Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
11 Conclusions and Recommendations Microorganisms Mankmd hap a long history of using microorganisms ~ food processing, agriculture, waste treatment, and in other beneficial am plications. New molecular methods for genetically modifying m~- croorg=~srns will expand the range of beneficial applications, for example, in control of plant disease ~d ~ biodegradation of toxic pollutants. In many respects, molecular methods resemble the classical methods for modifying particular strains of microorganisms, but many of the new methods have two features that make them even more useful than the classical methods. Precision allows scientists to make genetic modifications in ~n~crobial strains that can be cha~- acter~zed more fully, in some cases to the level of the DNA sequence. This reduces the degree of Certainty associated with any intended application. The new methods have greater power because they enable scientists to isolate genes and transfer them across natured barriers. The power of these new techr iques creates the opportunity for new applications of microorganisms. Despite some initial concerns over the use of recombinant methods in laboratory research' it is now clew that these methods in themselves are not intrinsically dangerous. The next step after laboratory experunentation is to test modi- fied microorganisms ~ the field, arid establishing a scientifically based 123
i24 framework for decisions on field testing has been a primary purpose in this report. No adverse effects of introductions have been seen and an extensive body of information documents safe introductions of some microorganisms, such as the rhizobia, mycorrhizal fungi, bacuToviruses, Bacillus thuringiensis, and Agrobacterium radiobac- ter. However, less is known about field tests of microorganisms than of plants. Thus, for unfamiliar applications, it is prudent to prepare for the control of the introduced rn~croorganisms. Questions concerning the effects of an introduced microorgan- ism arise whenever the intended introduction differs substantially from those with an established record of safety. Such questions as unintended persistence and possible adverse effects should be ad- dressed scientifically, and as the scientific community continues to accumulate information regarding the safety or risk of environmental applications of microorganisms in field tests, levels of oversight can be tuned to the needs of particular situations. ~ the recommendations that follow, a framework has been de- veloped as a basis for a workable and scientifically based evaluation of the safety of microorganisms intended for field testing. This frame- work has been developed from consideration of three criteria: (1) familiarity with the history of introductions similar to the proposed introduction (Chapter 7), (2) control over persistence and spread of the introduced microorganism as well as over exchange of ge- netic material with the indigenous microflora (Chapters 8 through 10), and (3) environmental effects, including potential adverse effects associated with the introduction (Chapters 9 and 10). The framework does not distinguish between classical and molec- ular methods of genetic manipulation, nor between modified and un- modified genotypes. The framework is product- rather than pracess- oriented, focusing on the properties of the microorganism rather than on the methods by which it is obtained. Knowledge of the methods used may nonetheless yield useful information concerning the precision of genetic characterization of the rn~croorganism, which in turn may be relevant for assessment of its si~nilarity to previous applications, persistence, and possible effects after introduction. The framework has not focused on other variables, often sug- gested as criteria for oversight, because they convey relatively less scientifically useful information for assessments: the sources of genes, whether recombinants are intra- or intergeneric, and whether cod- ing or noncoding regions of the genome have been modified. The
125 necessity of using, whenever possible, sunple and readily identifiable criteria for oversight is recognized. Terms such as Uncertainty, ~sufficient," and Significant are used in the framework without precisely defining their quantitative tenets. Any specific numerical values assigned would be arbitrary and subject to disagreement, as some underlying variables may be difficult to quantify precisely. In the final analysis, assignment of risk categories must mclude a rational examination of the relevant scientific knowledge for each introduction. In the framework, assessments of potential risks arising from the introduction of mucroorganisms into the environment are made ac- cording to the three major criteria of familiarity, control, and effects. Upon evaluation of these three criteria, a proposed introduction can be field-tested according to establisher] practice or it can be assigned to one of three levels of concern: low, moderate, or high uncertainty (Fig. 11-13. The framework is inherently flexible, allowing an appli- cation to be reassigned to a different category as additional scientific information is obtained that is relevant to any of the three criteria. Small-scale field tests can proceed according to established prac- tice if the microorganism used, its intended function, and the target environment are all sufficiently similar to prior Introductions that have a safe history of use (Fig. 11-2~. Rhizobium used for enhance- ment of nitrogen fixation in leguminous crops provides a familiar example. If an introduction does not satisfy the farn~liarity criteria, it is evaluated with respect to both our ability to control the m~croorgan- ism's persistence and disserrunation and the microorganism' poten- ti~ for significant adverse effects (Fig. 11-1~. For example, Rhizobium modified to encode an insecticidal toxin would not be a familiar ~ntro- duction, even though it might well prove to be safe. An introduction is considered to be in the low-uncertainty category if it satisfies appropriate criteria with respect to both controllability =d low po- tentia] to result In adverse effects. An Introduction ~ considered to be in the moderate-uncertainty category if it satisfies criteria for either controllability or potential effects, but not both. An mtroduc- tion ~ considered to be in the high-uncertainty category if it satisfies neither the control nor the effects criterion (Fig. 11-1~. The high uncertainty status implies that potential adverse effects exist and are coupled with potential inability to control the microorganism, and hence its potential effects.
126 em o ._ - o _ o C Z - - - Q Zen o ~ - w Cat Cal ' o o - ~o 0· Z ~ .=._ / ~, 1:5 o' _ o' _ _ ._ ~ 3 IS - 0 I: o ._ ~ _ ~ o ~ _ ~ - - _ ~ ._ o' ~ _ ~W - ._ C' _. _ o - c ._ - ._ - ._ - C' - J ~ . _ ~ ._ ~ 11 CO 0 - o' - ._ - 3 ~ o - Ct / W 1 ~1 =. Cd a, - / / ._ o C, Ct 0 \ ~ \ ~n o~ ~Z ~ - o' . ~ ~ _ ~ - ._ _ _ ~ Q ~ _ _ O _ _ Ct CO CO > ~ LU Ct ~ lL Ct - C' ._ - _ _ ._ O> ~ ._ ~ ~ O ~ ._ > ~: C.) ^ X O ~t a~ Q, _ ~V . _
127 AS - o a, CC ~ I ~.~ ~Ed ~_' _ <l,0 ' 0o c) o_ _o o' _ ~o __ ~_~ a, cts~ o _~ E a, 0 a' o ~0 ~_ ~o C" _ ~ ~ ~ ~ o ~ ._ ' o - o `:s 0 ~ - ._ E ~ ~ ~ ~_ ~ o ~ _ ~ _ ~ ~ _ it, ._ Ct ._ o ~_ ._ _ _ 0 ~ o' O ~ ~ _ Q - 0 C 0 ~ os ~ _ ._ _ ._ _ ._
128 UJ o G ~a' O o' ~o -LL.J ~ I_ < n <c G a U) a' Do Cot o DO o ._ C) - 1 v Cot Z /~\Z / Z \ ,`: ^ ~Cot ~ TV 0 I,LI! ·_ _ ~_ ~J o _ ~O o Cal ~_ .0 a' o' ~0 ~ r ~: ~' ~ ~V 0 ~C) ~ 0 C] _ ~_ .c, ~CtS ~ ~_ 4) o ~_ A A: _ - o a, ~o'
129 0 C, _ C ~- o a, ~ ~_ c, ~ 0 . ,£' _ ~ Z ~ o Z o ~ O ~ 0 o _ ~DO , ~ ~ _ 0 ·_ 0 o C~ o _ CId ~y \~ · ~ . C" C _ ~ ~UJ o ~ ~Z o ~ ~_ ~, Z .e ._ ~.O ._ ~O . ~ o ~o ' - _ ~ C" o - o ' oC ~ ·_ Z C, Z ~ ~Z ~.o - ~ n n ~E ~ ~I ~ ~- ° ': 0 ~'`,, 0 ~s~ ~m ~ ~ ~ C, 0 ,. ._ ~ E o D e ~ 40 0 ~ C c ,, = < ~ · - E ~ {~ 0 ~5 - o.~ ~ ° o' ~ ~ , ~_
130 Q o o ~._ cut ~ ^ cut 0 _ a' ~ E4, ~n~ n ,,: wc a, ~ 4,, as w 0~ 0 ~ m.> ~ ~ o) ~ - ._ C _e e ° ° O _ ~_ ~ ° ° 04= ~ a' __ C\5 O Q 0 _ .- Q =~=~_ , ,~ ~ , 2 . ~ · an, ~ 1~: 43,·- ~ Q E ~ O ~ ~ Q 0-= C ~ ~ _- O4 'A 01 1 1 ooo ~-A ~ C.
131 Specific criteria for evaluating control of the microorganism af- ter it is introduced must include the potentials for persistence of the introduced microorganism, genetic exchange between the intros duced and indigenous microorganisms, and spread of the introduced microorganism to nontarget environments (Fig. 11-3~. A series of questions to be addressed in evaluating the potential for unwanted persistence of an introduced microorganism Is illustrated In Fig. 11-4. Criteria for evaduat~g ejects must depend, at least in part, on the intended function of the introduced microorganism in its target environment (Fig. 11-5~. Thus, a proposed field test of a bacterium to be used for biodegradation of a toxic poDut ant should be preceded by definitive laboratory experiments and should be designed to determine whether toxic by-products of the degradation may be created and persist. As the agencies grant permission to introduce genetically mod- ified microorganisms in field tests, they will receive advice Dom panels of experts who can utilize the decision framework described here. With experience, familiarity wiD increase, and we anticipate this will be accompanied by adjustments in the rigor of oversight.