyloid A proteins (apo SAA), encoded by a multigene family (2), are major acute phase reactants in humans (3). Apo SAAs are small molecules (104 amino acids in humans) that circulate as major apoprotein components of high density lipoprotein (HDL) (4). No definitive function has been identified for the apo SAA proteins. However, the ability of apo SAA to displace apoprotein (apo) A-I, the major apoprotein component o| HDL (4), and the reduction in HDL levels during disease states in humans (4,5) and during induced acute phase reactions in experimental animals (6,7) has led to speculation that apo SAA may alter HDL function and metabolism. This is supported by observations that high plasma levels of apo SAA are associated with diminished activity of the plasma cholesterol esterifying enzyme, lecithin:cholesterol acyltransferase (8) and that the presence of apo SAA reduces HDL affinity for hepatocytes and enhances HDL affinity for macrophages (9). However, the impact of apo SAA on HDL-mediated cellular cholesterol efflux has not been investigated. The removal of cholesterol from peripheral cells is an important component of reverse cholesterol transport by which excess cholesterol is directed from the periphery to the liver for catabolism (10-12). Whereas it is generally accepted that HDL is the major acceptor of cholesterol effluxed from cells, the role of the individual exchangeable apoproteins in this process is the subject o4 controversy. Particles 40023645
Page 42: 40023646
containing apo A-I are the most efficient acceptors of celIular cholasterol; however, apo A-ll arid apo C also can promote cholesterol efflu× (13-15)o Furthermore, a dimer of an 18 amino acid peptide with no sequence homology to the apoproteins was reported to stimulate clearance of cholesterol from cells (16). The common feature of these diverse proteins is their amphipathio alpha helical content (16,17). These observations have led Rothblat and coworkers to propose a receptor-independent interaction between amphipathic helical motifs of apoproteins and specific domains of the plasma membrane as the mechanism underlying apoprotein-mediated cholesterol efflux (18). Studies by Oram and coworkers suggest that the hydrolysis and mobilization of intracellular cholesteryl ester stores leading to cellular cholesterol efflux is initiated through an HDL cell surface receptor (19) and a candidate receptor has been cloned and sequenced (20). We have recently demonstrated specific inhibition of cellular cholesterol efflux to HDL with monoclonal antibodies that bind a region containing two adjacent amphipathic helical repeats in the putative hinged domain of apo A-! (13). This suggests that specific regions of apo A-I are critical for HDL receptor binding or for HDL uptake of free cholesterol. These observations prompted us to explore lhe interactions between apo SAA- containing acute-phase HDL (AP-HDL) and macrophages with emphasis on th~ process of cellular cholesterol efflux. Because HDL particles are dramatically remodeled by the association of apo SAA and the displacement of apo A-I during acute phase reactions, we predicted that the presence of apo SAA would inhibit cholesterol efflux to HDL. To test this hypothesis, we examined the impact of AP-HDL compared to 40023646
Page 43: 40023647
normal HDL (N-HDL) on call association of HDL and on the process of cholesterol efflux from human mon~cyli~ THP-1 ceils. 40023647
Page 44: 40023648
METHODS HDL isolation and iodinatior~ Normal and acute phase plasma samples were drawn from the same donors before and following surgery. AP-HDLand N-HDLwere prepared by sequential ultracentrifugation as previously described (4). The respective HDLs were tudher sub- fractionated into HDL2 (d=1.063-1.13 g/ml), HDL~, (d=-1.130-1.155 g/ml) HDL3b (d=-1.155-1.18 g/ml) and HDL~ (d-=-1.18 g/ml). Lipoproteins were iodinated (lZSl) using a modified iodine monochloride method (4). Following SDS-PAGE separation, Coomassia Blue staining and pyridine extraction, the protein content of each apoprotein was calculated from standard curves of pyridine extracted apoproteins (21). The specific activity of each 12%apoprotein in N-HDL and AP-HDL was also calculated. purification of apo SAA and apo A-I Apo SAA was purified as described (21). Briefly, AP-HDL was delipidated at - 20°C with ethanol/diethyl ether (3:2) and the protein pellet dissolved in 0.02 M-Tris, 0.15 M NaCI, pH 8.4 containing 7 M urea. Dissolved apoproteins were applied to a column (lx100 cm) (Biorad) of Sephacryl S-200 (Pharmacia) and eluted at room temperature in the same buffer. SDS-PAGE analysis of the fractions was perlo~med. Pure apo SAA-containing fractions were dialyzed against 0.002 M Tris, 0.015 M NaCI pH 8.4, lyophilized to one tenth the original voiume and stored at 4°C. Apo A-I was purified similarly except that N-HDL was used as starting material. 40023648
Page 45: 40023649
Preparation of apo SAA-enriched HDL Aliquots containing 2 mg o! N-HDL~were incubated with 7 mg purifi~=d apo SAA (N-HDWSAA) or apo A-I (N-HDI..fAI) at 25°0 for 1 h in 0.02 M-Tris, 0.15 M NaCI, pH 7.4. Two mg aliquots of N-HDI_~ and AP-HDL~ were mock incubated as controls without apoprotein (N-HDL, and AP-HDL,, respectively). The density of the incubation mixtures was adjusted to 1.25 g/ml with solid potassium bromide and the HDL samples were refloated ultracentrifugally (4). Fractions containing HDL were collected and dialyzed extensively against 0.15 M NaCI, 0.002 M EDTA, pH 7.4. Protein concentrations of the particles were determined (22) and individual apoprotein content established by pyridine extraction of Coomassie Blue stained bands from SDS-PAGE as described (21). THP-1 Cell Culture THP-1 (~ells were obtained from the Amedcan Type Culture Collection and maintained in suspension in T-150 culture flasks (Costar, Cambridge, MA) at a cell density of 1.0 - 5.0 × 105/ml in 100 ml of RPMI-1640 containing 10% FCS (Irvine Scientific, Santa Ana, CA), 10 mM Hopes, 2 mM L-glutamine, 1 fi~M sodium pyruvate and 5 x 10"s M ~-mercaptoethanol. Cultures were maF~'~ained at 37°C in 5% CO2. Cells were not used beyond 20 passages; Com.oarative AP-HDI_~and N-HDL= association with THP-1 cells THP-1 cells (8 x 10s) were placed in borosilicate glass tubes with 1.0 ml of serum-free RPMI-1640 culture medium containing increasing amounts of 1251-N-HDL~ or 1251-AP-HDL=. Triplicate samples were incubated for 1 h at 37°C, centrifuged at 4°C 4OO23649
Page 46: 40023650
and the supernatants conta[n[ng unbound '~I-HDL were collected. Pellets were washed three times with coId PBS, 0.2% BSA at 4°0, resuspsnded in one ml of the same buffer and transferred to clean tubes at 4°C. They were again pelleted and washed a fu~her three times with PBS at 4" C. Cell pellets were then dissolved in 1N NaOH for gamma counting and protein measurement (23) or in SDS-sample buffer for SDS-PAGE analysis and autoradiography. Duplicate no-cell control experiments were performed using empty tubes. Cellular degradation of N-HDL~ and AP-HDL3 dudng the 1 h, 37°C incubation was assessed in all samples as described (23). Briefly, cell-free supernatants (1 ml) were precipitated with 12% (w/v) TCA, The TCA-soluble material was extracted with chloroform following oxidation with H202 and counted. Cholesterol Efflux Assays Cholesterol efflux from THP-1 cells was measured by monitoring the appearance in cell culture supernatants of cholesterol synthesized from 14C-acetate. This system has proven effective for quantitating both apo A-I-dependent and diffusional cholesterol efflux (13). Individual experiments were conducted under serum-free conditions by washing the cells three times and culturing them in RPMI-1640 medium in which the 10% FCS was replaced with 1% Nutddoma-HU (Boehringer Mannheim Biocher~icals, Indianapolis). Cells were seeded at 1-2 x 10s cells in 0.5 ml/well in 24-well tissue culture dishes (Costar, Cambridge, MA). ~4C-acetate (New England Nuclear, Boston, MA) was added at a final concentration of 500/~M acetate with a specific activity of 5 mCi/mmol. Each 0.5 ml culture of THP-1 cells received 2-3 t~Ci of 14C-acetate. Cholesterol 40023650
Page 47: 40023651
acceptors (HDL or remoc~eIed HDL) were adde~l after a two-day exposure of cells to '4C-acetate. Fo~towing a further 24-hour culture period, 0.5 ml of trypsin (0.5 mg/ml trypsin, 0.2 mgfml EDTA, Irvine Scientific, Santa Ana, CA} was added to the cultures for 5 minutes at 37=0 to ensure dissociation of SAA-containing HDL particles from the cell surface (23). The cells and supernatants were separated by centrifugation and collected. Cells were frozen for subsequent DNA assays. Supematants were- immediately transferred to tubes containing 1.0 ml of ethanol and stored at 4°C. Supernatant lipids were subsequently extracted into hexane and separated by thin layer chromatography on silica gel 60A plates (250 mm layer, 10 x 20 cm, Whatman, Maidstone, England) with a solvent of isopropyl ether:glacial acetic acid (96:4). cholesterol bands were identified autoradiographically, confirmed as cholesterol by co- migration with 3H-cholesterol, scraped and counted in 2.0 ml of Bio Safe NA (RPi Corp., Mount Prospect, IL). Supernatant ~4C-cholesterol was corrected for recovery based on a 3H-cholesterol internal standard and expressed as cpm/pg of cellular DNA. Cellular DNA Assay All cholesterol parameters were normalized to cellular DNA content to account for~ariability introduced by differences in cell plating, supernatant recoveries or extractions. Cellular DNA was measured using a colorimetric assay as described previously (24). DNA concentration was calculated from a standard curve prepared with calf thymus DNA (Sigma Chemical Co., St. Louis, MO). DNA values within each experiment varied by less than 10% suggesting that treatments with different HDL preparations for 24 h did not affect THP-1 cell division or viability. 40023651
Page 48: 40023652
The phosphotipid content of the particles v:as measured using a cotorimetric kit (WAKO, Osaka, Japan) and calculated from a curve prepared from standards provided with the kit. The cholesterol content of the particles was measured using a fluorometric enzymatic cholesterol assay (25). 10 40023652
Page 49: 40023653
RESULTS Effect ot AP-HDL~_J~:~,.~ on Cellular Cholesterol E|flux To investigate the possible impact of apo SAA on HDL-mediated cellular cholesterol efflux we examined THP-1 cellular responses to N-HDL~ and AP-HDL3 from the same subject. The HDL~ subfracticn was chosen for the studies because it is the predominant apo SAA-beadng particle (4). Dividing "I'HP-1 cells were cultured in the presence of l~C-acetate tor 48 hrs, the period required for a least one cell cycle (26), allowing for representative distribution of t~C-chotesterol among the differenl cellular cholesterol pools. Efflux of newly synthesized '~C-cholesterol was then monitored following a 24-hour exposure to increasing concentrations of HDL. As illustrated in Figure 1, the HDL-mediated efflux of 14C-cholesterol from THP-1 cells was concentralion gependent and equivalent for N-HDL3 and AP-HDL~ although apo SAA accounted for 27% of the AP-HDI~ protein. In both cases, 50 pg/ml of HDL protein induced a two-fold increase in supernatant l~C-cholesterol compared with control. The increase in cholesterol efflux was greater than three-fold in response to N-HDL= and AP-HDL~ at 400 pg/ml. This level of efflux is comparable to that observed with total HDL (13). Previous studies with this cell system have established that appearance of ~4C-cholesterol in the culture supernatants reflects a loss of cholesterol mass from the cells and not just exchange of HDL cholesterol with cellular 14C-cholesterol (13). Association of AP-HDL~L~ with THP-1 cells Figure 2 shows the association of ~SI-N-HDL~ and '251-AP-HDL~with THP-1 cells. 13_ 40023653
Page 50: 40023654
When ceils were exposed to increasing concentrations of ~i-HDL for 1 h at 37~C, signilicant{y more leSi-AP-HDL~ than l"~FN-HDl.~apoproteins associated wilh the cells at all concentrations tested. SDS-PAGE analysis and autoradiography' allowed for identification of the specific cell-associated apoproteins (Figure 3). When the individual apoprotein bands were excised and counted, the ratio of apo A-I to apo A-II in the cell- associated I~-~I-N-HD~ was found to be identical to that of the starting matedal (apo A-I: 70%; apo A-II: 30%} (Figure 3A). In contrast, the cell-associated I~SI-AP-HDL3 apoproteins were markedly enriohed for apo SAA compared with the starting material (Figure 3B}, The apoprotein distribution in the 12SI-AP-HDL3 presented to the cells was apo A-I: 50%; apo A-II: 14% and apo SAA: 36%. The THP-I" cell-associated apoprotein distribution was 27%, 8% and 65%, respectively. The ratio of cell-associated apoproteins remained the same at all concentrations tested. Empty tubes had <1% of associated counts and after prolonged autoradiographic exposure of SDS-PAGE, the AP-HDL~ material from these no-cell controls resembled the starting material indicating that the selective apoprotein distribution was cell specific. Degradation of both cell-associated N-HDL~ and AP-HDL3, assessed as the recovery of TCA soluble radioactivity, was less than 2% at all concentrations, suggesting that little internalization of HDL occurred during the one-hour incub&tions al 37~C. Effect of SAA-conteining particles on cholesterol efflux from THP-1 cells Despite Ihe apparent lack of inhibition of cholesterol efflux by apo SAA in initial experiments (see Figure 1), the demonstration of selective binding of the apo SAA-rich ~2 40023654
Page 51: 40023655
subfract[on of AP-HDL~ to THP-1 ceils (see Figure 3B) prompted us to examine the effect of HDL preparations with higher proportions of apo SAA protein. This was accomplished by remodeling N-HD~ with purified apo SA,k in vitro and reisolating the particles by ultracentrifugation. Apo A-! remodeled HDL (N-HDUAI) and reisolated (or "refloated") N-HDL= (N-HDL.) and AP-HDLo (AP-HDL,~) served as controls for the remodeled apo SAA particles (Figure 4B). Additionally, we subfractionated the N-HDL~ and AP-HDL3 by gradient ultracentrifugation (4) (Figure 4A), characterized the lipid and protein content of the particles, and tested them in the cholesterol efflux assay. Lipid and protein parameters for the preparations are summarized in Table I, As previously demonstrated (4), the apo SAA-enriched particles hadhigher protein:cholesterol ratios than the paired non-SAA-containing controls in all cases (compare N-HDL~= with AP- HDL~, HDIEAI with HDL/SAA, etc,, in Table I). In all but one case, N-HDL~ versus AP- HDL~, in which-limited material allowed only for a single assay sample, the apo SAA- containing HDLs had higher protein:phospholipid ratios than the matched N-HDL controls; however, the cholesterol:phospholipid ratios remained essentially unchanged. As expected, when these particles were added to THP-1 cells on the basis of their lipid content (60 pg/ml phospholipid), the N-HDL~, the "refloated" N-HDL3.~ (N-HE)L,) and the apo A-I remodeled particles (N-HDL/AI) induced cholesterol efflux in a similar manner with a greater than six-fold increase over control (Figure 5), Of the non-SAA-containing HDLs, only the N-HDL~ was less effective. However, in all cases the apo SAA- containing HDLs were less efficient than the matched N-HDL controls at promoting cholesterol efflux. The HDL preparation with the highest apo SAA content, N-HDL/SAA, 40023655
Page 52: 40023656
was characterized by the greatest inhibition of cetIular chotesterol efflux {30% decrease in efllu×, p<0.031, compared with N-HDIEAI). These results indicated that the presence of apo SAA en HDL had little impact on cellular cholesterol efflu× unless apo SAA represented more than 55% of the_HDL protein. 40023656
Page 53: 40023657
DISCUSSION Epidemiolagic data has revealed an inverse re!ationship between HDL levels and the risk of cardiovascular disease. This relationship applies to levels of both HDL cholesterol and apo A-I (27,28). Any decrease in HDL or disturbance in HDL composition may compromise cholesterol homeostasis and ultimately increase the risk of atherosclorosis. The acute phase response is characterized by both a decrease in HDL (5-7) and a displacement of apo A-! by apo SAA on HDL (4). The importance of thoso changes in HDL may be reflected in the clinical observation that pationts who experience these acute phase changes on an ongoing basis such as thoso with rheumatoid arthritis, have a high incidence of cardiovascular disease (29). We have recently demonstrated that HDL-mediated cholesterol efflux from peripheral cells'can be inhibited in part by antibodies specific to apo A-I (13). Therefore, we predicted that displacement of apo A-I by apo SAA on HDL from acute phase patients could decrease the effectiveness of the HDL in promoting cellular cholesterol efflux. The data presented here indicate that HDL efficiency in promoting cholesterol efflux was compromised only when the apo SAA content of the particles exceeded 50% of the total HDL protein. The finding that THP-1 cells selectively bound an apo SAA-onriched, apo A-I depleted subpopulation of particles from AP-HDL~ confirmed similar findings of selective AP-HDL= binding to neutrophils (2;3) and macrophages (9). Because the ratio of AP-HDL~ apoproteins bound to cells remained the same at all concentrations tested, 40023657
Page 54: 40023658
this selective binding is thought to reflect the preferential binding of a particular apo SAA-enriched subset of HDL particle~ rathgr than surface remodeling of the ceil- associated AP-HDL~ or transfer of apo SAA to the cell surface. Whatever the mechanism, the multiple equilibria precluded Scatchard-type analyses of this binding data. Whereas the mechanism of selective association of AP-HDI~ with THpol cells is not yet understood, the observation is important in reference to our cholesterol efflux data. Although significantly more AP-HD~ apoprotein was found associated with the cells at each concentration, the apo SAA-enriched particles were substantially more dense than the mean HDL (Table 1). Thus the number of particles bound may have been very similar despite the increase in protein bound. The presence of apo SAA appears to negatively influence the cell association of apo A-I (Figure 3) suggesting that apo SAA may bind the same cell surface site as apo A-I (and apo A-II). Neither the selective association of the apo SAA-enriched particles nor the concomitant decrease in apo A-I binding resulted in dramatic reduction of cellular cholesterol efllux. These studies do not permit a conclusion as to the nature of the binding site for the apo SAA-enriched particles. The binding may represent a lipid-lipid interaction, a non-specific apoprotein-lipid interaction such as that proposed by Rothblat an.el coworkers for normal HDI.. (18) or a specific receptor interaction. Several putative cell surface receptors for apo A-I have been identified; one on peripheral cells that presumably mediates cholesterol efflux (20,30) and two on liver membranes that presumably mediate HDL delivery of cholesterol to the liver (31,32). Recently, Vadiveloo and ¢oworkers (32) have demonstraled that apo A-II binds these same 40023658
Page 55: 40023659
receptors. The binding of apo A-I and apo A-II to the liver membrane proteins is proposed to involve ~-hel[cal regions in the carbo×'y-terminal of each apoprotein (32,33) that share sequence i~omology (34). Apo SAA has significant alpha-helical structure when associated with phospholipid (35,36) and may bind in a similar fashion.. Several interpretations of the cholesterol efflux data are relevant, the most obvious being that HDL-mediated cellular cholesterol efflux is dependent upon the presence of apo A-I and, therefore, will occur less efficiently when the number of apo A- I molecules per particle drops below a certain threshold. However, particles that contained as little as 4% apo A-1 promoted cholesterol efflux only 30% less efficiently than N-HDL3 (see N-HDL/SAA, Table I and Figure 5). Because the molecular mass of apo A-I is 2.3 times that of apo SAA, the molar ratio of apo SAA to apo A-I in the N- HDL/SAA preparation was 50:1. Therefore, our data could suggest that, like apo A-I! and apo C-Ill, apo SAA can promote cholesterol efflux although less efficiently than apo A-I. Alternatively, the diminished capacity of the SAA-enriched HDL to promote cholesterol efflux may be a function of the decreased lipid:protein ratio. (Table 1). The impact that these findings may have on the development of atherosclerosis during the chronic acute phase situation remains to be established. Although apo SAA can constitute up to 80% of HDL apoproteins in extreme circumstances (37), it ~'arely exceeds 50%, the concentration at which cellular cholesterol efflux can be impaired. Our data support the contention that a number of HDL apoproteins can share the capacity to associate with cells and to mediate cholesterol efflux. This sharing of function is most likely based on similarities in the tertiary structure of the proteins and 40023659
Page 56: 40023660
reflects th~ common characteristic of atpha amphipati~ic helical domains. !8 40023660
Page 57: 40023661
ACKNOWLEDGEMENTS The authors wish to thank Audrey Black and Winnie Lee for technical assistance, and Anna Mayers for preparation of tha manuscript. This work was supported by NIH grants HL-50060 (CLB}, HL-43815 (LKC), AR:40379 (FCdeB) and the Council for Tobacco Research (FCdaB). This is manuscript #8576 from the Department of Immunology, The Scripps Research Institute. 40023661
Page 58: 40023662
REFERENCES 1. Ste-~l, D. M. and A. S. Whitehead. 1994. Tt~ major acute phase reactants: C-reactive protein, serum amyloid P component and s.~rurn amyloid A protein, fmmunoL Teday 15:81-88. 2. Strachan, A. F., W. F. Brandt, P. Woo, D. R. van der Westhuyzen0 G. A. Coetzee, M. C. de Beer, E. G. Shepard, and F. C. de Beer. 1989. Human serum amyloid A protein. The assignment of six major isot(~rms to three published gene sequences and evidence for two genetic loci. J. BioL Chem. 264:18368 3. Malle, E., A. Steinmetz, and J. G. Raynes. 1993. Serum amyloid A {SAA): an acute phase protein and apolipoprotein. Atherosclerosis 102:131-146. 4. Coetzee, G. A., A. F. Strachan, D. R. van der Westhuyzen, H. C. Hoppe, M. S. Jeenah, and F. C. de Beer. 1986. Serum amyloid A-containing human high density lipoprotein 3. Density, size, and apolipoprotein composition. J. BioL Chem. 261:9644-9651. 5. Rosenson, R. S. 1993. Myocardial injury: the acute phase response and lipoprotein metabolism. J. Am. Coil. CardloL 22:933-940. 6. Cabana, V. G., J. N. Siegel, and S. M. Sabesin. 1989. Effects of the acute phase response on the concentration and density distribution of plasma lipids and 2o 40023662
Page 59: 40023663
apoIi~oproteins. J. L~pfd Res. 30:39-49. 7. Holfmari, J. S. and E. P. Benditt. 1982. Changes in high density I]poprotein content following endotoxin administration in the mouse. FormatiQn of serum amylQid protein-rich subfractions. J. Cell BioL 257:10510-10517. 8, Steinmetz, A., G. Hocke, R. Saile, P. Puchois, and J. -C. Fruchart. 1989. Influence of serum amyloid A on cholesterol esterification in human plasma. Biochim. Biophys. Acta 1006:173-178. 9. Kisilevsky, R. and L. Subrahmanyan. 1992. Serum amyloid A changes high density lipoprotein's cellular affinity. A clue to serum amyloid A's principal function. Lab Invest. 66:778-785. 4- 10. Glomset, J. A. 1968. The plasma lecithin:cholesterol acyltransferase reaction. J. Lipid Res. 9:155-167. 11. Johnson, W. J., F. H. Mahlberg, G. H. Rothblat, and M. C. Phillips. 1991. Cholesterol transport between cells and high-density lipoproteins. Biochim. Biophys. Acta 1085:273-298. 12. Pieters, M. N., D. Schouterl, and T. J. C. Van Berkel. 1994. In vitro and in vivo 21 40023663
Page 60: 40023664
evidenc~ for the ro~e of HDL in revers~ cholesterol transport. B[~chfm. B[ophys. Acta MoL Basis Dis. 1225:125-134. 13. Banka, C. L, A. S. Black, and L. K. Curtiss. 1994_. Localization of an apolipoprotein A-I epitope critical for lipoprotein-mediated cholesterol efflux from monocytic cells. J. Bio/. Chem. 269:10288-10297. 14. Mahlberg, F. H. and G. H. Rothblat. 1992. Cellular cholesterol eftlux: role of cell membrane kinetic pools and interaction with apolipoproteins AI, All, and Cs. J. BioL Chem. 267:4541-4550. 15. Mahlberg, F. H., J. M. Glick, S. Lund-Katz, and G. H. Rothblat. 1991. Influence of apolipoproteins AI, All, and Cs on the metabolism of membrane and lysosomal cholesterol in macrophages. J. BioL Chem. 266:19930-19937. 16. Mendez, A. J., G. M. Anantharamaiah, and J. F. Cram. 1992. Synthetic amphipathic helical peptides that stimulate clearance of cholesterol from cells. Circulation 86:1-7 17. Segrest, J. P., M. K. Jones, H. DeLoof, C. G. Brouillette, Y. V. Venkatachalapathi, and G. M. Anantharamaiah. 1992. The amphipathio helix in the exchangeable apolipoproteins: a review of secondary structure and function.. J. Lipid Res. 33:141-166. 22 40023664
Page 61: 40023665
18. Rothblat, G. H., F. H. Mahlberg, W. J. Johnson, and M. C. Phillips. 1992. Apotipoproteins, membrane choI~stero! domains and lhe regulalion of cholesterol efflu~c J. l.J'pid Res. 33:1091-1097. 19. Oram, J. F., A. J. Mendez, J. P. Slotte, and T. F. Johnson. 1991. High density I[poprotein apolipoproleins mediate removal of sterol from intracellular pools but not from plasma membranes of cholesterol-loaded fibroblasts. Arterioscler. Thromb. 11:403-414. 20. McKnight, G. L., J. Reasoner, T. Gilbert, K. O. Sundquist, B. M. Hokland, P. A. McKernan, J. Champagne, C. J. Johnson, M. C. Bailey, R. Holly, P. J. O'Hara, and J. F. Oram. 1992. Cloning and expression of a cellular high density lipoprotein-binding protein that is up-regulated by cholesterol loading of cells. J. BioL Chem. 267:12131-12141. 21. Godenir, N. I., M. S. Jeenah, G. A. Coetzee, D. R. van der Westhuyzen, and F. C. de Beer. 1985. Standardization of the quantitation of serum amyloid A protein (SAA') in human serum. J. ImmunoL Methods 83:217-225. 22. Lowry, Oo H., N. J. Rosenberg, A. L Farr, and R. J. Randall. 1951" Protein measurement with the folin phenol reagent. J. Biol. Chem. 193:265-275. 23. Shephard, E. G., F. C. de Beer, M. C. de Beer, M. S. Jeenah, G. A. Coetzee, and D. R. van der Westhuyzen. 1987. Neutrophil association and degradation o! normal and acute 23 40023665
Page 62: 40023666
phase high density lipoprotein. Bfochem. J. 248:919-926. 24. Labarca, C. and K. Pargen. 1980. A simpte, rapid, and sensitive DNA assay. Anal Biochem. 102:344-352. 25. Gamble, W., M. Vaughan, H. S. Kruth, and F. Avigan. 1978. Procedure for determination of free and total cholesterol in micro- or nanogram amounts suitable for studies with cultured cells. J. Lipid Res: 19:1068-1057. 26. Banka, C. L., A. S. Black, C. A. Dyer, and L. K. Curtiss. 1991. THP-1 cells form foam cells in response to coculture with lipoproteins but not platelets. J. Lipid Res. 32:35-43. 27. Buring, J. ~..; G. T. O'Connor, S. Z. Goldhaber, B. Rosner, P. N. Herbert, C. B. Blum, J. L. Breslow, and C. H. Hennekens. 1992. Decreased HD~ and HDL3 cholesterol, apo A-I and apo A-II, and increased risk of myocardial infarction. Circulation 85:22-29. 28. Stampfer, M. J., F. M. Sacks, S. Salvini, W. C. Wilier, and C. H. Hennekens. 1991. A prospective study of cholesterol, apolipoproteins, and the risk of myocardial infar(~tion. N. Engl. J. Med. 325:373-381. 29. Wolfe, F., D. M. Mitchell, J. T. Sibley, J. F. Fries, D. A. Bloch, C. A. Williams, P. W. Spitz, M. Haga, S. M. Kleinheksel, and M. A. Cathey. 1994. The mortality of rheumatoid 2~ 40023666
Page 63: 40023667
arthritis. Arthritis Rheum 37:481-494. 30. S~otte, J. P., J. F. Oram, and E. L. B[erman. 1987. Binding of high density lipoproteins to cell receptors promotes translooation of cholesterol from intracellular membranes to the cell surface. J. BioL Chem. 262:12904-12907. 31. Tozuka, M. and N. Fidge. 1989. Purification and characterization o4 two high-density-lipoprotein-binding proteins from rat and human liver. Biochem. J. 261:239-244. 32. Vadiveloo, P. K., C. M. Allan, B. J. Murray, and N. H. Fidge. 1993. Interaction of apolipoprotein All with the putative high-density lipoprotein receptor. Biochemistry 32:9480-9485: 33. Morrison, J., N. H. Fidge, and M. Tozuka. 1991. Determination of the structural domain of apoAI recognized by high density lipoprotein receptors. J. BioL Chem. -~56:187B0-18785. 34. Allan, C. M., N. H. Fidge, and J. Kanellos. 1992. Antibodies to the carboxyl terminus of human apolipoprotein A-I. The putative cellular binding domain of high density lipoprotein 3 and oarboxyl terminal structural homology between apolipoproteins A-I and A-II. J. BfoL Chem. 267:13257-13261. 25 40023667
Page 64: 40023668
35. Segrest, J. P., H. J. Pownall, R. L. Jackson, G. G. G1snner, and P. S. Pollock. 1976. Amyloid A: amph~pathic helixes and lipid binding. Biochemistry 15:3187-3191. 36. Baussarman, L. L., P. N. Herbert, T. Forle, R. D. Klausner, K. P. W. J. McAdam, J. C. Osborne, and M. Rosseneu. 1983. Interaction of the serum amyloid A proteins with phospholipid. J. Biol. Chem. 258:10681-10688. 37. Strachan, A. F., F. C. de Beer, G. A. Coetzee, H. C. Hoppe, M. S. Jeenah, and D. R. van der Westhuyzen. 1986. Characteristics of apo-SAA containing HDL~ in humans. Protfdes BioL Fluids 34:359-362. 26 40023668
Page 65: 40023669
FOOTNOTES Abbreviations used in this paper: apo SAA, apoprotein serum amyloid A; apo, apoprotein; AP-HDL, acute phase HDL; N-HDL, normal HDL. 2"/ 40023669
Page 66: 40023670
TABLE 1. Apoproteln and Lipid Distribution Among HDL Particles Particles Apo A-I" Apo A-II" SAA* Phospholipidt Cholesterolt (%) (%) (%) N-HDL~ 71 AP-HDI_~56 N-HDL~ 82 AP-HDI..~ 39 N-HDL~= 69 AP-HDL,t58 N-HDUAI 73 N-HDLJSAA 4 29 608 ± 14 125 ± 11 17 27 570 ± 13 102 ± 2 18 331§ 81 ± 1 6 55 411~ 56 ± 3 31 636± 1 15±35 17 25 589 ± 39 131 ± 16 27 608 ± 14 149 ± 11 10 86 383 ± 39 70 ± 8 * % total protein 1"/~g/mg protein, mean ± S.D. :l: particles refloated by ultracentrifugation § sample limitation resulted in single determinations 28 40023670
Page 67: 40023671
FIGURE LEGENDS Figure 1: Dose dependence of HDL-mediated cellular cholesterol efflux. THP-1 cells were cultured in serum-free medium at 1.5 x 105 cells per 0.5 ml in 24-well plates. Following a 2-day exposure to 14C-acetate, the cells were incubated with medium (0 mg/ml) or increasing concentrations of normal (N-HDL=) or acute phase (AP-HDL~) HDL=. After 24 h, supernatant 14C-cholesterol was extracted and measured as described in Methods. Values are normalized to DNA content of the cells and represent the mean ± S.D. of four replicate cultures. Figure 2:THP-1 cellular association of 12SI-N-HDL3 and 12SI-AP-HDL3 apoprotein. THP-1 cells (8 x 105/borosilicate glass tube) were exposed to increasing concentrations of 1251-HDL for I h, pelleted, washed extensively, dissolved in 1N NaOH and counted." Each concentration point represents mean -,-S.D. of triplicate determinations. Figure 3: Preferential association of a subpopulation of apo SAA-enriched AP-HDL parlicles with THP-1 cells. Autoradiograph of cell-associated lzSI-HDL separated on reduced 4-30% gradient SDS-PAGE. Samples were from experiments conducted as described in Figure 2. Lanes 1-6 represent the cell-bound apoprot~ins following incubation with 2, 5, 10, 20, 60, 200 microgram of ~zSI-N-HDL~ (A) or I=SI-AP-HDL3 (B) proteirdml respectively. Lanes S contain 5 pg of HDL protein from standard preparations of '~I-N-HDL~ (A) or ~2SI-AP-HDL.~ (B). Illustrated is one representative 29 40023671
Page 68: 40023672
e×p~riment ef two. Figure 4: Electrophoret[c characterization of HDL particles used in cellular chotestero! efflux assays. Photographs of Coomassie Blue stained reduced 4-30% gradient SDS- PAGE. Panel A: HDL~ uitracentrifugally subfractionated. Lane 1, N-HDL standard; lane 2, N-HDL~; lane 3, N-HDI_~; lane 4, N-HDL~; lane 5, AP-HDL~,; lane 6, AP-HDL=~; lane 7, AP-HDL:~; and lane 8, AP-HDL= standard. Panel B: remodeled apo SAA-enriched and control particles. Lane 1, N-HDL~, refloated (rf); lane 2, apo A-I enriched N-HDL~ (N-HDLEAI); lane 3, apo SAA-enriched N-HDL=, (N-HDL/SAA);; and lane 4, AP-HDL3 refloated Cd). Figure 5. Effect of apo SAA-poor and apo SAA-enriched particles on cholesterol efflux from THP-1 cells. THP-1 cells were cultured and supematant 14C-cholesterol was quantitated as described for Figure 1. Values are normalized to DNA content of the cells and represent tl~e mean ± S.D. of four replicates. Cells were exposed for 24 hr to medium alone (open bar), normal and acute phase HDL subpopulations 3a and 3c. reisolated normal and acute phase HDL3,(rf) or normal HDL~ remodeled with apo A-I or apo SAA. a) p<0.01 compared with matched control; b) p<0.001 compared witl~ matched control. 3O 40023672
Page 69: 40023673
• ~5000 0 ~ 10000 5000 20000 --/'/ • N-HDL3 - [] AP-HDL3 0 50 100 200 400 HDL (pg/ml) Figure I 40023673
Page 70: 40023674
1.0' 0.0 0 I T" IIII I I 200 400 600 800 000 HDL (pg/ml) Figure 2 _32 40023674
Page 71: 40023675
A S=S 1 ~, 2 , 3 4 S 6 ~=..,-*-. °.-Apo A-I -Apo -'Apo SAA $ $ Figure 3 33
Page 72: 40023676
A 1 B 2 3 4 5 6 7 ~ - Apo A-I ~- Apo SAA ~- Apo A-II 8 - Apo A-I 3 4 -/~po SAA - po A-II Figure 4 34
Page 73: 40023677
C~ IOJ~,uoo e£ "IC]H'N e~ "laH-dV o¢ "IQH-N o~: "IQH-dV .P "IQH-N .P "IQH-dV ~ IV I "IQH-N WS I "IGH-N Supern'atant ..C-Cholesterol (CPMIpg DNA) I I I
Page 74: 40023678
J. Lipid Res 1994 IN PRESS CHARACTERIZATION OF CONSTITUTIVE HUMAN SERUM AMYLOID A PROTEIN (SAA4) AS.AN APOLIPOPROTEIN *De Beer, M.C., "D/'uan, T., *Kindy, M.S., §Asztalos, B.F., §Roheim, P.S. and t-$De Beer, F.C. Departments of *Biochemistry and ~'Medicine, University of Kentucky, Lexington, K¥ and the ~:Veterans Administration Hospital, Lexington, KY: Department of §Physiology, Division of Lipoprotoin Metabolism and Pathophysiology, Louisiana State University Medical Center, New Orleans, LA. Running Title: SAA4 as an Apolipoprotein To whom correspondence should be addressed: Dr. F. C. de Beer Department of Medicine J515, Kentucky Clinic Limestone Street Lexington, KY 40536 Tel: (606) 323-6700 Fax: (606) 323-1020 Abbreviations: HDL: high density lipoprotein; LDL: low density lipoprotein; VLDL: very low density lipoprotein; IEF: isoeleetde focusing; SAA: serum amyloid A protein. 40023678
Page 75: 40023679
,ABSTRACT Serum amyioid A proteins (SAAs), a family of homologous molecules, are apolipoproteins of HDL. They can be divided into two groups. The first group comprises the well-characterized acute phase SAAs that associate with HDL during inflammation, thereby remodeling the HDL particle by displacing apo-AI. The second group consists of the recently discovered constitutive SAAs -- mouse SAA5 and human SAA4. They exist as minor apolipoproteins on HDL but constitute more than 90% of the total SAA during homeostasis. We have characterized human SAA4 as an apolipoprotein. During homeostasis SAA4 is. synthesized only in the liver. Purification of SAA4 has been described and its plasma concentration established at 55 _+ 13/zg/ml in 26 healthy individuals. It was present on all HDL density classes and VLDL but was absent from LDL. Using two-dimensional eleetrophoresis and phosphofimaging, SAA4 was found to be associated with a specific subpopulation of only three HDL particles. These HDL particles are not involved in the initial cholesterol transfer from cells. SUPPLEMENTARY KEY WORDS: high density lipoprotein, acute phase 40023679
Page 76: 40023680
INTRODUCTION High density lipoprotein (HDL) plays a central role in Iipid metabolism by continuously exchanging components with cells and other iipoproteins (1). HDL particles display a dynamic polydispersity with respect to size, hydrated density and apolipoprotein composition (2). Apolipoproteins fulfill important biological roles by acting as tigands for receptors or co-factors for enzymes (3). Serum amyloid A proteins (SAAs), a family of homologous molecules, are all apolipoproteins of HDL (4,5). They can be divided into two groups (6). The first group includes the well-characterized classical acute phase SAAs that increase dramatieally during an acute phase response due to cytokine-driven hepatic synthesis (7). They displace apolipoprotein AI (apo-AI) with resultant remodeling of HDL, yielding larger particles with a higher hydrated density (2); even becoming the major apolipoprotein component of acute phase HDL (8). Th~ second group comprises the recently discovered constitutively expressed SAAs, namely mouse SAA5 (9) and human SAA4 (6). They exist as minor apolipop~teins of normal HDL comprising 1%-2% of the total apolipoprotein component during homeostasis (6,9). Four human SAA genes are located on chromosome 11 (10,11). The acute phase SAAs are encoded by two genes, SAA1 and SAA2. Allelie variation at these two loci accounts for the acute phase SAA isoforms identified to date (12,13). The locus designated SAA3 appears to be a pseudogene because of the presence of an extra base in exert 2 (14). This is corroborated by the fact that 40023680
Page 77: 40023681
2 neither the protein product rtor message for this gene has been identified. The SAA4 locus identified recently encodes the constitutively expressed SAA4 (10). In this paper, we localized and characterized SAA4 on HI)L s.ubpopulations, a prerequisite for future functional analyses. We define the synthesis site of SAA4 and describe techniques for its purification. During the acute phase response, when the eytokine-indueible SAAs dominate on a molar basis, SAA4 remains associated with HDL as a minor apolipoprotein. Using two- dimensional eleetrophoresis and phosphorimaging, SAA4 is found to be associated with a distinct subclass of HDL particles unrelated to those involved in the initial cholesterol transfer from cells (15). ,,MATERIAI~,S Al~D ,METHODS Pre, paration, of Lipoprote, ins Lipoproteins were isolated by sequential ultracentrifugation from the blood of healthy donors or from patients experiencing an acute phase response after surgery (16). HDL was further subfraetionated according to plasma density (p); HDI-/2 (p=1.063-1.13 g/ml), HDL3A (p=1.13-1.155 g/ml), HDL3B (p=1.155-1.18 g/mi) and HI)L3c (p>l.18 g/ml) by reeentrifugation of the total HDL fraction into a linear KBr gradient as described previously (2). .P..ur.ification. of SAA,, SAA4 was isolated from HDL by Rotofor (Biorad, Richmond, CA, USA) preparative isoelectdc focusing as per the manufacturer's instructions. Briefly, 40023681
Page 78: 40023682
3 200 mg of normal HDI, was delipidated with ethanol:ether (3:2, v/v) and protein pellets suspended in 15 ml 8 M urea, 1% (w/v) decyl sodium sulphate (Eastman Kodak Co., Rochester, NY, USA) and 5% (v/v) 2-mercaptoethanol. The Rotofor running buffer consisted of 8 M urea, 1.2% Ampholines pH 4-6.5, 1.2% Ampholines pH 7-9 and 0.6% Ampholines pH 3-10. The anionic and cationic chambers contained 0.1 M phosphoric acid and 0.1 M sodium hydroxide respectively. The sample was eleetrofoeused at constant power (12W) until equilibrium was reached. Twenty fractions were harvested and analyzed by SDS- PAGE. Those containing SAA4 were pooled and subjected to molecular sieve chromatography to separate contaminating apelipoproteins. Briefly, the SAA4- containing fractions were dialyzed against 15 mM NaCi, 2 mM Tris/HCI pH 8.4 and lyophilized. The lyophilized pellet was suspended in 2 rnl 7 M urea, 150 mM. NaCi, 20 mM Tris/HCI pH 8.4 and subjected to molecular sieve chromatography on a I X 120 cm Sephaeryl $200 column as per the manufacturer's instructions (Pharmaeia LKB Bioteehnology, Piseataway, N J, USA) (17). SAA~. Assay SAA4 was measured in plasma samples with an immunoradiometric method using rabbit anti-human SAA4 antibody as described for inflammatory SAAs (18). A standard curve was obtained by using SAA4-enfiched HDL instead of purified SAA4 that is relatively insoluble and aggregates in solution. Th.is was prepared by incubating 1 mg purified SAA4 with 1 mg normal HElL in 20 mM Tris/HCl pH 7.4, 150 mM NaCI for 1 h at 22" C with gentle shaking. Th.e HDL was 40023682
Page 79: 40023683
separated from the free SAA4 by ultmcentrifugal flotation (2). Aliquots of this SAA4-enriched HDL were subjected to SDS-PAGE using a 5% ~o 20% acrylamide gradient, and the Coomassie-stained SAA4 bands were excised and quantitated by pyridine extraction of-the dye as described (18). Electrofocusing Aliquots (50-400pg) of lipoproteins were freeze-dried and delipidated with 0.5 ml chloroform:methanol (2:1, v/v) (19). The delipidated proteins were suspended in sample buffer consisting of 8 M urea, 1% (w/v) deeyl sodium sulphate (Eastman Kodak Co., Rochester, NY, USA) and 5% (v/v) 2- mercaptoethanol. Samples were electrofoeused on 0.3 mm polyacrylamide gels containing 7 M urea and an Ampholine gradient consisting of 20% (.v/v) Ampholines pH 3-10, 40% (v/v) Ampholines pH 4-6.5 and 40% (v/v) Ampholine.s" pH 7-9 (Pharmacia-LKB Biotechnoiogy, Piseataway, N J, USA) as described (16). Immunochemieal Analysis The SAA isoform distributions of lipoproteins were investigated by means of immunoehemieal analyses (16). Fifty micrograms of various lipoproteins were freeze-dried, delipidated and subjected to isoeleetrie focusing as described above. Samples on eleetrofoeused gels were pressure-blotted onto 0.2 Fm-pore-size nitrocellulose membranes (Schleieher and Sehuell, Keene, NJ, :USA) for 20 h at room temperature. The membrane was wetted with 25 mM Tris/HCl, pH 8.3, 192 mM glyeine and 15% (v/v) methanol. Following pressure-blotting, membrane binding sites were blocked overnight at 4 "C with 5% (wlv) non-fat 40023683
Page 80: 40023684
5 dry milk in PBS containing 2% (w/v) BSA. Screenings for SAA isoforms were performed with a l:I000 dilution of one of the following antibodies: our rabbit anti-human SAA4 antibody (17), or a monoclonal anti-human SAA4 antibody, or a monoclonal anti-human AA antibody. An alkaline phosphatase-conjugated goat anti-rabbit lgO antibody was used as secondary antibody (A8025, lot no. 39F- 88961; Sigma Chemical Co., St. Louis, Me, USA). The ehromogenie substrates for alkaline phosphatase, 5-bromo-4-ehloro-3-indolyl phosphate p-toluidine salt and nitroblue tetrazolium chloride (Bethesda Research Laboratories Life Teehnologies, Bethesda, MD, USA), were applied according to the manufacturer's instructions. Northern Blot Hybridization, ,bmalysis Human multiple tissue Northern blots containing poly A+ RNA from a variety of human tissues, as well as peripheral blood ieukocytes, (#7760-1 and 7759-1, Clonteeh Laboratories, Inc., Pale Alto, CA, USA) were probed with our radiolabeled SAA4 eDNA clone according to the manufacturer's instructions. Radiolabeled B-aetin eDNA was used as a control probe. The blots were washed according to the manufacturer's instructions prior to exposure to film. Two-Dime~nsional Sepa ,ra,ti0n of,,. HD, L Particles Two-dimensional separation of HDL particles was achieved as previously published (20). In the first dimension, plasma lipoproteins were separated by charge using agarose eleetrophoresis. In the second dimension, a separation by 40023684
Page 81: 40023685
size was achieved using 3% to polyaerylamide gel electrophoresis. 6 36% non-denaturing concave gradient Separated lipoproteins were transferred to a nitrocellulose membrane, localized by labeled monospeeifie antibodies, and quantitate.d by phosphorimaging (Phosphorimager.SF, Molecular Dynamics, Sunnyvale, CA, USA). Data were expressed as pixel points by computer analysis and were linearized with the dpm of the l:25I-labeled antigen-antibody complexes. Radioactivity was integrated by the Molecular Dynamics Image QuantTM computer program. This method is capable of resolving several apo-AI-containing HDL subpopulations, including subclasses of or- and pre-ct-migrating HDL, and a number of pre-B-migrating HDL subclasses. Particles weredefined according to their relative Rf to albumin (first dimension) and their size (second dimension). Immunol~recil~itation of S..AA4.-,co.ntaining HDL Particles SAA4-containing HDL particles were isolated from fresh human plasma by immunoabsorption. Briefly, 8 mg lgG fraction from rabbit anti-human SAA4 antisera was coupled via its carbohydrate moieties to A.ffi-g©i HZ (~-i'brad Laboratorios, CA, USA) according to the manufacturer's instructions. Fresh human plasma batches (1 ml in 9 ml PBS or 10 ml) were incubated respectively overnight with 3 ml immobilized IgG fraction at 4"C with gentle rotation. Bound SAA4-containing particles were eluted after extensive 4"C PBS washing with 7 M Urea, 20 mM Tris/HCI, 150 mM NaCI, 1 mM EDTA, pH 8.4 and analyzed by SDS-PAGE using a 5% to 20% aerylamide gradient gel. In one experiment 40023685
Page 82: 40023686
7 eluted SAA4-containing particles were dialyzed against 4mM Tris/HCl.. 30raM NaCI, pH 8.4 followed by freeze-drying to one fifth the original volume. These particles were re-applied to the immunoabsorber, eluted and analyzed as above. RESULTS SAA~ Purifica,ti.on In order to eharaeterize SAA4 as an apolipoprotein, we developed a method to purify it from normal HDL by using preparative isoeleetrie focusing (preparative IEF) and subsequent molecular sieve chromatography. We isolated 200 mg of total HDL from 240 rnl of normal human plasma (2). The final yield of SAA4 was 2.9 rag, which constitutes a 25% purification efficiency given the starting concentration of 51/zg/ml. Analytical IEF indicated that SAA4 isoforms have relatively basic pl's of 8.1, 7.9 and 7.3, while apolipoprotein C (ape-C) an.d" apolipoprotein A[I (ape-All) have much more acidic pI values (pH <6:0). Therefore, ape-All and apo-C that will co-chromatograph with SAA4 on molecular sieve chromatography, can be separated dearly from SAA4 in the fractions eluted from the Rotofor apparatus. Figure 1 shows that when the fractions were analyzed by SDS-PAGE, apo-C and apo-AIl appeared in the acidic fractions (fractions number 2-5) (Figure I, top), while SAA4 appeared in the basic fractions (fractions number 13-19) (Figure 1, bottom): The darkening at the bottom of the gel is the result of Ampholines. At this stage SAA4 was contaminated by apo-Al and ampholines that were removed by molecular sieve chromatography. 40023686
Page 83: 40023687
8 Plasma Concentration of SAAa Plasma SAA4 concentrations were measured with an immunoradiometric assay similar to our method reported previously for inflammatory SAAs (I8). Standardization was achieved by creating artificial SAA4-enriched HDL particles as depicted in figure 2, lane 2. (The apparent decrease in apoprotein AI[ results from gel "smiling."). In 26 healthy volunteers the plasma concentration of SAA4 was 55+13/~g/ml. Thus, the concentration wa~ very comparable to that of the apolipoproteins C. Tissue Expression ,of,,S ,AA~t Poly A+ RNA from 15 different human organs and peripheral blood leukocytes were hybridized to the radiolabeled SAA4 eDNA clone CSI. Northern blot analyses showed SAA~ expression c[ufing homeostasis only in liver tissue (Figur.e 3). The size of the SAA4 mRNA was similar to the approximately 700 bases reported previously (6). The Northern blot obtained from the first eight organs was subsequently hybridized to radiolabeled g-actin eDNA to verify the intactness of the RNA, showing the presence of l~-actin in the organs analyzed, as well as ct- actin in heart and muscle (Figure 3). Distribution of.S.AA~. Amont~ Livol~rotein Classes and HDL.S.ubelasses Immunochemical staining of isoelectric-focused LDL and VLDL obtained from normal individuals showed that SAA4 was present also in VLDL but not in LDL (Figure 4). In addition to the major SAA4 isoforms (pI 7.3, 7.9, and 8.1), minor isoforms of pI 6.8, 6.25, and 6.20, confirmed to be SAA4 by amino-~erminal 40023687
Page 84: 40023688
9 amino acid sequencing, are also evident in VLDL. These minor isoforms constitute approximately 5% of the total SAA4 component. The same results were obtained with LDL and VLDL from patients in acute phase (data not shown). SDS-PAGE analyses of HDL particles with a wide range of densities (p=1.063-1.21 g/ml) showed that both the glyeosylated and non-glyeosylated molecules are present in similar ratios and amounts in HDL2, HDL3A, and HDL3B (data not shown). SAA4 is Present on Acute Phase ,HDL SAA4 is a minor apolipoprotei.n component of normal HDL constituting 1% to 2% of the total apolipoproteins of this particle. It was the dominant form of SA/k on normal.HDL, the acute SAAs being virtually undetectable (Figure 5A). In- patients mounting an acute phase response, SAA4 was not displaced by the vastly increased number of inflammatory SAA molecules on HDL; its presence was masked by the dominance of the acute phase SAAs (Figure 5B). We also analyzed the acu~" phase SAAs in patienls undergoing an acute phase response. We observed that both SAA4 and the acute phase SAAs co-localize to the same HDL subpopulations (unpublished observation, Asztalos,B., Roheim,P.S., and De Beer,F.C.). The acute phase SAAs are encoded by two.genes with alleli¢ variation possible at each locus (12,13). Each allele gives rise to two isoforrns, the primary translation product and a post-translational modification, thus accounting for eight possible isoforms (12,13). In the immunoblot presented in 40O23688
Page 85: 40023689
10 Figure 5C, the acute SAA isoforms on the HDL of three patients were identified with a monoelonal anti-hmnan AA antibody specific for acute phase SAA1 and SAA2. These patients had respective acute phase SAA concentrations of 167, 579, and 290/~g/ml. All three patients were homozygous at the SAA1 gene locus. The protein product of this gene (SAAI) was represented by the primary translation product (pl 6.4) and its post-translational modification (pl 6.0). Patients #1 and #3 were also homozygous at the SAA2 gene locus. The protein encoded by this allele, SAA2ct, was represented by the primary translation product (pl 7.5) and its post-translational modification (pl 7.0). Patient #2, however, was heterozygous at the SAA2 gen~ locus. Here SAA2ot and SAA2B were represented by the respective primary translation products (pl 7.5 and 8.0) and their post-translational modifications (pl 7.0 and 7.4). Figure 5B is a- Coomassie stain of the HDL from the same three patients represented in Figure 5C, showing the acute SAA isoforms evident in Figure 5C, as well as the SAA4 isoforms evident in F~gure 5A. In patients heterozygous at the SAA2 gene locus (such as patient #2), the presence of the SA/L¢ isoforms are masiced by the oven~xpression of particularly the basic acute phase SAA2~ isoforms. However, in patients who are homozygous at the SAA2 gene locus, the constitutive SAA4 can be readily distinguished from acute SAA isoforms by IEF (patient #1; patient #3). The amount of SAA4 on the HDL particles during the acute phase is unaltered from that on normal HDL; SAA4 is thus not appreciably displaced. 40023689
Page 86: 40023690
11 SAA~. is Present on a Particular Subpopulation of HI)L Particles Ultracentrifugal separation of lipoproteins is not as sensitive a method of lipoprotein fraetionation as two-dimensional eleetrophoresis. During ultracentrifugation, losses and/or incomplete separation of so-me functionally important particles may occur (15,20). We subjected fresh, normal plasma from three individuals to two- dimensional eleetrophoresis using 0.7% agarose in the first dimension to separate particles according to mobilities and non-denaturing polyaerylamide gradient gels in the second dimension to separate particles of different sizes. With the aid of a monospeeific rabbit anti-human SAA4, we determined a remarkably selective distribution of SAA4 to three distinct particles (Figure 6). Two of these particles (1 and 2) had similar sizes, but different charges. Particle 3 was similar in" charge to particle 2, but not in size (Table 1). From the first-dimensional distribution, (top insert, Figure 6), it was apparent that there were two separate a-migrating particles, which represented over 95% of SAA4. Presence of pre-13- migrating particles in the first dimension was indicated. This went undetected on two-dimensional electrophoresis. This probably represents VLDL-associated SAA4. It should be noted that the Rf's of the SAA4-earrying particles were very similar in the three individuals studied, but the percent distribution of SAA4 between individuals varied somewhat (Table 1). 40023690
Page 87: 40023691
12 Characterization of the Apolipoprotein Component of SAArI= Containing Particles Solid phase immunoabsorption using mono-specific anti-SAA4 antibodies revealed that SAA4 was present on particles that had apo-Al, ape-All and apo-C present in ratios indistinguishable from those of normal HDL3 prepared by ultracentrifugation (data not shown). When small volumes of plasma (1 ml) were incubated with the solid phase, the composition of the eluted particles remained unaltered when compared to offering 10 ml of plasma. When eluted particles were re-constituted in a physiological buffer and re-applied to the immunoabsorber, elution and analysis indicated that the apolipoprotein ratios were unaffected from the original (data not shown). This indicated it was unlikely that SAA4-enriehed particles existed as reported for inflammatory SAAs- during the acute phase (2). DIS,,CUSSIQN Until now, the view was prevalent that SAA molecules were present on normal HDL in very insignificant amounts and that these same molecules increased dramatically during inflammatory events to become even the major apolipoproteins on acute phase HDL. Our finding that th6re are two distinct groups of SAA links the function of the SAA family more intimately to that of HDL. Minor apolipoproteins can have major biological roles (3). The potential for SAA4 to play a similar role in normal HDL function merits consideration. 40023691
Page 88: 40023692
13 Presumably the acute phase SAAs have a different but related function that equips the HDL particle for host defense during i.nflammatory states. A number of studies have suggested that the function of SAA is linked to the inflammatory process and that HDL is-merely a carrier for this moleci~le (21-23). Teleologic considerations, however, suggest that SAA is involved in I-IDL metabolism per se'. This has been supported by studies showing altered HDL binding to eel|s when SAA is present on the particles (24,25) and the significant influence that SAA has on lecithin cholesterol aeyltransferase (LCAT) (26). The discovery of the constitutive SAA_ group, that comprised more than 90% of SAA on normal HDL (6,9), linked the function of the SAA family more directly to that of normal HDL. The distribution of the constitutive SAA molecules, that were restricted to HDL subclasses and VLDL, was similar to that of apo-C (27). and provides support for this contention. The concentration of human constitutive SAA4 was comparable to that of apo-C (27). The basic isoeleetrie points of the constitutive SAA4 isoforms probably contributed to the delay in recognizing the existence of this group .on normal I-IDL. Only a single gene has be~n identified (28), and the isoforms are probably the result of differential glyeosylation because all these, isoforms were identical through nine cycles of amino-terminal sequencing (data not shown). The constitutive SAAs of human and mouse have been found to be structurally similar to each other, but distinct from the inflammatory SAA group (6,9). All inflammatory SAAs in all species studied were conserved between amino acids 33 40023692
Page 89: 40023693
and 44 (29). The constitutive SAA molecules are unique in having substitutions in this region, as well as the characteristic additional octapeptide insert (6,9). This suggests a distinct function for this group. The induction of the constitutive SAA group also differed from that of the inflammatory group (6,9). Human SAA4 was not induced by cytokines, whereas, constitutive mouse SAA5 was only modestly induced (8). However, it has recently been shown that the presence of mouse SAA5 in HDL during the peak of the ~eute phase response was prevented either by translational interference or by displacement from the particle and rapid clearance (9). This suggests that mechanisms operate to ensure the domination of either of the SAA groups on HDL, but not both at the same time (9). Of interest was the relatively high tyrosine content of the constitutive SAAs. The inserts, for instance, had a double tyrosine motif (6,9). It merits consideration whether this might have functional implications given the recent data showing oxidative tyrosylation of I-IDL by peroxidase enhanced cholesterol removal from cultured fibmblasts and macrophages (30). ,~, The advent of two-dimensional separation of lipoproteins from fresh plasma, on the basis of charge in the first dimension and size in the second dimension (20), has allowed for a much greater definition of the polydispersity of HDL particles (15). Thus, apo-Al-eontaining particles were divided into 12 distinct groups (15). It is remarkable that SAA4 was associated with only three discreet, closely related particles. These particles were distinct from the particles that were shown 40023693
Page 90: 40023694
15 to be involved in the initial acceptance of cholesterol from cells (20). Thus it is unlikely that constitutive SAA4 was involved in this process. Additionally, our immunoabsorption data indicated that the ratios of apolipoproteins on SAA4- carrying particles did not differ from the ratios obtained when a total HDL3 population of particles was prepared from this plasma. Given that phospholipids have recently been identified as important factors in imparting charge to HDL particles (31), it makes it likely that these particles carry a distinct phospholipid component different from other HDL particles. This could be of considerable interest given that phospholipid transfer between lipoprotein particles remains ill- defined even though of obvious importance (32). We propose that the function of the SAA family is linked to that of HDL. Studies indicating that inflammatory SAA-bearing HDL increased binding to cells (24,25)" raises the question whether lipid flow between cells and HDL is altered by the presence of inflammatory SAA on these particles. Constitutive SAA4, on the other hand, merits consideration as a factor that might be involved in lipid transfer between lipoprotein classes. "-" 40023694
Page 91: 40023695
16 ACKNOWI,EDGEMENTS: We wish to thank Ms. Susan Allen for excellent editorial assistance during the preparation of this manuscript, as well as Dr. C. Banka, Scripps Research Institute, La Jolla, CA for the monoclonal anti-human SAA4- antibody, and Dr. Mordechai Pros, Sackler Facility of Medicine, Tel Aviv University, Israel, for the monoelonal anti-human AA antibody. This work was supported in part by Grant 3375 from the Council for Tobacco Research, Veterans Administration Medical Research Funds, and United States Public Health Service Grants AR 40379 (F.C. de B.) and HL 25596 (PHR). 40023695
Page 92: 40023696
REFERENCES !.. Karathanasis, S.K. 17 1992. Lipoprotein Metabolism: High Density Lipoproteins. In Molecular Genetics of Coronary Artery Disease (A.J. Lusis, J.l. Rotter, and R.S. Sparkes, editor) Karger, Bas~l. pp. 140-171. 2. Coetzee, G.A., A. F. Strachan, D. R. van der Westhuyzen, H. C. Hoppe, M. S. Jeenah, and F.C. de Beer. 1986. Serum amyloid A-containing human high density lipoprotein 3: Density, size and apolipoprotein composition. J. Biol. Chem. 261:9644-9651. 3.Brunzell,J.D. 1989. Familial lipoprotein lipase deficiency and other causes of the ehylomicronemia syndrome.In The Metabolic.Basis of Inherited Disease. C.R. Scriver,A.L. Baudet, W. S. Sly, and D. Valle. editors. McGraw-Hill,New York, 1165-1180. 4. Eriksen, N., and E.P. Benditt. 1980. Isolation and characterization of the arnyloid-related apoprotein (SAA) from human high density lipoprotein. Proc. Natl. Acad, Sei, USA. 17:6860-6864. 5. Hoffman, J.S., and E.--P. Benditt. 1982. Secretion of serum amyloid protein and assembly of serum amyloid protein-rich high density lil~oprotein in primary mouse hepatoeyte culture. J. Biol. Chem. 257: 10510-10517. 6. Whitehead, A.S., M. C. de Beer, D.M. Steel, M. Ritz, J.M.. Lelias, W.S. Lane, and F.C. de Beer. 1992. Identification of novel members of the serum amyloid A protein supeffamily as constitutive apolipoproteins of high density lipoproteins. J. Biol. Chem. 267: 3862-3867. 40023696
Page 93: 40023697
18 7. McAdam, K.P.W.J., and J.D. Sipe. 1976. Murine model for human secondary amyloidosis: Genetic variability of the acute phase serum protein SAA response to endotoxins and casein. J. Exp. Med. 144: 1121-1127. 8. Strachan, A.F., F.C. de Beer, G. A. Coetzee, H.C. Hoppe, M.S. Jeenah, and D, R. van der Westhuyzen. 1986. Characteristics of apo-SAA containing HDL3 in humans. Protides Biol. Fluids. 34:359-362. 9. De Beer, M.C., M,S, Kindy, W.S. Lane, and F.C. de Beer. 1994. Mouse serum amyloid A protein (SAA5): Structure and expression. J. Biol. Chem. 269:4661-4667. 10. Steel, D. M., G.C. Sellar, C.M. Uhlar, S. Simon, F.C. de Beer, and A.S. Whitehead. 1993. A constitutively expressed serum amyloid A protein gene (SAA4) is closely linked to, and shares structural similarities with an acute-phase.. serum amyloid A protein gene (SAA.2). Genomics I6:44-7-454. 11. Kluve-Beekerrnan, B., S.L. Nayior, A. Marshall, J.C. Gardner, T.B. Shaws, and M.D. Benson. 1986. Localization of human SAA gene(s) to chromosome 11 and dntecfion of DNA polymorphisms. Biochem. Biophys. Res. Commun. 137:1196-1204. 12. Straehan, A.F., W.F. Brandt, P. Woo, D.R. van der Wosthuyzen, G.A. Coetzee, M.C. de Beer, E.G. Shephard, and F.C. de Beer. 1989. Human serum amyloid A protein: The assignment of the six major isoforrns to three published gene sequences and evidence for two genetic loci. J. Biol. Chem. 264: 18368- 18373. 40023697
Page 94: 40023698
19 13. Beach, C.M., M.C. de Beer, J.D. Sipe, L.D. Loose, and F.C. de Beer. 1992. I-[uman serum amyloid A protein: Complete amino acid sequence of a ne~v va.riant. Bioehem. J. 282:615-620. 14. Kluve-Beckerman, B., M.L. Drumm, and M,D. Benson. 1991. N6n- ex.pression of the human serum amyloid A three (SAA3) gene. DNA and Cell Biol. 10:651-661. 15. C~stro, G.R., and C.J. Fielding. 1988. Eady incorporation of cell-derived cholesterol into pre-B-migrating high-density iipoprotein. Biochemistry. 27:25- 29. 16. Strachan, A. F., F.C. de Beer, D.R. van der Westhuyzen, and G.A. Coetzee. 1988. Identification of three isoform patterns of human serum amyloid A protein. Bioehem. J. 250:203-207. 17. Strachan, A. F., E.G. Shephard, D.U. Bellstedt, G.A. Coetzee, D.R. van der Westttuyzen, and F.C. de Beer. 1989. Human serum amyloid A protein: behavior in aqueous and urea-containing solutions and antibody production. Biochem. J. 263:365-3"70. 18. Godenir, N.L., M.S. Jeenah, G.A. Co~tzee, D.R. van der Westhuyzen, A.F. Straehan, and F.C. de Beer. 1985. Standardization of the quantitation of serum amyloid A protein (SAA) in human serum. J. lmmunol. Methods. 83:217-225. 19. Foleh, J., M. Lees, and G.H. Sloane Stanley. 1957. A simple method for the isolation and purification of total lipids from animal tissues. J. Biol. Chem. 226:497-509. 40023698
Page 95: 40023699
20 20. Asztalos, B.F., C.H. Sloop, L. Wong, and P.S. Roheim. 1993. Two- dimensional electrophoresis of plasma lipoproteins: Recognition of new apo-Al containing subpopulations. Bioehim. Biophys. Aeta. 1169:291-300. 21. Aldo-Benson, M.A., and M.D. Benson. 1982. SAA suppression of immune response in-vitro: Evidence for an effect on T-e.ell maerophage interaction. J. lmmunol. 128:2390-2392. 22. Linke, R. P., V. Book, G. Valet, and G. Rothe. 1991. Inhibition of the oxidative burst response of N-Formyl peptide-stimulated neutrophils by SAA. Biochem. Biophys. Res. Commun. 176:1100-1105. 23. Zimliehman, S., A. Danon, 1. Nathan, G. Mozes, and R. Shainkin- Kestenbaum. 1990. Serum amyloid A, an acute phase protein, inhibits platelet activation. J. Lab. Ciin. Med. 116:180-186. 24. Shephard, E.G., F.C. de Beer, M.C. de Beer, M.S. Jeenah, G.A. Coetzee, and D.R. van der Westhuyzen. 1987. Neutrophil association and degradation of normal and acute-phase high-density lipoprotein 3. Bioehem. J. 248:919-926. 25. Kisilevsky, R., and L. Subrahmanyan. 1992. Serum amyloid A changes high density lipoprotein's cellular affinity. Lab. Invest. 66:778-785. 26. Steinmetz, A., G. Hocke, R. Saile, P. Puehois, and J.C. Fruehart. 1989. Influence of serum amyloid A on cholesterol esterification in.human plasma. Bioehim. Biophys. Acta 1006:173-178. 4OO23699
Page 96: 40023700
21 27. Ohta, T., S. Hattori, S. Nishiyama, and I. Mmsuda. 1988. Studies on the lipid and apolipoprotein compositions of two species of apo A-l-containing lipoproteins in normolipedemie males and females. J. Lipid Res. 29:721-728. 28. Sellar, G.C., S.A. Jordan, W.A. Biekrnore, J.A. Fantes, V. van Heyningen, and A.S. Whitehead. 1994. The human serum amyloid A protein (SAA) superfamily gene duster: Mapping to chromosome 1 Ip15.1 by physical and genetic linkage analysis. Genomies 19:1-6. 29. Watson, G., S. Coade, and P. Woo. 1992. Analysis of the genomie and derived protein structure of anovel human serum amyloid A gene, SAA4. Scand. J. [rnmunol. 36:703-712. 30. Francis, G.A., A.J. Mendez, E.L. Bierman, and J.W. Heine.eke. 1993. Oxidative tyrosylation of high density lipoprotein by peroxidase enhances cholesterol removal from cultured fibroblasts and macrophage foam cells. Proc. Natl. Acad. Sci. USA. 90:6631-6635. 31. Davidson, W.S., D.L. Sparks, S. Lund-Katz, and M.C. Phillips. 1993. Molecular basis for the difference in charge between pre-beta- and alp.ha-high density lipoprotein. Circulation 88: Suppl 2:1437. 32. Tall, A.R. 1990. Plasma high density lipoproteins: Metabolism and relationship to atherogenesis. J. Clin. Invest. 86: 379-384. 40023700
Page 97: 40023701
LEGENDS TO FIGURES Figure [ (Top an,,d Bottom): SDS-PAGE ,analysis .of apolipoproteins separated bY, preparative isoelectric focusing Normal human HDL (200 rag) was subjected to preparative isoelectric focusing as described in "Materials and Methods" and aliquots of each of the eluted fractions were analyzed in 5% to 20% aerylamide gradient reduced SDS gels. The Coomassie-stained gels depict the separation achieved for apo-Al, apo-AIl, and SAA4. (Apo-C is obscured by ampholines). SAA4 eluting in fractions 13-19 show apo-Al contamination. S-normal human HDL (10 ~ug) loaded as a standard. Figure 2: ~-,,enriched.. HD.L. .... parti, gle used in immunoradiomet.ric SAA4_ This Coomassie stain of a 5% to 20% aerylamide gradient reduced SDS-gel depicts the SAA4-enriehed particle used in immunoradiometrie SAA4 assays (lane 2). This particle was. generated by incubating purified SAA4 with normal HDL as described in "Materials and Methods". The quantity of SAA4 on this particle was determined by pyfidine extraction of the dye from Coomassie- stained gels (18). Lane 1, normal human HDL and lane 3, acute phase human HDL. Each lane contains 5/~g of protein. 40023701
Page 98: 40023702
Figure 3: Northern blot analysis of hu, man tissues and peripheral blood ieukocvtes A radiolabeled SAA4. eDNA was hybridized to blots containing p01y A+ RNA from 15 human tissues and peripheral blood leukoeytes, showing that SAA4 mRNA is produced only in haman liver. Radiolabeled human B-aetin eDNA was also hybridized to the first blot to verify the intactness of the RNA, showing the presence of gaetin RNA in all tissues, as well. as ct-actin in heart and skeletal muscle. Fieure 4: Immungblot of ,.electrofoeus, ed.. lipoproteins Lipoproteins were subjected to isoelectrie focusing and pressure blotting. The major SAA4 isoforms (pl 7.3, 7.9, and 8.1) as well as the minor isoforms (pl 6.20, 6.25, and 6.8) were identified with a rabbit anti-human SAA4 antibody. Lane 1, 50/zg normal HDL3A; lanes 2 and 3, 50 #g normal LDL; and ~'ane 4, 50 pg normal VLDL. (Microgram quantities refer to prolein). This immunoblot shows the absence of SAA4 from LDL. 40023702
Page 99: 40023703
Fij~ure 5: Presence of SAA4._ on acute phase ,,HDL Figure 5A is a Coomassie stain of 400 ~ug eleetrofoeused HDL from a healthy individual showing the presence of SAA4 molecules with pI's- 8.1, 7.9 and 7.3. The basic ampholine gradient chosen does not allow for separation of ihe more acidic apo-AI, aim-All and C apoproteins, which focus at the very acidic end of the gel. Figure 5B is a Coomassie stain of 200/tg eleetrofocused HDL from three patients with varying concentrations of acute phase SAAs (167/~g/ml patient I; 579 /~g/ml patient 2; 290/~glrnl patient 3). Patient 2 is heterozygous at the SAA2 gene locus. The resultant basle isoforms obscure SAA4. Patients 1 and 3 are homozygous at the SAA2 gene locus, and the SAA4 isoforrns are clearly visible as quantitatively significant proteins. Figure 5C represents an immunochemieal stain of the HDL from the same three patients as in 5B, verifying the acute SAA isoforn~ disposition of these three patients. Only 100/tg HDL was focused, and the acute SAA isoforms were specifically identified with a monoelonal anti-human AA antibody (1:1000 dilution). This confirms SAA4 molecules are present on acute phase HDL. Their presence is masked by the preponderance of acute SAA, particularly in individuals heterozygous at the SAA2 gene locus, and thus also in pools of acute phase HDL. 40023703
Page 100: 40023704
Figure 6: Two-dimensional ,separation of SAA4 subpopulations Plasma was electrophorvsed in the first dimension (0.7% agarose), followed by the application of the agaros~ strip to th~ top of the non-denaturing concave gradient polyacrylamide gel (3% - 369'0) and subse, quenfly electrophoresed. This figure shows the SAA4 distribution in the second dimension. Separated SAA4 subpopulations are labeled 1, 2, and 3. The star indicates the position of human albumin obtained subsequent to the SAA4 immunolocalization. The horizontal insert on the top represents SAA4 distribution on a duplicate ag~ose stdp. On the left side of the panel, the internal standards are: (1) thyrogiobulin (17nm); (2) fcrdtin (12.2nm); (3) eatalase (9.5Into); (4) lactate dchydrogenase (8.16nm); (5) albumin (7.1nm); (6) ovalbumin (4.66nm). Non-specific reactions of- antibody to gamma globulin also appear. 40023704
Page 101: 40023705
Table 1. Two-Dimension~l Coordinates and Percent Distribution of SA~4 1 0.80 ± 0.005 8.76 ± 0.18 b'1.3 _ 15.9 9. 0.95 ± 0.020 8.65 ± 0.13 33.0 4- 19.0 3 0.97 .',- 0.020 8.01 4-'0.10 13.6 ± 2.0 Values represent the mean + S.D. or three sul)jccts dctcrmln¢ct thr~e times. S.D. = standard deviation Plasma samples were subjected to two-dimensional electrophoresis as described in "M=~terials and Methods." Relative Rf to albumin was obtained after immunolocalization with anti-albumin. The size of the particles were determined by constructing vertical reotangles around the Internal standards and the HDL particles (20) and quantitating the incorporated redioactivity as set out in "Materials and Methods.*~Medal diameters were calculated using computer-g~~erated internal curves {20). Using the coordinated Rf and size, each area was delineated and the pixel volume and % of total immunoradioactivity calculated (20). 40023705
Page 102: 40023706
Fig. 1~ Top $ 1 2 3 4 5 6 7 8 9 Fig. 1 Bottom APO-AI SAA 4¢ 10 1"1 12 13 14 1~ 1~ 1~ 18 19 40023708
Page 103: 40023707
• ! I_ heart brain Dlacenta !lver .~keietal muscle .k.'ldnby pancreas 'ispleen ihymus :,'. ;prostate • ."' • ~estis ovary small intestine colon peripheral blood leukocyte //'1 I" I o 0 Z 0 rn
Page 104: 40023708
• !
Page 105: 40023709
---
Page 106: 40023710
Fig. ~ ~-~ 17.0 - 12.2- 9.51 - 8.16- 7.10- 3 40023710
Page 107: 40023711
' ~ Biochcm. ~. • ~-~,.. • Submitted (rapid communication) SERUM AMYLOID A PROTEIN ENHANCES THE ACTIVI'W OF SECRETORY NON-PANCREATIC PHOSPHOUPASE A= Waldemar Pruzansld+, Fredericl~ C. de Bee#', Mada C. de Beer" Eva Stefanski+, and Peter Vadas+. Inflammation Research Group, University of Toronto, Canada+ and Department of Medicine, University of Kentucky College of Medicine, Lexington, KentuckyA Short Title: Index words: Serum amyloid A enhances activity of phospholipase A2 Acute phase, SAA, HDL, phospholipase A~ atherogenesis Supported by grant-in-aid from "The Medical Research Council of Canada, (W.P. ~'nd P.M.), U.S. Public Health Services grant AR 40379 (F.C.D) and by the Council for q'obacco Research Grant 3375 (F.C.D) Correspondin.q Author: W. Pruzanski, M.D., The Wellesley Hospital, 160 Wellesley Street East, Jones Building Room 104, Toronto, Ontario, Canada, M4Y 1J3, TEL: 416-926-7785, FAX: 416-966-5046 40023711
Page 108: 40023712
2 SYNOPSIS The acute phase proteins serum amyloid A protein (SAA) and secretory phospholipasa A= (sPLA~ are simultaneously expressed during inflammatory conditions. SAA associates with HDL altering its physical-chemical composition. We found that purified acute phase SAA, but not constitutive SAA, markedly enhances the lipolytic activity of sPLA= in a dose-related manner using PC:lyso PC or PE:lyso PE liposomal substrates..Normal HDL was found to reduce activity of sPLAz in a dose-dependent manner. When acute phase HDL that contained 27% SAA was tested, it enhanced sPi_A= activity. Immunopudlied monospecific antibodies against SAA completely abolished the enhancing activity of SAA and of acdte phase HDL Given the central role of HDL in lipoprotein metabolism, the interaction between HDL, SAA and sPI.A= may account f~r changes detected in lipoprotein metabolism during the acute phase. 40023712
Page 109: 40023713
3 INTRODUCTION The serum amyloid A proteins (SAAs) compr'~se a family of molecules that are apolipoproteins of HDL (1,2,3,4). They can be divided into two groups. The first group is the well-characterized acute phase SAAs that increase dramatically during inflammatory events (1-3). They displace apolipoprotein AI (apo AI) with resultant remodelling of the HDL particle yielding larger particles with a higher hydrated density, and even becoming the major apolipoprotein component of acute phase HDL (1-¢). The second group comprises the recently discovered mouse SA~ (5) and human SAA4 (6) group that are constitutively expressed as minor apolipoproteins associated with HDL during homeostasis. They are structurally and probably functionally distinct from the acute phase SAA group. The function of acute phase SAA on HDL is still unknown, but it seems likely that this apolipoprotein equips the HDL particle for an altered role to promote host defense. In defining the function of acute phase SAA, one needs to take cognizance of the fact that as SAA increases in the circulation, there is concomitant cytoklne driven synthesis and extraceilular release of non-pancreatic (group II) secretory phospholipase Az (sPLAz) (7,8). Both SAA and sPI.Az are rapidly induced during inflammatory events (7,8,9,10) with mRNA for both detectable at 2 hours and the resultant proteins similarly produced with respect to time (5,11). Whereas acute phase SAAs are practically only produced in the liver (5,6), sPLAz is produced by a variety of mesenchymal cells that would be present at sites of inflammation (7,8). The concentration of sPLA~ can increase hundreds of fold in inflammatory fluids (7,8,12) and in the circulation (13,14). 40023713
Page 110: 40023714
4 The influence of phosphoI~pase act;vily on HDL is important in the understanding of the function of th~s parole (15,16}. Dudng homeostasis, the influence of phospholipasa activity on normal HDL has been shown with respect to hepatic lipase (15016,17). Hydrolysis of surface phospholipid molecules was associated ~h redistribution of cholesterol from the core to the surface of the HDL pafdcle and increased polar group segmental motion of surface pflospholipids (15). 3"he result is a shift in the equilibrium of free cholesterol between HDL and the plasma meml~'ane with an increased net delivery of free cholesterol to the cell by a surface transfer process (15). Given, that the phospholipase activity of hepatic lipase during homeostasis would be f'med on the surface of hepatocytes, the net result was postulated to be cholesterol delivery to the liver for excretion, an important part of reverse cholesterol transport (15,16,17). During the acute phase the situation is ~astically changed. SAA now dominates on the HOL particle and a new phospholipase activity, that of sPLA~ is highly induced "by cytokines at inflammatory sites. The interaction between SAA-bearing acute phase HDL and sPLA~ is thus important to define altered HDL function during inflammation. Herein we report that SAA can promote the activity of sPLA~. Although a number of agents have been described that can block the activity of this enzyme (7',8), SAA is the first protein identified that can promote enzymatic action of sPLA~. These data hold the potential to advance our u.nderstanding of the major changes in lipoprotein metabolism that occur during inflammation. 40023714
Page 111: 40023715
5 MATERIALS AND METHODS 1,2-Dipalmitoyl-phosphatidylchoIine (dipalmitoyl PC) was obtained from Avanti Polar Lipids (Birmingham, AL). Phosphatidylchotine L-¢=-dipalmitoyl [2-palmitoyl-l-14C] (55.5 mCi/mmol) and oleic acid [1-1~C] (40 - 60 mCi/mmol) were purchased from DuPont NEN Prodgcts. L-3-phosphatidylethanolamine l-palmitoyl [2-1~C linoleoyl] (50 -'60 mCi/mmol) was obtained from Amersham, Canada. Recombinant human PLA= (rh-Pl_A~) was the generous gift of Dr. Jeffrey Browning, Biogen Corporation (Cambridge, MA). Rh-PLA2 was produced from a stable transfected Chinese hamster ovary cell line and purified. Bio-Rad protein assay was purchased from Bio-Rad (Richmond, CA). lysophosphatidylcholine palmitoyl, L-~x-phosphatidylethanolamine-S-linoleoyI-y-palmitoyl~ L-~,-lysophosphatidylethanolamine palmitoyl, bovine serum albumin and silica gel were purchased from Sigma (St. Louis, MO). All other reagents were analytical grade or better. Preparation of acute phase HDL (APHDL). and normal HDL (NHDL) The respective HDUs were isolated by sequential ultra-centrifugation from the blood~of healthy donors or from patients experiencing an acute phase response after surgery. HDLwas further subfractioned according to plasma density (p); HD~ (p=1.063- 1.13 g/ml), HDL~ (p=1.13-1.55 g/ml), HDL3a (p=1.155-1.18 gird) by recentrifugation of the total HDL fraction into a linear KBr gradient as described previously (18). Only HDL~ fractions were used in these studies. Protein concentrations were determined and individual apolipoprotein content established by pyridine e~'action of Coomassie Blue stained bands from SDS-PAGE as described (4). Normal HDLa, contained 71% ape A 40023715
Page 112: 40023716
6 I and 29% apo AlL Acute phase HDL~ contained 56% Apo I, 17% Apo All and 27% acute phase SAA. Purification of Constitutive SAA, (¢SAA) SA/h was isolated from normal human HDL (6) by Rotofor (Biorad, Richmond, CA, USA) preparative isoelectric focusing as per the manufacturer's instructions. Two hundrad mg of normal HDL was delipideted at minus 2(PC with ethanol:ether (3:2, v/v) and the protein pellets suspended in 15 ml 8 M urea, 1% (w/v) decyl sodium sulphate (Eastman Kodak Co., Rochester, NY) and 5% (v/v) 2-mercaptoethanol. "l'he Rotofor running buffer consisted of 8 M urea, 1.2% Ampholines pH4-6.5, 1.2% Ampholines pHT-9 and 0.6% Arnpholines pH 7-9 and 0.6% Ampholines pH 3-10. 3"he anionic and cationic chambers contained 0.1 M phosphoric acid and 0.1 M sodium hydroxide respectively. The sample was electrofocused at constant power (12W) until equilibrium was reached. Twenty fractions were harvested and analyzed by SDS-PAGE. The SA/h-containing fractions were dialyzed against 15 mM NaCI, 2 mM Tris/HCI pH 8.4 and lyophilized. The lyophilized pellet was suspended in 2 ml 7 M urea, 150 mM NaCI, 20 mM Tris/HCl pH 8.4 and subjected to molecular sieve chromatography on a 1 X 120 cm Sephacryl $200 column as per the manufacturer's instructions (Pharmai:ia LKB Biotechnology, Piscataway, N J, USA). PreDaratl0n of acute phase SAA (aSAA) and antibodies 1~0 Acute phase HDL~ was delipidated and the SAA separated by molecular sieve chromatography as described previously (6,18). Monospecific rabbit anti-human antibodies were raised (18). lmmunopuritied antibodies were prepared by coupling 40023716
Page 113: 40023717
7 purified SAP, to cyanogen bromide activated Sepharose 4B (Pharmacia LKB) according to manufacturers instructions. Antiserum was passed over the column, washed in PBS, uncoupled with 0.2 M glycine HC! and dialyzed into PBS. ,Llposome Assays Aliquots of dipalmitoyl PC, [14C] dipalmitoyl PC, with or without lysoPC (or phosphatidylethanolamine (PE) with or without LysoPE) were prepared in chloroform and evaporated to dryness. Multilamellar liposomes were prepared by dispersing the resulting lipid mixture in 100 mM Tris HCI buffer, pH 8.0, followed by heating for 2 min at 41°C and vortexing for 2 rain before use. Only freshly prepared liposomes were used~ Assays were carded out in a total volume of 0.2 mi of 100 mM Tris HCl, pH 8.0 containing 2.5 or 10 mM CaCI~, 0.1% bovine serum albumin and 5 - 30 nmoles of PC vesicles (containing 2-16 nCi of [14C] dipalmitoyl PC per assay). The optimal concentration was found to be 20 nmoles/assay and this was used in all experiments. If acute phase SAA's or other agents were included, they were preincubated with 20 nmoles of liposomes for 1 hr at 41°C before the assay. The reaction was then started by addition of 20 pl of rh sPLA= stock solution with final sPLA~ concentrations ranging between 10 - 200 rig/200/~1 assay volume unless otherwise stated. The reaction mixture was incubated for 30 rain at 41=0. The reaction was stopped by the addition of 1.32 ml isopropanol/heptane/0.5. M H~,S04 40:10:1 (v/v/v). The mixture was heated for 1 rain at 60~0 before addition of 0.66 ml H=O and 0.8 ml heptane. The two phases were allowed to separate and after centrifugation for 10 min at 1500 rpm, 0.8 ml of the upper phase was added to 1.0 ml heptane containing 100 mg sitica gel. The mixture was spun again for 10 rain at 1500 40023717
Page 114: 40023718
8 rpm and 1.0 ml of the supematant was used for scintillation counting of 14C-labeIted free palmitic acid. All assays were done in triplicate (19). Inhibition Studios Twenty #1 of SAA (concentration 250 t~g/ml) were mixed with 30 #1 of appropriately diluted immunopurified monospecific rabbit anti-human SAA antibody and the mixture incubated at room temperature for 60 min. Final concentrations of antibody were 25, 50 and 300/~g/ml. Fifty #1 of 0.1 M TrisHCI buffer, pH 8.0, 20 ~1 of 100 mM Ca=+, 20 #1 of BSA (concentration 10 mg/ml) and 40/~g of substrate were added and the final mixture was further incubated for 30 min at 41°C. The results were expressed in nmol/30 rain PLA~. The controls included antibody alone, ~. microglobulin 25/~g/ml alone and mi~oglobulin/anti P~-microglobulin complex. Antihuman ~z-microglobulin ant~ody (mouse IgG1) (done BM-63) was obtained from Sigma (St. Louis, Mo). Working dilution was at least 1:1000 per #g of antigen in an indirect ELISA assay. In assays employing anti-SAA antibodies the antigens were preincubated with appropriately diluted antibodies for 60 min. at room temperature. Uposomal substrate was added and the mixture was further incubated for 60 rain. at 41°C. The sPl_Az was finally added and the-reaction was allowed to proceed for 30 rain. at 41°C.. Statistioal analysis was done using standard tests such as Student t test and correlation coefficients. Results Influence of normal and acute phase HDL on sPLA~ activity Normal HDL (NHDL) was found to inhibit the activity of sPLA= in a dose-dependent 40023718
Page 115: 40023719
9 manner v~th los0 o! 10 g/ml. Using PC:Lyso PC liposomes as a substrate (Fig. 1), inhibitory act;'v~t has already been observed at NHDL concentration of 1 #g/m1, whereas concentration of 100 ~g/ml inhibited sPLA~ a~vity by 80% (p < 0.001). When PE:Lyso PE liposomes were used as a substrate, the inhibition was very similar to that obtained using PC:Lyso PC liposomes (10 /=g/ml - 52%, 50 /~g/ml - 58%, 100/=g/talc 70%). Preincubation with phosphorylcholine (10 aM) did not abolish inhibitory activity of NHDL. Inhibitory activity of NHDL was related to the amount of sPLA~ used in the assays. Keeping a constant NHDL concentration of 10 t~g/ml, the inhibition of hydrolytic activity by 10 ng of sPLA2 was 59%, by 100 ng - 34% and by 200 ng- 11%. Since during the acute phase, newly synthesized and released extracellularly acute phase SAA combines with HDL resulting in the formation of acute phase HDL (APHDL), we compared the effect of physiological NHDL and APHDL on sPLAz activity. The former (15/=g/ml) gave 62% inhibition of sPLAz enzymatic activity, (p < 0.001) whereas APHDL (15 ~g/ml) caused enhancement of 188% (p < 0.001). Influence of purified SAA on sPi.A~ activity Purified inflammatory SAA (aSAA) enhanced the activity of sPL_A= in a dose- dependent manner (Fig. 2). Using PC:LysoPC liposomes (proportion 2:1) as a substrate, 135% enhancement was seen at a aSAA concentration of 3 ~g/ml, whereas a concentration of 10/~g/ml doubled (195%) sPLA= activity (p < 0.001). When PC alone was used as a substrate, the enhancement was less pronounced with 10 #g/mi of aSAA increasing sPI.Az activity to 130%, 20 #g/ml to 151% (p <0.01) and 50 #g/ml to 185% (p < 0.001). Liposomes composed of PE: Lyso PE (proportion 2:1) were also hydrolysed 40023719
Page 116: 40023720
10 but less actively than PC:Lyso PC. Ten #g/ml of aSAA enhanced sPLA= a~vity to 129% and 50 #g/ml to 167% (p < 0.001). Constitutive SAA (cSAA) in the range of 10 #g/ml to 50/~g/ml did not sh3nificantly enhance sPI_A= activity (9-34%) (p > 0.05), in either PC:Lyso PC or PE:Lyso PE system. Inhibition of aSAA - mediated sPI.A, enhancement by anti-SA.A antibody aSAA preincubated with increasing concen~ations of anti-SAA antibody gradually lost its sPI.A~ enhancing activity (Fig. 3). Neither antibody alone nor irrelevant ~ microglobulin / anti ~,, microglobulin complex of similar molecular weight had an impact on sPLA= activity. To support ~he notion that the enhancing activity of APHDL on sPl.~ is caused by inflammatory SAA associated with HDL, APHDL~ (25 /~g/ml) was incubated with monospecific, immunopurified anti-SAA antibody (Fig 4). When no antibody was added, the enhancement of sPLAz activity by APHDL~ was 165% (p < 0.001). Incubation Of APHDL~ with 100/~g/ml of anti~SAA antibody reduced the enhancement of sPLA= activity to 12%; incubation with 200 #g/ml of antibody abolished completely enhancing activity and actually resulted in inhibition of sPLA~ activity by 29%. Antibody per se had no effect on sPLA= activity. 40023720
Page 117: 40023721
11 DISCUSSION During the acute phase the concentration of acute protein SAA can increase up to a thousand-fold in t:ody fluids (1). The presence of SAA on HDL remodels the particle (1) and alters binding characteristics of SAA-bearing HDI_, augmenting its association with macropheges (20) and neutrophils (21) and reducing its affinity for hepatocytes (20). SAA thus holds the potential to drastically alter HDL function. Concomitantly with SAA, sPL,~ activity increases markedly in conditions which induce the acute phase reaction (7,8). Circulating sPI_A~ activity may increase more than 3000-fold in septic shock (13). Extremely high activity of sPLA= was found in malaria (14), multi-organ dysfunction due to salicylate poisoning (22) and in inflammatory processes such as peritonitis (23) or inflammatory arthritis (24). "l'here is paucity of specific information about the possible link between sPLAz activity and lipoprotein meiabolism, even though a substantial body of evidence exists that this group of lipases alter HDL - mediated lipid flow (15,16,17). It has been reported that the hydrolysis of HDL phospholipids by hepatic lipase, an enzyme which possesses PI-Az-like activity, results in the appearance of HDL with enhanced ability to deliver cholesterol to cells (15,16). This process was found to be associated with relocation of cholesterol molecules from the core of HDL to the surface, inducing the shift in equilibrium of free cholesterol between HDL and the plasma membrane (15). It was also reported that stimulation of the release of free fatty acids from PC-labelled human platelets by HDL, is probably mediated through the activation of membrane-associated PLA~ (25). 4OO23721
Page 118: 40023722
12 At inflammatory sites there would be an enormous increase In the fluid phase sPLA= activity in addition to an altered HDL containing SAA (APHDL). Since APHDL enhances lipolytic activity of sPLAz, one envisages a mechanism whereby lipid, and pa~cularly cholesterol, would flow from HDL towards cells at such sites. Such lipid delivery might in the short term be of potential benefit during the acute inflammatory event, providing materials for repair of damaged membranes of cells. Given the central role of HDL in lipoprotein metabolism (26), the interaction of sPLA= with SAA and HDL, could potentially account for some changes detected in lipoprotein metabolism dudng the acute phase. This is of particular relevance for patients with chronic inllammatory diseases such as rheumatoid arthritis who have prolonged periods of high sPLA2 activity and SAA induction (27,28). In these patients a high incidence of atherosclerosis and high mortality from cardiovascular diseases have been reported (29). Our findings may additionally be important for events leading to the development of atherosclerotic lesions in the vessel wall. The presence of acute phase SAA molecules in atherosclerotic lesions in mouse and man have been reported (30,31).~ In situ hybridization studies indicated production of SAA in endothelial cells, macrophage-dedved 'loam" cells and vascular smooth muscle cells in lesions (31), presumably induced by oxidized lipids present in the milieu (32). When one considers that vascular smooth muscles cells are most probably a major source of sPLA= synthesLs (|1), the enhancement of sPLA~ activity by SAA at such sites, could lead to aberrant lipid delivery and conceivably promote ffoam" cell formation, a key event in atherogenesis. 40023722
Page 119: 40023723
13 The finding that conslJtutiva SA~ had no significant effect on sPLAz actMty, suggests that the very c~nserved region of acute phase SAA between am|no acids 35 and 48 could be the active site for promotion of this activity as the constitutive SAPs are the only members of SAA family with differences in this region (6}. The m~chanism by which SAA promotes sPI-Az activity is unknown, sPLAz hydrolyses phospholipids only when they are in a specific spatial orientation. Unperturbed cell membrane phospholipids for instance, resist this activity. It is conceivable that the amphiphilic SAP,, when associated with cell membranes, alters their fluidity, making phospholipids accessible to sPLAz - induced hydrolysis. Another possibility is that the SAA region between amino" acids 48 and 51, that is homologous to the Gly~ - X - Gly=z - Gly= phospholipid and calcium binding site of PLAz molecules, is involved. This possibility is lessened by the fact that constitutive SAA (6) has an identical sequence in the above region but does not activate sPLA=. Our data indicate the existence of a potential mechanism based on SANsPI_A~ interaction, that may significantly alter lipid flow dudng Inflammation. 40023723
Page 120: 40023724
14 LEGENDS 1"O FIGURES Inhibition of Iipolytic activity of phospholipase/~ by NHDI_ Enhancement of lipolytic actMty of phospholipase A= by serum amyloid A (SAA) protein. Closed circles - PC : Lyso PC substrate, triangles - PE : Lyso PE substrate. Inhibition of SAA enhancing activity by anti-SAA antibody (Ab). Inhibition of APHDL enhancing activity by anti-SPA antibody. 4OO23724
Page 121: 40023725
o o o ° 15 REFERENCES Coetzee GA, Strachan AF, van der Westhuyzen DR, Hoppe HC, Jeenah MS, De Beer FC: Serum amyloid A-containing human high density lipoprotein 3: density, size, and apolipoprotein composition. J. Biol. Chem. 261:9644-9651, 1986. Cabana VG, Siegel JN, Sabesin SM: Effects of the acute phase response on the concentration and density distribution of plasma lipids and apolipoproteins. J. Lipid Res° 30:39-49, 1989. Malle E, Steinmetz A, Raynes JG: Serum amyloid A (SAA): an acute phase protein and apolipoprotein. Atherosclerosis. 102:131-146, 1993. Godenir NL, Jeenah MS, Coetzee GA, van der Westhuyzen DR, Strachan AF, de Beer FC: Standardization of the quantitation of serum amyloid A protein (SAA) in human serum. J. Immunol. Methods, 83:217-225, 1985. De Beer MC, Kindy MS, Lane WS, De Beer FC: Mouse serum amyloid A protein (SAAs): structure and expression. J. Biol. Chem. 269:4661-4667, 1994. Whitehead AS, De Beer MC, Steel DM, Ritz M, Lelias JM, Lane WS, De Beer FC: Identification of novel members of the serum amyloid A protein (SAA) superfamily as constitutive apolipoproteins of high density lipoprotein. J. Biol. Chem. 267:3862-3867, 1992. Vadas P, Browning J, Edelson J, Pruzanski W: Extracellular phospholipase A= expression and inflammation: the relationship with associated disease states. J. Upid Mediators 8:1-30, 1993. 40023725
Page 122: 40023726
16 8. Pruzanski W, Vadas P, Browning J: Secretory non-pancreatic group II phospholipase A=: role in physiologic and inflammatory processes. J. Lipid Mediators 8:161-167, 1993. 9. Rokita H, Loose LD, Battle LM, Sipe JD: Synergism of interleukin I and intedeukin 6 induces serum amyloid A production while depressing fibrinogen: A quantitative analysis. J. Rheumatol. 21:400-405, 1994. 10. Smith JW, McDonald "13.: Production of serum amyioid A and C-reactive protein by HepG= cells stimulated with combinations of cytokines or monocyte conditioned media: the effects of prednisolone. Clin. Exp. Immunol. 90:293-299, 1992. 11. Nakano T, Ohara O, Teraoka H, Adta H: Group II phospholipase A= mRNA synthesis is stimulated by two distinct mechanisms in rat vascular smooth muscle cells. FEBS Lett. 261:171-174, 1990. 12.. Pruzanski W, Scott K, Smith G, Rajkovic I, Stefanski E, Vadas P: Enzymatic activity and immunoreactivity of extracellular phospholipase A~ in inflammatory synovial fluids. Inflammation, 16:451-459, 1992. 13. Vadas.P, Pruzanski W: Induction of group II phospholipase Az expressiorr~nd the pathogenesis of the sepsis syndrome. Circulatory Shock 39:160-167, 1993. 14. Vadas P, Taylor T, Molyneux M, Stefanski E, Pruzanski W: Serum phospholipase A= and disease severity In children with falciparum malaria. Am. J. Trop. Mad. Hyg. 49:455-459, 1993. 40023726
Page 123: 40023727
17 15. Bamberger M, Lund-Katz S, Phillips MC, Rothblat GH: Mechanism of the hepatic lipase induced accumulation of h~gh-density lipoprotein cholesterol by cells in culture. Biochemistly 24:3693-3701, 1985. 16. Bamberger M, Gtick JW, Rothblat GH: Hepatic lipase stimulates the uptake of high density I[poprotein cholesterol by hepatoma cells. J. Lipid. Res. 24:869-876, 1983. 17. Collet X, Perret BP, Simard G, Vieu C, Douste~Blazy L: Behaviour of phosphollpase-modified HDL toward cultured hepatocytes. I. Enhanced transfers of HDL sterols and apoproteins. Biochim. Biophys. Au--'ta. 1043:301-310, 1990. 18. Strachan AF, Shephard EG, Bells.tedt DU, Coetzee GA, van der Westhuyzen DR, De Beer FC: Human serum amyloid A protein. Behaviour in aqueous and urea- containing solutions and antibody production. Biochem. J. 263:365-370, 1989.. 19. Pruzanski W, de Beer FC, Kennedy B, Stefanski E. Vadas P: Modulation of proinflammatory phospholipase Az by acute phase reactants and microtubular disruptures. In: "Upid Mediators in Health and Disease (LMHD)". ED. U. Zor. pp 33-38. Freund Publishing House Ltd. London, Tel Aviv, 1994. 20. Kisilevsky. R, Subrahmanyan L: Serum amyloid A changes high density lipoprotein's cellular affinity. Lab. Invest. 66:778-785, 1992. 21. Shephard EG, de Beer FC, de Beer MC, Jeenah MS, Coetzee GA, van der Westhuyzen DR: Neutrophil association and degradation of normal and acute- phase high-density lipoprotein 3. Biochem. J. 248:919-926, 1987. 40023727
Page 124: 40023728
23. 25. 26. 7° 28. 18 Vadas P, Schouten B, Stefanski E, Scott K, Pruzanski W: Association o~ hyperphospholipasemia A~ with multisystem organ dysfunction due to s~ticylate intoxication. Crit. Care Med. 21:1087-1091, 1993. Vadas P, Pruzanski W, Stefanski E, Johnson L, Seilhamer J, Mustard R Jr., Bohnen J: Phospho~ipa~e~ A~ in acute bacteriaJ peritonitis in m~n. In: Cell activation and signal initiation: Receptor and phosphol[pase control of inositol phosphate, PAF, and eicosanoid production, pp 311-316, 1989. Pruzanski W, Vadas P, Stefanski E, Urowitz MB: Phospholipase A2 activity in sera and synovial fluids in rheumatoid arthritis and osteoarthritis. J. Rheumatol. 12:211 - 216, 1985. Magret V, Le Barzic R, Nazih H, Sanderson-Nazih F, Fruchart JC, Delbart C: Role of HDL~ in platelet PLA~ activation. In: Nato Advanced Research Workshop Esterases, Lipases and Phospholipases: From Structure to Clinical Significance. Univ. of Bordeaux II, France. pp 92. Sept, 22-24, 1993. Tall AR: Plasma high density lipoproteins. Metabolism and Relationship to Atherogenesis. J. Clin. Invest. 86:379-384, 1990. Pruzanski W, Keystone EC, Bombardier C, Snow K, Stemby B, Vadas P: Phospholipase A= correlates wl~ disease activity in rheumatoid arthritis. J. Rheumat. 15:1351-1355, 1988. De Beer FC, Fagan EA, Hughes GRV, Mallya RK, Lanham JG, Pepys MB: Seru~ amyloid-A protein concentration in inflammatory diseases and its relationship to the incidence of reactive systemic amyloidosis. Lancet I1: 231-237, 1982. 40023728
Page 125: 40023729
19 29. Wolfe F, MitcheIi D, Siblay JT, Fdes JF, B]och DA, Williams CA, Spitz PW, Hager M, K]einheksel SM, Cathey MA: The mortality of rheumatoid arthritis. Arthr. Rheumat. 37:481-494, 1994. 30. Qiao J-H, Cast'ellani LW, Fishbein MD, Lusis A J: Immune-complex-mediated vasculitis increases coronary artery lipid accumulation in autoimmune-prone MRL mice. Arterioscler. Throm. 13:932-943, 1993. 31. Meek RL, Urieli-Shoval S, Benditt EP: Expression of apolipoprotein serum amylold A mRNA in human atherosclerotic lesions and cultured vascular cells: Implications for serum amyloid A function. Proc. Nat. A~ad. Sci. USA 9I:3186-3190, 1994. 32. Liao F, Andalibi A, de Beer FC, Fogelman M, Lusis A J: Genetic control of inflammatory gene induction and NF-kappa B-like transcription factor activation in response to an atherogenic diet in mice. J. Clin. Invest. 91:2572-2579, 1993... 40023729
Page 126: 40023730
Inhibition of sPLA 2 (%) (u!uJ 08/IOLUU) ~ V es~d!loqdsoqcl
Page 127: 40023731
Figure 2. 4.0 ,3.0 2.5 2.0 ~10 0.5 0 • PC:Lyso PE ~• PC:Ly,_.so £1~ . ~_ I I I I I I 10 20 30 40 50 60 gg/ml 40023731
Page 128: 40023732
Figure 3. 200 - 2_-. 150 ~::::....:.~::: :::::::::4::: ~ 100 ::~:.'.':i:::" 50 :~:::~:~!~ 0 SAA 25 Ab 0 :8:::::::::: 25 25 ~g/ml 25 50 ~i~it~:::::::::;: :+.:.::::... f~:~:~:i:! 25 3OO 40023732
Page 129: 40023733
• Figure 4. 0 lO0 p.g/ml 2OO 40023733

Text Control

Highlight Text:

OCR Text Alignment:

Image Control

Image Rotation:

Image Size: