Environmental Health PerspectivesVol. 18, pp. 111-116, 1976

Techniques for Assessment of TeratogenicEffects: Developmental Enzyme Patternsby F. D. Andrew*6

Most studies designed for assessing teratogenic effects focus on only three of four types of finalmanifestations of abnormal development; namely intrauterine death, malformations, and growthretardation. Developmental toxicity evaluations generally do not include functional deficits.Current techniques are inadequate for assessing functional capability during perinatal develop-ment, and there is a need for improved measures. Thus, measurement of developmental enzymepatterns is proposed as an approach that directly evaluates acquisition of metabolic competence offundamental organ systems. During ontogenesis most organs acquire their full complement of en-zyme activities in a programmed sequence which corresponds to attainment of complete func-tional capability. The usual alterations in enzyme activity that characterize these patterns occur atone of three time periods, namely, late fetal, early neonatal, or late suckling. Qualitative or quan-titative changes in these patterns at any time by one or more key enzymes of a tissue could be in-dicative of developmental toxicity. Factors are outlined relating to consideration of developmentalenzyme profiles as indices of maturation capable of reflecting the action of toxic agents. Thispresentation: (1 I reviews the current state of the art for evaluating effects on development, (2) con-siders the applicability of enzyme patterns for biochemical assessment of development, 434characterizes the target tissues and those metabolic pathways and/or specific enzymes most sen-sitive and adaptable to practical toxicity evaluations, and (44 describes the steps being taken tovalidate this system. The ultimate objective of this approach is to determine whether alterations indevelopmental enzyme profiles will provide a technique with improved capability for assessingdevelopmental toxicity.

IntroductionThere is a growing recognition of a need for

more sensitive and meaningful methods for as-

sessing the adverse effects of chemical agents on

human health in general and on developing in-dividuals in particular. This presentation willconsider the use of enzyme profiles as indices ofmaturation, which, in turn, may reflect the actionof toxic agents. This approach, measurement ofenzyme patterns in developing tissues, directlyevaluates the acquisition of metabolic compe-tence by fundamental organ systems.Most studies designed for assessing terato-

genic effects, regardless of the type of agent,focus on only three of four final manifestations of

*Biology Department, Battelle, Pacific NorthwestLaboratories, Richland, Washington 99352.

abnormal development; namely, intrauterine orneonatal death, malformations, and growth retar-dation (1). For these studies the agents are givenduring major organogenesis, thereby producingexposures of the embryo. Evidence is ac-cumulating which suggests that late exposure ofthe fetus and possibly even neonatal exposures tosome agents may have deleterious developmentaleffects. Perinatal mortality is still a very realproblem for infants throughout the world, andovert symptomology, such as prematurity withrespiratory distress and/or jaundice, is seenfrequently. Other effects may be less obvious.For instance, minimal brain dysfunction (MBD) inchildren may be an example of a subtlemanifestation of a molecular, if not a cellular,lesion. While a chemically induced animal modelfor MBD has been developed in neonatal rats, itsetiology in children is uncertain (2, 3).

December 1976 III

Much of the work related to developmentaltoxicity has been directed toward assessing mor-phological and cyotogenetic changes and to alesser extent, the behavioral aspects. We knowvery little regarding effects on biochemicalsystems, that is, the effects of agents upon theprocess of development after majororganogenesis (1, 4). Unlike the gross structuralmanifestations of embryotoxicity, effects at themolecular level are not readily evident, and maybe revealed only by special tests. A molecular orbiochemical malformation may remain dormantfor a period, then manifest itself later in life as afunctional impariment, such as: behavioral disor-der, learning disability, infertility, tumor suscep-tability, or immunological defect. Most organsduring gestation are potential targets for insultsthat may produce cellular and/or molecular le-sions with profound developmental disturbancesfor the individual.Developmental toxicity evaluations generally

do not include these types of functional deficits.Current techniques are inadequate for assessingfunctional capability during the perinatal period,and there is a need for improved measures (4).These problems require evaluation by whatevermethods seem most appropriate. One approach isto focus on age-dependent changes in enzyme ac-tivity. Developing tissues contain distinctive en-zyme complements that exhibit dramatic changesin their activity coinciding with the criticalphases of the perinatal period. Greengard (5) hasrecommended that altered enzyme patterns beconsidered "biochemical malformations" thatmay serve as more subtle, sensitive and reliableindicators of perinatal dysfunction than ob-servations of growth and survival rates. To datethere have been few attempts to monitor effectsof chemical agents on enzymes during develop-ment other than those involved in oxidative drugmetabolism (6, 7). Significant impairment ofneonatal rat liver and brain enzymes have beenreported after irradiation in utero (8-10). Never-theless, there have been no comprehensive at-tempts to fully explore and characterize this ap-proach to developmental toxicity testing.

Several aspects must be considered in outliningany proposed model for relating alterations infunctional maturation with altered enzymeprofiles. This requirement has been considered asa series of specific questions that can be an-swered analytically and experimentally (Table 1).The first need is to consider the applicability oftissue enzyme patterns as biochemical indicatorsof functional capability. Second, it is necessary to

Table 1. Specific aims: to find answers to these questionsregarding the proposed system.

ApplicabilityHow are perinatal studies important?What relevance have they to humans?What agents alter development?When do the agents produce their effects?

CharacterizationWhich tissues are preferentially affected?Which enzyme pathways are altered?How should specific enzymes be selected for analysis?Are qualitative or quantitative effects more important?

Validation: how can this approach be used to:Ask environmentally relevant questions?Design practical test procedures?Evaluate environmentally related agents?Predict human health hazards?

characterize the presumptive target tissues andthose pathways, especially their key enzymes,that are most suitable for practical toxicityevaluations. Third, the system still requiresvalidation by initially testing classical physicaland chemical teratogens before attempting todemonstrate broad applicability by testingputative agents.

ApplicabilityThe proposed system is based on the hypoth-

esis that chemically induced changes in activity ofperinatal enzyme systems during developmentcan be detected and can serve as biochemical in-dicators of insult to developing animals.During mammalian ontogenesis most organs

acquire their full complement of enzyme activi-ties in a programmed sequence which corre-sponds to attainment of complete functionalcapability (4, 10-13). Marked variations in manymetabolic functions also occur during develop-ment. These changes are particularly dramatic atbirth, with the abrupt transition from intra-uterine to extrauterine environment and duringthe weaning period, when great variations innutrition occur. These can be further divided intothe late fetal, neonatal, and late suckling periods.A separate cluster of enzymes attains approxi-mately adult levels of activity during each ofthese periods, coinciding with achievement of in-creased metabolic competence (4, 13).To date, most perinatal studies involving en-

zyme patterns have used either normal or en-docrine-modified animals to reveal the nature ofthe regulatory factors for gene expression duringontogenesis (11). Thus, the approach is relativelynew to developmental toxicity evaluations.

Environmental Health Perspectives112

Some toxic substances of current interest arealdrin/dieldrin, DDT, endrin, cadmium, lead,mercury, zinc, alkyl sulfonates, hexachlorophene,PCBs, poly(vinyl chloride), TCDD, ethanol,radionuclides, and in addition, malnutrition.While malnutrition is not an agent, it is one of themost common causes of perinatal problems (14).The other items listed represent examples ofpesticides, industrial effluents, solvents, or sur-factants with potential hazard as developmentaltoxicants. Studies on developmental toxicityhistorically have involved use of single agents ad-ministered under controlled conditions.Generally, the animals used were supplied withadequate to excess amounts of all essentialnutrients and were maintained in a uniform en-vironment. The general population does not liveunder such ideally controlled conditions. Actual-ly, those most likely to experience inadvertantexposures to harmful agents are those most likelyto suffer dietary imbalances, incontrolled drugusage, endocrine variations, or pathophysiologi-cal states that would increase their susceptabilityto an adverse reaction.

Test systems need to include some of thesepotentials for interactions along with usualprocedures to determine their role in modifyingthe animal's response (15, 16). Identification ofalterations in response following such exposureswill be helpful in predicting the synergistic orprotective actions of multiple influences withrespect to developmental toxicity.

It has generally been accepted that theperinatal and neonatal periods are among themost sensitive stages of life. The lack of ap-propriate data on the response of these immaturestages to exposure to toxic agents has preventedthe full utilization and consideration of tissue en-zyme ontogenesis as a test system.

CharacterizationA number of factors justify the need to charac-

terize developmental enzyme patterns. Of par-ticular significance is the determination of whichorgan(s) represent(s) the major site(s) for ad-verse effects on maturation during differentstages of development.There is need to consider the distribution of

selected enzymes in organs other than liver, in-cluding brain, lung, kidney, and intestine.However, the value of comparative organ studieswill be influenced by the physicochemical proper-ties of the agent, its probable route of entry, andits pharmacokinetics. Current studies are at-

tempting to characterize effects on the develop-mental patterns of several enzymes in varioustissues at different perinatal stages with classicalphysical and chemical agents. These types ofstudies also need to investigate the most com-monly occurring and biologically significant path-ways. Information gained from such studies canaid in predicting how the conceptus will respondto certain chemicals and whether this response isrelated to fetal disposition of toxic agents.

Analysis of enzyme patterns during hormonalor nutritional regulation and the control of geneexpression during differentiation and especiallyin neoplastic tissues have been conducted byWeber et al. (17-19). These studies, led to theidentification of a number of key enzymes invarious intermediary metabolism pathways thatwere considered to be the main targets ofmetabolic control. Some of the features thatcharacterize them have recently been discussed(18, 19). A molecular correlation concept wasdeveloped to provide a means for studying geneexpression at the molecular level and was basedon the assumption that various cellular functionscould be analyzed, correlated and understood inappropriate terms. Likewise, it is recognized thatfor an understanding of the regulation of thebiochemical patterns and their correlation withappropriate biological characteristics of develop-ment, experimental efforts must concentrate onkey enzymes and critical metabolic pathways.

Focusing on the key enzymes defined by Weber(17-19) precludes the necessity of assaying alarger proportion of the approximately 160 en-zymes identified in rat liver alone (20). Rather, aquartet of key gluconeogenic enzymes and a trioof key glycolytic enzymes may adequatelyevaluate effects on carbohydrate metabolism(21). Similar groups of enzymes operating inpurine, pyrimidine, DNA, ornithine, and cAMPmetabolism have been identified as key enzymes(21). The ultimate selection of target tissue en-zymes for testing thus requires consideration ofcharacteristics of the presumed target tissue aswell as those factors requiring specificknowledge, such as the probable action of the testagents. Such additional criteria for enzyme selec-tion are: (1) low levels related to human disor-ders; (2) easily measured by standard assays; (3)distributed in more than one tissue; (4) active insoluble or bound form; (5) stability of appropriatepreparations; (6) detectable in amniotic fluid.

Primarily, control of metabolism in mammals isachieved through modulation of activity and/oramount of regulatory enzymes. Thus, a

December 1976 113

significant quantitative change would be one inwhich the shift from control activity levels dif-fered by several orders of magnitude or a tem-poral deviation of more than one day was seen.Qualitative changes of note would be shifts inmetabolic pathways from primarily synthetic tomore degradative activity or shifts in isozymepatterns. Some of these aberrations might bepermanent, but the stage of the conceptus at thetime of exposure would influence the degree ofresponse. Qualitative or quantitative changes inthe enzyme patterns at any time by one or morekey enzymes of a tissue could be indicative ofdevelopmental toxicity.These studies should provide information con-

cerning: (a) which organ or organs are mostsusceptable to alteration of constituent enzymesat different ages, (b) what pathways are mostlikely to be affected in a particular perinatalperiod, and (c) what is the inter- and intralittervariation in response of offspring to gestationalinsult?

ValidationA number of factors are being considered in the

validation of this test system (Table 2). The in-sults selected for testing include salicylate andhydroxyurea (two classical chemical teratogens),

Table 2. Factors considered for design of developmental tox-icity evaluations.

Selection of test speciesOntogenic similarities to manPharmacokinetics of agent

Developmental stages for agent administrationMajor organogenesisLate fetal periodEarly postnatal period

Exposure ParametersAt least three doses ranging from maximum tolerated to

no apparent effectRoutes related to anticipated human exposure and proper-

ties of agentInclude both nontreated and vehicle controls

Dosage Schedule-Minimum RequirementMore than one developmental stage per doseAt least five animals per dose and stage

"Routine teratology" evaluation criteriaToxicity: maternal weight changeEmbryolethality: embryo resorptions and fetal deathsMalformations: external, visceral, and skeletalGrowth retardation: fetal weight

Postnatal parameters: functional capacityGrowth rate and survivalBehavior and learning abilityBiochemical and physiological processesInmunological determinants

radiation (a classical physical teratogen) andmalnutrition (a common environmental con-dition). There is experimental evidence that bothchemicals affect postnatal behavior in rats, whenadministered under the appropriate conditionsthat induce malformations, but given at lowerlevels (22, 23). Intrauterine exposure of rats andbeagles to x- or y-radiation has been reported toalter levels of several enzymes in liver and brainof neonatal animals (8-10). The potential fordevelopmental problems following maternalmalnutrition alone or in combination with suchphysical and chemical agents was discussedpreviously.Our initial studies will be restricted to rats. As

appropriate, other species may be included fortheir variations in degree of maturity of offspringat term and duration of gestation (4, 24). For thesubsequent investigations, pharmacokineticstudies may be conducted concurrently with en-zyme studies, depending on the agents selectedfor evaluation.

Test agents will be administered to the dameither during major organogenesis or during thelate fetal period or, possibly, during lactation.Administration directly to the offspring duringthe first 48 hr postnatally will also be con-sidered for some agents. Radiation will be giveneither as a single whole-body exposure or it willbe limited to the exteriorized gravid uterus.Magnitude and duration of malnutrition willdepend on the dietary components to be deleted,such as proteins, fats, carbohydrates, minerals,or vitamins.A general outline of the proposed experimental

design (Table 3) gives the major points regardingthe exposure parameters and dosage schedule.Generally, dose levels tested will range fromthose toxic to the dam to those that appear to bewithout effect to the conceptus and will include atleast one intermediate level. These studies willfocus on the more subtle effects on biochemicaland physiological processes that result from ad-ministration of low doses to immature animals ascompared to adults. Control animals will receivean equal volume of vehicle or an appropriatesham procedure on the same gestation days asthe treatments or no exposure. Body weights ofdams will be monitored periodically throughoutgestation. Generally, five to six litters will be in-cluded in each group.As indicated in Table 2, "routine teratology"

evaluations will be conducted on representativelitters to correlate morphological events with theenzyme activity data. These evaluations will in-clude: counting the implantation sites, noting the

Environmental Health Perspectives114

Table 3. General experimental design.

Gesta- Agenttional or

Developmental postnatal Salicy- Hydroxy- Radia-stage day late urea tion

Fertilization 1

Majororgan- 9 EXPbogenesis 10 INJ

111213 INJ

Late fetal 1819 INJ/SAC INJ/SAC EXP/SAC20 SAC SAC SAC21 SAC d SAC d SAC d22 SAC SAC SAC

Postnatal Birth1 INJ INJ EXP23 SAC SAC SAC4 SAC SAC SAC5 SAC SAC SAC6 SAC SAC SAC7 SAC SAC SAC

Preweaning 18 SAC SAC SAC19 SAC SAC SAC20 SAC SAC SAC21 SAC SAC SAC22 SAC SAC SAC

INJ = Developmental period for agent administration atappropriate dosage levels.

EXP = Developmental period for exposure to x ory-irradiation.

c SAC = Developmental periods for enzyme analysis and/ormeasurement of other parameters.

d"Routine" teratology evaluations are performed onrepresentative animals at gestation day 21.

general condition and viability of each conceptus,recording sex and weight for each live fetus,examining each pup for external malformations,and processing representative pups of each litterto detect visceral and/or skeletal alterations.

All enzyme measurements will be made byusing standard procedures whenever possible,but modifications of some procedures may beneeded. All enzyme assay methods will beverified under optimal conditions using purifiedenzymes, if commercially available. Generally,four to five consecutive ages of animals will beanalyzed at each cluster. Thus, sudden fluc-tuations in enzyme activity can be detected andactivity differences can be monitored throughoutspecific test periods.

There are a number of conceptual and technicalerrors that could lead to a lack of correlation bet-ween biochemical malformations and metabolic

dysfunction: (a) marked changes in indicators ofbiochemical function may occur without affectingfunctional reserve capacity; (b) significantchanges in enzyme activity may reflect reversiblefunctional effects without permanent sequela,: (c)altered enzyme activity may reflect a generalizedresponse to stress and not a specific response tothe toxic agent; (d) marked species variations inenzymatic alterations may occur in response to aspecific insult, especially if there are species dif-ferences in pharmacokinetics; (e) permanent ef-fects on enzymes may depend on a specific timefor the insult.

Thus, the effect of relevant gestational insultson perinatal enzyme patterns will be studied inreference to the following specific objectives.Developmental profiles of certain key enzymeswill be traced in hepatic and some extrahepatictissues at various time periods. Changes in en-zyme activity profiles will be correlated withother functional indicators. Enzyme activities ofcorresponding maternal tissues will be measuredsimultaneous with those of their pups at the ap-propriate times. The importance of alteredprofiles to growth rates and survival for times upto and including sexual maturity will be in-vestigated.

In summary, the ultimate objective of this ap-proach is to determine whether alterations in en-zyme profiles will provide a technique with im-proved capability for assessing developmentaltoxicity.

In October 1975, National Institute of Environmental HealthSciences (NIEHS) sponsored a conference entitled, "EnzymePatterns in the Evalution of Developmental Toxicity" at theResearch Triangle Park, N.C. The informative comments, ob-servations, and discussions of the invited speakers and otherparticipants were helpful in formulating my ideas. However,the statements given in this presentation are my own opinions,biases, and prejudices and are not intended to reflect those ofthe workshop participants. Special thanks go to Dr. Robert L.Dixon and other staff members at NIEHS for their contribu-tions to the success of the workshop.

In addition to NIEHS staff scientists, participants included:K. P. Chepenik, M. Feigelson, 0. Greengard, R. W. Hanson, S.Harris, A. Herzfeld, E. M. Johnson, E. Kun, J. Murray, B. A.Schwetz, W. J. Scott, B. R. Sonawane, K. Taketa, J. B. War-shaw, and G. Weber.

REFERENCES1. Wilson, J. G. Environment and Birth Defects. Academic

Press, New York, 1973, Chapt. 2.2. Shaywitz, B. A., Yager, R. D., and Klopper, J. H. Selective

brain dopamine depletion in developing rats: An experi-mental model of minimal brain dysfunction. Science 191:305 (1976).

December 1976 115

3. De La Cruz, F. F., Fox, B. N., and Roberts, R. H. Minimalbrain dysfunction. Ann. N.Y. Acad. Sci. 205: 5 (1973).

4. Nelson, N. Principles for Evaluating Chemicals in the En-vironment. Report of the Committee for the Working Con-ference on Principles of Protocols for EvaluatingChemicals in the Environment. National Academy ofScience, Washington, D.C., 1975, Chapt. 10.

5. Greengard, 0. Enzymic and morphological differentiationof rat liver. In: Perinatal Pharmacology-Problems andPriorities. J. Dancis and J. C. Hwang, eds., Raven Press,New York, 1974, pp.15-26.

6. Fouts, J. R. Microsomal mixed-function oxidases in thefetal and newborn rabbit. In: Fetal Pharmacology. L. 0.Boreus, Ed., Raven Press, New York, 1973, pp. 305-320.

7. Lucier, G. W., et al. Postnatal stimulation of hepaticmicrosomal enzymes following administration of TCDD topregnant rats. Chem. Biol. Interact. 11: 15 (1975).

8. Smith, C. H., and Shore, M. L. Impaired development ofrat liver enzyme activities at birth after irradiation inutero. Radiat. Res. 29: 499 (1966).

9. Nair, V. Effects of exposure to low doses of X-radiationduring pregnancy on the development of biochemical sys-tems in the offspring. In: Radiation Biology of the Fetaland Juvenile Mammal. M. R. Sikov and D. D. Mahlum,Eds., U. S. Atomic Energy Commission, Oak Ridge, 1969,pp. 899-910.

10. Garner, R. J., Graves, J. P., and Lane, C. E. Irradiationand enzyme ontogenesis in the rat and beagle. In: Radia-tion Biology of the Fetal and Juvenile Mammal. M. R.Sikov and D. D. Mahlum, Eds., U. S. Atomic Energy Com-mission, Oak Ridge, 1969, pp. 975-983.

11. Greengard, 0. Effects of hormones on development of fetalenzymes. Clin. Pharmacol. Therap. (Part 2)14: 721 (1973).

12. Moog, F. The control of enzyme activity in mammals inearly development and in old age. In: Enzyme Synthesis

and Degradation in Mammalian Systems. M. Recheigl,Ed., Karger, Basel, 1971, pp. 47-76.

13. Greengard, 0. Enzymic differentiation in mammalian tis-sues. Essays Biochem. 7: 159 (1971).

14. Latham, M. C. Protein-calorie malnutrition in children andits relation to physiological development and behavior.Physiol. Rev. 54: 541 (1974).

15. Shakman, R. A. Nutritional influences on toxicity of envi-ronmental pollutants. Arch. Environ. Health 28: 105(1974).

16. Newberne, P. M. Influence on pharmacological experi-ments of chemicals and other factors in diets of labora-tory animals. Fed. Proc. 34: 209 (1975).

17. Ferdinandus, J. A., Morris, H. P., and Weber, G. Behaviorof opposing pathways of thymidine utilization in differen-tiating, regenerating, and neoplastic liver. Cancer Res. 31:550 (1971).

18. Weber, G. Ordered and specific pattern of gene expressionin neoplasia. Adv. Enzyme Reg. 11: 79 (1973).

19. Weber, G. Molecular correlation concept. In: The Molec-ular Biology of Cancer, H. Busch, Ed., Academic Press,New York, 1974, Chapt. 12.

20. Knox, W. E. Enzyme Patterns in Fetal, Adult, and Neo-plastic Rat Tissue. S. Karger, Basel, 1972.

21. Weber, G. Control of gene expression in carbohydrate,pyrimidine and DNA metabolism. Adv. Enzyme Reg. 9: 63(1971).

22. Butcher, R. E., Vorhees, C. V. and Kimmel, C. A. Learningimpairment from maternal salicylate treatment in rats.Nature New Biol. 236: 211 (1972).

23. Butcher, R. E., et al. Postnatal effects in rats of prenataltreatment with hydroxyurea. Teratology 7: 161 (1973).

24. Hanson, R. W. The choice of animal species for studies ofmetabolic regulation. Nutrition Rev. 32: 1 (1974).

116 Environmental Health Perspectives