Tobacco Documents Online

~~7y~that seemed to us to be hazardous might '2;Y be perfonned before adequate coid :ns- eration had been given to the potential dangers. Contrary to what was believed y many observers, our concerns per- ~.'3'~! „'.. ''T;,, - tatned to a few very specific types of exariments that could be carried out with the new techniques, not to the tech- niques themselves ` Guidelines have long been available to ~ ~; protect laboratory workers and the gen- eral public against known hazards asso- ciated with the handling of certain chem- icals, radioisotopes, and pathogenic mi- ~`eroorganisms; but because of the new- ki ~; ;ness of recombinant DNA techniques, no guidelines were yet available for this research. My colleagues and I wanted to be sure that these new techniques would ;~not be used, for example, for the con- ° F,struction of streptococci or pneumo- , :;cocci resistant to penicillin, or for the ~ creation of Escherichia coli capable of synthesizing botulinum toxin or diph- theria toxin. We asked that these experi- ments not be done, and also called for deferral of construction of bacterial re- combinants containing tumor virus genes until the implications of such experi- ments could be given further consid- eration. , . , ' - . During the past 2 years, much fiction has been written about "recombinant DNA research." What began as an act of responsibility by scientists, including a number of those involved in the devel- opment of the new techniques, has be- t with the lsvel of anticipated hazard; the containment procedures used for patho- .~ genicbacteria, toxic substances, and ra- dioisotopes attempt to do this However ., ,~,,~ the containment measures used in these :' =~areas address themselves only to known hazards and do not attempt to protect against the unknown. If the same prin- Stanley N. Cohen .;, ciple of protecting only againit known or expected hazards were .follovlled in re- ~;,,. ,j :~combinant DNA research, there would :;_: be no containment whatsoever except .. . . . - H come the breeding ground for a horde of ; for a very few experiments. In this in- publicists-most poorly informed, some ': stance, we are asking not only that there well-meaning, some self-serving. In this . be no evidence of hazard, but that there article I attempt to inject some relevant : be positive evidence that there is no ~ could be carried out with newly devel- facts into the extensive public discussion ; hazard. In developing guideGnes for re- ~~oped techniques for the propagation of of recombinant DNA research ^~ combinant DNA research, we have at- !.' , genes from diverse sources in bacteria (/). Because of the newness and relative simplicity of these techniques (2), we Some Basic Information ~t` Almost 3 years ago, I joined with a group of scientific colleagues in publicly ~ealling attention to possible biohazards of certain kinds of experiments that The author is a molecular geneticist and Professor of Medicine at the Stanford University School of Medicine, Stanford. Califomia 94305. This article is adapted from a statement prepared for a meet- ing of the Committee on Environmental Health of the California Medical Association. 18 November /97b. - t., . ~tempted to take precautionary steps to „ : protect ourselves against hazards that are not known to exist-and this unprec- ; were concerned that experiments in- edented act of caution is so novel that it :': volving certain genetic combinations Recombinant DNA research is not a has been widely misinterpreted as im- plying the imminence or at least the likeli- hood of danger. aez, Much has been made of the fact that; even if aa particular recombinant' DNA molecule shows no evidence of being yn~; hazardous at the present time, we are f unable to say for certain that it will not devastate our planet some years hence. Of course this view is correct; similarly, we are unable to say for certain that the vaccines we are administering to millions of children do not contain agents that will produce contagious cancer some years hence, we are unable to say for certain that a virulent virus will not be brought to the United States next winter by a traveler from abroad, causing a nationwide fatal epidemic of a hitherto unknown disease-and we are unable to '_7 say for certain that novel hybrid plants being bred around the world will' not : suddenly become weeds that will over- ' ; come our major food crops and cause ~s ; worldwide famine. .. The statement that potential hazards : t could result from certain experiments "' involving recombinant DNA techniques is akin to the statement that a vaccine injected today into millions of people could lead to infectious cancer in 20 years, a pandemic caused by a traveler- borne virus could devastate the United States, or a new plant species rnald un- ~ controllably destroy the world's food N supply. We have no reason to expect ~ that any of these things will happen, but ~ we are unable to say for certain that they -~ will not happen. Similarly, we are unable ~ to guarantee that any of man'sefforts to influence the earth's weather, explore space, modify crops, or cure disease will not carry with them the seeds for the ultimate destruction of civilization. Can ~`single entity, but rather it is a group of techniques that can be used for a wide variety of experiments. Much confusion has resulted from a lack of understanding of this point by many who have wntten about the subject. Recombinant DNA techniques, like chemicals on a shelf, are neither good nor bad per se. Certain expenmentsthat can be done wtth these techniques are likely to be hazardous (just as certain experiments done with combinations of chemicals taken from the shelf will be hazardous), and there is universal agreement that such recombi- nant DNA experiments should not be done. Other experiments in which the very same techniques are used-such as taking apart a DNA molecule and putting segments of it back together again-are without conceivable hazard, and anyone who has looked into the matter has con- cluded that these experiments can be done without concern. Then, there is the area "in between." For many experiments, there is no evi- dence of biohazard, but there is also no certainty that there is not a hazard. For these experiments, guidelines have been developed in an attempt'to match a level of containment with a degree of hypo- thetical risk. Perhaps the single point that has been most misunderstood in the controversy about recombinant DNA re- search, is that discussion of "risk" in the middle category of experiments relates entirely to hypothetical and speculative possibilities, not expected consequences or even phenomena that seem likely to occur on the basis of what is known. Unfortunately, much of the speculation has been interpreted as fact. There is nothing noveLabout the prin- ciple of matching a level of containment

nant DNA. ~,~,Unfortunately, the public has been led to believe that the biohazards described a<. , in various scenarios are likely or prob- able outcomes of recombinant DNA re- search. "If the scientists themselves are concerned enough to raise the issue," goes the fiction, "the problem is prob- ably much worse than anyone will ad- ;mit." However, the simple fact is that ,: there is no evidence that a bacterium carrying any recombinant DNA mole- cule poses a hazard beyond the hazard that can be anticipated from the known ~1properties of the components of the re- combinant. And experinients involving -~~;~ ` genes that produce toxic substances or +>~- pose other known hazards are prohibit- ~ ~. .- ed. .. _ . ,.. . - r . , 4i ;xwe in fact point to one major area of ered the possibility of hazard could exer- How About the Benefits? ~r human activity where one can say f'or cise appropriate restraint. While most cei tht th iik? Ptititiddfdthirihttf rtanaeres zero rsoen- scensswoueenergoree- `tially, we could respond to such risks by ' dom of scientific thought and discourse, ~` 'taking measures such as prohibiting for- I do not know of anyone who has pro- :eign travel to reduce the hazard of dead- posed that scientists should be free to do Iy virus importation and stopping experi- whatever experiments they choose re- , miih hbridl I idlfh entaton wty pants.ts pos- garess o te consequences. sible to develop plausible "scare sce narios" involving virtually any activity ~4," '' :or process, and tttese would have as Interference with "Evolutionary '` much (or as little) basis in fact as most of „"' Wisdo the scenarios involving recombinantm DNA. But we must distinguish fear of } z°~Some critics of recombinant DNA re- ,!the unknown from fear that has some 'search ask us to believe that the process basis in fact; this appears to be the crux "`of evolution of plants, animals, and mi- of the controversy surrounding recombi- '"crobes has 'remained delicately con- ° t alld filliofd that roeor mns o years, an the , `construction of recombinant DNA mole- cules now threatens the,~master plan of evolution. Such thinking, which requires a belief that nature is endowed with wisdom, intent, and foresight, is alien ~ to most post-Darwinian biologists (3). f" Moreover, there is no evidence that the ' evolutionary process is delicately con- trolled by nature. To the contrary, man has long ago modified the process of ` evolution, and biological evolution con- tinues to be influenced by man. Primitive man's domestication of animals and culti- vation of crops provided an "unnatural" advantage to certain biological species and a consequent perturbation of evolu- tion. The later creation by man of hybrid - plants and animals has resulted in the propagation of new genetic combinations that are not the products of natural evolu- tion. In the microbiological world, the use of antimicrobial agents to treat bacte- rial infections and the advent of mass Freedom of Scienti6c Inquiry issue has been raised repeatedly ~iv`-~. -•i during discussions of recombinant DNA Y~ research. "The time has come," the crit- charge„ "for scientists to abandon their long-held belief that they should be .~. _ "` - free to pursue the acquisition of new ~ knowledge regardless of the con- t ~~J as pr s~ ::.sequences." The fact is that no one has proposed that freedom of inquiry should extend to scientific experiments that en- -".danger public safety. Yet, "freedom of -.seientific inquiry" is repeatedly raised as 'a straw-man issue by critics who imply that somewhere there are those who ar- gue gue that there should be no restraint whatsoever on research. Instead, the history of this issue is one - of self-imposed restraint by scientists from the very start. The scientific group that first raised the question of possible hazard in some kinds of recombinant DNA experiments included most of the acientists involved in the development of the techniques-and their concern was made public so that other investigators who might not have adequately consid- t ld FEBRUARY 1977 For all but a very few experiments, the risks of recombinant DNA research are speculative. Are the benefits equally speculative or is there some factual basis for expecting that benefits will occur from this technique? I believe that the anticipation of benefits has Isubstantial basis in fact, and that the benefits fall into two principal categories: (i) advance- ment of fundamental scientific and medi- cal cal knowledge, and (ii) possible practical applications. 7-' In the short space of 3 rs years, the use of the recombinant DNA technology has' already been of major importance in the 'advancement -of fundamental knowl-, '° edge. We need to understand the struc- ture and function of genes, and this meth- odology provides a way to isolate large ~ quantities of specific segments of DNA in pure form. For example, recombinant DNA methodology has provided us with much information about thastructure o `plasmids that cause antibiotic resistance in bacteria, and has given us insights into how these elements propagate them- selves, how they evolve, and how their genes are regulated. In the past, our'J inability to isolate specific genetic re- "`;« gions of the chromosomes of higher orga- nisms has limited our understanding of the genes of complex cells. Now use of recombinant DNA techniques has pro- vided knowledge about how genes are organized into chromosomes and how gene expression is controlled. With such knowledge we can begin to learn how defects in the structure of such genes , alter their function. ~ ' immunization programs against viral dis- On a more practical level, recombi- ease has made untenable the thesis of nant DNA techniques potentially permit b delicate evolutionary control. the construction of bacterial strains that ' A recent letter (4) that has been widely can produce biologically important sub- quoted by critics of recombinant DNA stances such as antibodies and hor- research asks,_ "Have we the right to mones. Although the full expression of ~ counteract irreversibly the evolutionary higher organism DNA that is necessary wisdom of millions of years ...?" It is to accomplish such production has not this so-called evolutionary wisdom that yet been achieved in bacteria, the steps gave us the gene combinations for bubon- that need to be taken to reach this goal ic plague, smallpox, yellow fever, ty- are defined, and we can reasonably ex- ' phoid, polio, diabetes, and cancer. It is pect that the introduction of appropriate _ this wisdom that continues to give us "start" and "stop" control signals into N ~ uncontrollable diseases such as Lassa recombinant DNA molecules will enable fever,, Marburg virus, and very recently the expression of animal cell genes. On ~.~ the Marburg-related hemorrhagic fever an even shorter time scale, we can ex- ~~ virus, which has resulted in nearly 100 pect recombinant DNA techniques to ''N percent mortality in infected individuals revolutionize the production of antibiot- =y; ics, vitamins, and medically and indus- trially useful chemicals by' eliminating the need to grow and process the often exotic bacterial and fungal strains cur- rently used as sources for such agents. We can anticipate the construction of modified antimicrobial agents that are WN' ~; , 4 in Zaire and the Sudan. The acquisition and use of all biological and medical knowledge constitutes an intentional and continuing assault on evolutionary wis- dom. Is this the -warfare against na- ture" that some critics fear from re- combinant DNA?

.. . ~ ~ ^ - ~ ~ ~ . ., ~ .... . .. ~~' not destroyed by the _ antibiotic in-:'' been misled into taking what I believe to trances, negative air pressure, and spe- ' activating enzymes responsible for drug be an antienvironmental position on the cial air' filtration devices. Facilities resistance in bactena tssue of recombinant DNA + r ~~i'where P3 experiments can be performed ,,:,In the area of vaccine production, we `~ ~,4 ~~~, ~ f~~~~$ , ~~'are limited in number, but they exist at "tcan anticipate the construction of specif- ~'%;, : : &j ~`'~;~~ ~~ ~~ some universities. 7 ~ ~ ic bacterial strains able to produce de- w The NIH Guidelines 4) Experiments in which the hazard is -~+#~- sired antigenic products, eliminating the ,;,,;~~ onsidered unlikely to be serious even if jj ~~, present need for immunization with Even if hazards are speculative and it occurs still require laboratdry proce• ~ 4killed or attenuated specimens of dis- ' the potential benefits are significant and dures (P2 containment) that 11ave for F_ease-causingviruses ~~`r convincing' wouldn't it still be better to ` years been considered sufficient for re- w ... '~~'~~€.~ ,. 4'; One ptactical application of recom- carry out recombinant DNA expen , search with such pathogenic bacteria a ~s Tbinant DNA technology in the area of ments under conditions that provide an "Salmone!!a r phosa Clostrrdium bot ,- ;Wvaccinc production is already close 6 ~J added measure of safety-just in case ulinum, and Cholera vibrio. The NIH a•; L being realized. An E. coli plasmid coding some of the conjectural hazards prove to ':; guidelines require that P2 facilities be or an . enteric toxin fatal to livestock ~~ be real? s.~ ~.~'~~~:~~~;~used for work with bacteria carrying in- has been `taken apart and the toxin. ~This is exactly what is required under ~terspecies recombinant DNA molecules , gene has been separated from the re- " the NIH (National Institutes of Health) that have shown. no evidence of being mainder of the plasmid. The next step .'.guidelines (5) for recombinant DNA re- . hazardous-and even for some recombi- is to cut a`way a smali segment of the search s~y~~,t;:~u +"1'':nant DNA experiments in which there is, ' toxin-producing gene sothat the sub-r r111) These guidelines prohibit experi- substarrtial evidence of lack of hazard. j~stance produced by the resulting gene in,; ments in which there is,some scientific `_ S) The PI (lowest) level . of con- f coli will not have toxic properties but basis for anticipating that a hazard will tainment can be used only for recombi- will be immunologically active in stimu- : occur. In addition, they prohibit experi-"nant DNA molecules that potentially can _ lating antibody production Fments in which a hazard, although it. be made by ordinary biological gene ex-. ~,' ~,,Other benefits from recombinant DNA :~might be entirely speculative, was -change in bacteria. Conformity to even "research in the areas of food and energy y. judged by NIH to be potentially serious - this lowest level of containment in the :'production are more speculative. How-' enough to warr•ant prohibition of the ex- laboratory requires decontamination of ~~~'x=ever, even in these areas there is a scien- `"periment. The types of experiment that work surfaces daily and after spills of =AWtific basis for expecting that the benefits were the basis of the initial "moratori- biological materials, the use of inechani- 4 . + ; will someday be realized. The limited um" are included in this category; con- cal pipetting devices or cotton plugged '~-g ~. availability of fertilizers and the potential trary to the statements of some who have pipettes by workers, a pest control pro- ~. •fiazards associated with excessive use of _ written about recombinant DNA re- ; gram, and decontamination of liquid and n fertilizers now limits the yields search there has in fact been no lifting of solid waste leaving the laboratory nitroge,. A~ of grain and other crops, but agricultural the original restrictions on such expen ':.,(,In other areas of actual or potential ,: 1 ; experts suggest that transplantation of inents. biological hazard, physical containment the nitrogenase system from the chromo- -~ 2) The NIH guidelines require that a is all that microbiologists have had to ' somes of certain bacteria into plants or large class of other experiments be car- rely upon; if the Lassa fev ~er virus were into other bacteria that live symbiotically ried out in P4 (high level) containment to be released inadvertently from a P4 ;.. ,.;. with food' crop plants may eliminate the facilities of the type designed for work facility, there would be no further barrier need for fertilizers. For many years, Sci- • with the most hazardous naturally occur- to prevent the propagation of this virus sntists have modified the heredity of ring microorganisms known to man which is known to be deadly and for ; plants by comparatively primitive tech- (such as Lassa fever virus, Marburg vi- which no specific therapy exists. How- niques. Now there is a means of doing rus, and Zaire hemorrhagic fever virus). ever, the NIH guidelines for recombi- this with greater precision than has been It is difficult to imagine more hazardous nant DNA research have provided for an '?'al '"possible previously. ^,'' self-propagating biological agents than additional level of safety for workers and Ceain algae are known to produee such viruses, some of which lead to near - the public: This is a system of biological ~ rt using sunlight as ly 100 hydrogen from water, percent mortality in infected indi- containment that is designed to reduce energy. This process potentially can viduals. The P4 containment requires a by many orders of magnitude the chance yield a virtually limitless source of pollu- specially built laboratory with airlocks of propagation outside the laboratory of tion-free energy if technical and biochem- and filters, biological safety cabinets, microorganisms used as hosts for re- ical problems indigenous to the known clothing changes for personnel, auto- combinant DNA molecules. . `~ . hydrogen-producing organisms can be claves within the facility and the like An inevitable consequence of these ,. solved. Recombinant DNA techniques This level of containment is requiredtor containment procedures is that they offer a possible means of solution to recombinant DNA experiments for have made it difficult for the public to these problems. which there is at present no evidence of appreciate that most of the hazards un- It is ironic that some of the most vocal hazard, but for which it is perceived that der discussion are conjectural. Because opposition to recombinant DNA re- the hazard might be potentially serious if in the past, governmental agencies have search has come from those most con- conjectural fears prove to be real. There often been slow to respond to clear and cerned about the environment. The abili- are at present only four or five installa- definite dangers in other areas of tech- :~ ty to manipulate microbial genes offers tions in the United States where P4 ex- nology, it has been inconceivable to sci- the promise of more effective utilization periments could be carried out. - entists working in other fields and to the of renewable resources for mankind's 3) Experiments associated with a still public at large that an extensive and food and energy needs; the status quo lesser degree of fiypothetical risk can be costly federal machinery would have offers the prospect of progressive and conducted in P3 containment facilities. been established to provide protection in continuing devastation of the environ- These are also specially constructed lab- this area of research unless severe haz- ment. Yet, some environmentalists have oratories requiring double door en- ards were known to exist. The fact that ~ 6% . -^- ` ~S(.9ENCE VOL. 193 C

,.ri....5. ~. N 5 f ~ recombinant DNA research has prompt, imaginable circumstances than one can feted by ancicnt and new diseases. ,' and-~` ed international meetings, extensive cov- prove that currently administered vac- by malnutrition and pollution; recombt erage in the news media, and govern- cines do not contain an undetected self- nant DNA techniques offer a reasonable "! M- `mental intervention at the federal level propagating agent capable of producing expectation for a partial solution to some ~ ,'-.ltas been perceived by the public as cancer in the future, or that a hybrid of these problems. Thus, we must ask ' rima facie evidence that this research plant created today will not lead to disas- whether we can afford to allow pre ' must be more dangerous than all the rest. • trous consequences some years hence. occupation with and conjecture about The scientific community's response has "~No matter1what evidence is collected to hazards that are not known fo exist, to ~ been to establish increasingly elaborate document the safety of a new therapeutic limit our ability to deal with haiards thaO ~ procedures to police itself-but these agent, a vaccine, a process, or a particu- do exist. Is there in fact greater risk in ' - - to be necessary. ~~~:;ed; the experiments that can be done to proposed by some. We must then exam'' ftids; ~ l kidfbint DNAleldi jdiily or it ~~ very acts o scientific cauon an repon-arn o recoman mocue, proceengucous,n no pro- N.sibility have only served to perpetuate one can always conjure up the possibility ceeding at all? We must ask whether .~ ' and strengthen the general belief that the of future hazards that cannot be dis- there is any rational basis for predicting ; `~ hazards under discussion must be clear- ~proved• When one deals with conjecture, , the dire consequences of recombinant ;~.cut and imminent in order for such steps , the number of possible hazards is unlimit- DNA research portrayed in the scenarios `~ ~ ~~, 1~ It is worth pointtng out that despite establish the absence of hazard are finite ine the "benefit" " side of the picture and ` ,,' predictions of imminent disaster from ~in number. "."!P"u•~~`~.;k~' weigh the already realized benefits and ` i ddi h fa th f i l D i bl e expectat a ona e reasona on o t NA exper ments, t e recombinant ct Those who argue that we should not sr.L"z' ;~,remains that during the past 3th yeacs, use recombinant DNA techniques until benefits, against the vague fear of the `imany billions of bacteria containing a or unless we are absolutely certain that unknown that has in my opinion been the ~ :wide variety of recombinant DNA mole- there is zero risk fail to recognize that no focal point of this controversy. r3~~PT+J.~. . ) ., _,,;.. . ,. .. . , . :r 3. :-.r.~. eules have been grown and propagated in one will ever be able to guarantee total the United States and abroad, incorporat- . freedom from risk in any significant hu- ~ 1 P Berg, D_ Baltimore, H. W Boyer. S. N.' .., iag DNA from viruses, protozoa, in- man activity. All that we can reasonably Cohen. R W. Davis, D. S. Hogness, ID• Na-~ j,. sects, sea urchins, fro s, yeast mam- '. ex ect is a mechanism for dealin re- ~~s,R•RobGn 7•D•WUson,S.Weissman,N .: g + P g i< D. Zinder, Proc• Natl•:AcaQ Sci. U.S.A• 71, mals, and unrelated bacterial species in- sponsibly with hazards that are known to 2s93 (1974)• = 2. S. N: Cohen. A. C. Y. Chang, H. W. Boyer, R. to E. eoli, without hazardous con- exist or which appear likely on the basis , B Heuin ibid 70 3240 (1973); S N Cohen , g .,...., em .. Beyond Sci. Am. 233 (`lo: 7). 24 (i973). sequences so far as I am aware And the of information that is known - . - k. ~~; 3. lf we accept the view that any natural barriers to ;,4r ,., majority of these experiments were car- this, we can and should exercise caution the propas,tion ofgenetic mater;aldet>ived from ~ re- 5`` u^rclated species do not owe thcir existence to rior to the strict containment in an activit that carries us into `ried out Despite the experience thus far, it will a new drug or vaccine, or bringing a limitations to gene exchange have evotvea bo- i of genes from diverse orga- ; ause the mixing r always be valid to ar ue that recombi- s aceshi back to Earth from the moon. g P P nisms is biologically undesirable-not in a morat nant DNA molecules that seem safe • Today, as in the past, there are those or theological sense as some observers would in- today may prove hazardous tomorrow. who would like to think that there is voi ~a us believe-but to those organisms tn-~ ~ 4 4 ~. ?~~lb .Ff;,~jG!f'X >X sS~ • y y p p , }:the intent of nature, we can reason that evolu- 'a ~ procedures specified in the current feder- viously uncharted territory, whether it is ~•~ tion has created and maintained such barriers :, al uidelines. recombinant DNA research creatiomof ~a°"D oppoltunities for genetic mizing ocwr ..~ g + in nature. Furthermore, wa must conclude that . 938 (197 1 E Chff Sr 192,6),; One can no more prove the safety of a freedom from risk in the status quo..:arga,cence .• S. Fed.Reg•41(176)(9September1476),pp•3842tL- °: : patticular genetic combination under all However, humanity continues to be buf _: 38483. +~ 4 r~ ~ c,; , r K~ ! ra f- r • ~ . . , ..... • .. .. . . .. . _ .. . X~ri_ .. .. ~:i . t.. ..~'l r - - . '. ' . .. ~ ~ . . . - . . : ~ 'r. '4 x >w~+..[ l . .. ~, y~ . s. ~. p r: _ r x~t r s Y e- '+y f S i . .w'74 r. _ Y .n t. , i" ~a}! 4 4 >.AeF, •:-3 ~. ~ ~ i.Yl(i f ' -.i~.. `X~.. ....'a i}.. . . i;! ~d-r f w°~.T .~ JJL~ [~a.54Yr ~a r<: r L~ J r ,t, ...J.....n ' ! . ; '~.µ'ss'GS.'Y~ ff