14

Teratogenesis¹

Charles P. Dagg

Teratologists are concerned primarily with the causes development, and anatomy of morphologically abnormal individuals. Most investigators have focused their attention upon gross structural defects produced either by mutant genes, by noxious environmental stimuli, or by a combination of genetic and environmental conditions. This chapter deals mainly with studies on environmentally induced skeletal abnormalities in mice. The coverage is restricted to a few papers selected to illustrate some of the major factors that must be considered in teratogenic experiments with these animals.

Abnormalities arising from the action of single or multiple genetic factors are discussed in Chapters 8 and 15. Reviews summarizing experimentation with other animals and covering aspects of teratogenesis not dealt with in this chapter have been published by European Society ( 1963), Fishbein ( 1960, 1963), Giroud and Tuchmann-Duplessis ( 1962), Kalter and Warkany ( 1959), Nishimura ( 1964), Wilson and Warkany ( 1964), and Wolstenholme and O'Conner ( 1960).

As with many other types of investigations, the advantages in using mice are found in their relatively low cost, small size with consequent reduction in space required for breeding and maintenance, in the availability of genetically uniform strains, and in the large number and diversity of mutants affecting the development of structures of interest to teratologists.

Malformations induced by teratogens are the end result of morphogenetic interactions between the genes of the embryo which confer the potentialities for normal or abnormal development and the environment of the embryo which includes the dam's genotype and environment, in addition to the teratogen. Specific effects of some of these factors are known. The ones that will be considered are: time of treatment, dose-response relationships, route of administration, interaction between agents, protective treatments, diet, seasonal effects, maternal influences, the maternal genotype and the embryonal genotype.

ENVIRONMENTAL FACTORS

Time of treatment

A list of teratogenic treatments and their principal morphological effects in various strains of mice is presented in Table 14-1. Descriptions and photographs of abnormalities of the head and skull can be found in Kalter and Warkany ( 1957) and Kalter ( 1963). Abnormalities of axial skeleton are illustrated in Kalter and Warkany ( 1957), Kalter ( 1963), and Murakami and Kameyama ( 1963), and photographs of various limb and foot deformities are presented in Kalter and Warkany ( 1957), Nishimura and Kuginuki ( 1958) and Kageyama ( 1961a).

In general, each organ of the embryo passes through a period of development during which it is particularly susceptible to teratogenic treatments. This developmental phase is commonly referred to as the critical period for that organ. In most cases the critical period corresponds to the time at which the organ is developing most rapidly, so that by the time organogenesis is complete the organ no longer responds to teratogens. Occasionally the organ may be malformed by treatment prior to the appearance of its earliest visible rudiment. As used by various investigators, the term critical period refers to either the time of maximum frequency of response or to the entire period during which a particular malformation can be produced. In the latter case the length of the critical period is often dependent upon the intensity of the stimulus. An increase in intensity frequently extends the critical period slightly to earlier and later developmental stages.

Critical periods for some of the teratogens that produce gross morphological changes are shown in Table 14-2. There is some confusion in the literature about specific critical periods because not all investigators have used the same method for counting the days of gestation. Some have regarded the first 24 hours as day 0 while others have designated the same day as day 1. To facilitate comparisons of critical periods for the different teratogens listed in Table 14-2 the first 24 hours has been called day 1 in every case.

As shown in Table 14-2, the critical periods for various organs may overlap. Each organ, however, has a distinctive and relatively limited period of susceptibility to teratogens. Occasionally, a teratogenic treatment has shown two separate periods of maximum response for a particular malformation. For example, there are two distinct periods of high frequencies of cleft palate following treatment with either X-radiation, excess vitamin A, or 5-fluorouracil.

The over-all critical period and the time of maximum response for a particular malformation may be different for different agents. Cleft palate is a noteworthy example of a malformation for which the critical period is markedly dependent upon the nature of the teratogen ( Table 14-2). These great differences in critical periods may be explained by assuming that different biochemical and, perhaps, developmental events were being interfered with in separate cases.

The period during which a particular malformation can be induced in an organ is usually of much shorter duration than the critical period for the organ as a whole. For example, polydactylous hind feet were produced by treatment of 10-day embryos with low doses of fluorouracil. Treatment the following day resulted in both polydactyly and oligodactyly, and treatment of 12-day embryos caused oligodactyly only (Dagg, 1960, 1963). Furthermore, as expected, the location of defects in the axial skeleton has varied with the time of exposure. Similar malformations — fusions, deletions, and reductions of vertebral elements — were found in different and relatively localized areas of the spinal column depending on the age of the embryos when subjected to X-radiation or to hypoxia ( Murakami et al., 1961; Murakami and Kameyama, 1963).

Dose-response relationships

The response to a teratogenic agent is dependent upon the intensity and the duration of the stimulus. Typically the range of doses can be divided into a subthreshold, a teratogenic, and a lethal range. The subthreshold range includes all doses not teratogenic according to criteria established by the observer. Obviously, the type of abnormalities that are looked for and the standards used to distinguish normal from abnormal development will vary with the investigator, and an embryo classified as normal in one laboratory might be regarded as defective in another. In the teratogenic range, an increase in intensity or duration of treatment usually is reflected in an increase in frequency and severity of the defect. With some agents, malformations can be produced without appreciably affecting the survival of the embryos. whereas other agents appear to be embryocidal over the entire teratogenic range. Generally, the frequency of embryonal deaths rises progressively as the intensity or duration of the stimulus is increased until all embryos are destroyed. At this point the pregnant dam may herself show signs of being affected.

In order for an agent to be teratogenic it must show some specificity in regard to the tissues affected. The damage produced must be differential so that certain structures can be malformed without, at the same time, destroying the entire conceptus. In some instances the teratogenic range is so narrow that it is difficult to demonstrate that the agent is teratogenic at sublethal doses.

Occasionally, dose-response curves have shown a plateau, so that after a given frequency of malformations was reached, further increases in dosage were not accompanied by higher proportions of deformed embryos. As an example, 200 mg/kg of 5-chlorodeoxyuridine given to A/J and BALB/cJ dams at 10 days after mating produced malformed hind feet in approximately 40 per cent of the fetuses ( Nishimura, 1964). The incidence of malformed hind feet did not change significantly when 400 or 600 mg/kg was given, although the incidence of cleft palates did increase n strain A/J showing that the plateaued response curve was not characteristic of all organs. It is likely that the hind feet at this stage of development were capable of response in only 40 per cent of the A/J and BALB/cJ embryos. A plateaued response curve was not seen in three other strains of mice: C57BL/Ks, C57BL/10Gn, and C57BL/6J.

Similarly, in an experiment designed to test the interaction of four different factors on the cleft palate frequency, Warburton et al. ( 1962) used two dose levels of cortisone, two mouse strains, two maternal weight categories, and two commercial brands of food. With one brand of food and with the heavy mothers an increase in the dose of cortisone greatly increased the frequency of cleft palate in strain A/J (from 26 to 100 per cent), whereas there was little or no increase in strain C57BL (from 52 to 59 per cent).

The schedule of doses can be a factor in determining embryonal responses. Isaacson and Chaudhry ( 1962) gave pregnant A/J mice intramuscular injections of cortisone commencing at 11 ½ days of gestation. Four injections at 6-hour intervals of 0.625 mg for a total of 2.5 mg produced a higher incidence of cleft palates than a single dose of 2.5 mg. The two most common methods of administering drugs are to give amounts in proportion to the mother's body weight and to give the same absolute amount to all animals. An inverse relationship between the maternal weight and the frequency of cleft palate induced by a given dose of cortisone was found by Kalter ( 1956, 1957), and a similar relationship for 6-aminonicotinamide was reported by Pinsky and Fraser ( 1959). Kalter ( 1964) found that when the dosage of cortisone was based on maternal body weight, the influence of weight was largely eliminated. This effect of body weight is not universal. Dagg ( 1963) gave fluorouracil on a weight basis and found that the fetuses in the larger females, receiving proportionately larger amounts of the teratogen, had the higher frequencies of malformed hind feet. In this example the response appeared to be more closely related to the absolute amount of the teratogen than to the mother's body weight.

The response to teratogenic treatments has been shown to vary quantitatively and qualitatively at different times of the year. Kalter ( 1959a) found a seasonal variation in the frequency of cortisone-induced cleft palate in the offspring of (C57BL/6J x A/J)F₁ females crossed to A/J males. When the treatment was given during winter months, November to April, 56 per cent of the offspring were malformed. A lower incidence, 36 per cent cleft palate, was obtained during the summer, may to October. Ingalls et al. ( 1953) observed a seasonal variation in anomalies produced by exposure of pregnant females to 5 hours of hypoxia on the 14th day of gestation. Strain C57BL/6J responded with a higher rate of rib and vertebral malformations during the winter than during the summer. In these examples the variation was quantitative, in contrast to a qualitative variation found by Kalter and Warkany ( 1961) in experiments with vitamin A. Excessive amounts of vitamin A were administered during the early stages of embryonic development to A/J, DBA/1J, and C3H/J strains. Treatment during winter months and during late March and April caused frequent malformations of the palate and pinna and mild and moderate microstomia. However, none of the defects of the central nervous system, anus, and tail, observed during the winter months, appeared in the fetuses treated in late March and April. In all of these examples the teratogens were more effective during winter than during summer.

The nutritional status of the pregnant female would be expected to play an important role in teratogenesis. The influence of two diets, designated A and B, on the teratogenic action of cortisone has been reported by Warburton et al. ( 1962). On diet A the incidence of cleft palate was higher in strain A/J (72 per cent) than in C57BL/6 (6 per cent). On diet B the frequency was lower in strain A/J (26 per cent) than in C57BL/6 (52 per cent). It was concluded that both genetic and nutritional factors influenced the embryos' responses to cortisone, and genetic factors affected the influence of the nutritional factors.

Interaction between agents

Investigators studying the effects of simultaneously applied combinations of teratogenic agents have found four modes of interaction: addition, potentiation, noninteraction, and interference. Examples of additive interaction, in which the combined effects of two agents approximated the sum of the individual effects, were found by Runner and Dagg ( 1960). Strain 129/Rr pregnant females were fasted during the ninth day of gestation and 24 per cent of the embryos developed a syndrome of abnormalities of the ribs and axial skeleton. A similar syndrome was noted in 47 per cent of the embryos whose mothers were kept in a hypoxic atmosphere, in 17 per cent of the embryos from mothers injected with trypan blue, and in 13 per cent of those irradiated with 100 R of X-rays. Combination of fasting with each of the other three treatments had an additive effect: Fasting with hypoxia gave 75 per cent abnormals; with trypan blue, 44 per cent; and with 100 R of X-radiation, 33 per cent.

Other investigators have observed potentiation in which the combined effects of two agents exceeded the sum of their individual effects. Kalter ( 1960) fed a restricted diet, amounting to approximately 40 per cent of the normal intake during the middle third of pregnancy to (C57BL/6J x A/J)F₁ females mated to A/J males and found approximately 6 per cent cleft palates. Intramuscular cortisone injections at 0.5 and 1.0 mg for 4 consecutive days starting at 11 1/3 days of gestation produced 1 per cent and 12 per cent cleft palates, respectively. The combination of the restricted diet and the two separate doses of cortisone gave 37 per cent and 51 per cent cleft palates, clearly more than the expected frequencies of 7 per cent and 18 per cent had the effects been arithmetically additive. Smithberg and Runner ( 1963) also observed potentiation in interaction and discovered differences between strains in this regard. Tolbutamide (a sulfonamide) and nicotinamide together showed potentiated interaction in strain 129/Rr and in strain C57BL/6J but not in strain BALB/cRr. Insulin was potentiated by nicotinamide in strain BALB/cRr but not in strain 129/Rr (C57BL/6 was not tested).

The potentiation of a teratogenic agent by a second agent not itself teratogenic or presented in subteratogenic doses was seen by Woollam and Millen ( 1960a). Thyroxine, given on days 11 and 12 in doses of 0.1 mg, had no deleterious effect on the survival of young, but it considerably reduced the chances of survival of young also exposed to X-radiation. A comparable situation was reported by Dagg and Kallio ( 1962). A teratogen, 5-fluoro-2'-deoxyuridine (FUDR), caused cleft palate, malformed hind feet, and abnormal tails in strain 129/Rr mice when injected 10 days after mating. The natural metabolite, thymidine, was not teratogenic alone, but when it was injected at a dose of 160 mg/kg immediately after FUDR at 20 mg/kg the percentages of surviving embryos with cleft palates and malformed hind feet were markedly increased. The frequency of malformed tails, however, was decreased, indicating that the potentiating interaction had some degree of organ specificity.

Examples of noninteraction of two agents, in which the combined effects were approximately those produced by the single treatment having the larger effect, have been presented by Runner and Dagg ( 1960) who found that iodoacetate and 9-methyl folic acid were nonadditive with 24 hours of fasting in strain 129/Rr.

The type of interaction is not solely a function of the properties of the agents but is also dependent upon the genotype of the treated animals. As noted above, the experiments by Smithberg and Runner ( 1963) with tolbutamide and nicotinamide demonstrated potentiation in strains 129/Rr and C57BL/Ks. However, in strain BALB/cRr the teratogens were nonadditive.

The fourth type of interaction, interference, in which the combined effects are less than those produced by the single treatment having the larger effect has not been observed in mouse embryos. In this regard it is of interest to note that hypoxia will protect the embryo against X-irradiation on day 11 ½ of pregnancy ( Russell et al., 1951; Russell and Russell, 1954). The mice were kept in an atmosphere containing 5 per cent oxygen for 10 minutes before and during the radiation treatment. The hypoxia by itself had no effects on the embryo under these conditions. Combined with irradiation, however, hypoxia provided marked protection against the radiation effects on viability, birth weight, tail length and shape, and forefoot and hind foot structure.

Protective treatments

The damage produced by some teratogens can be prevented by the administration of other agents. Typically, investigations of this sort have either or both of two purposes: to gain information about the modes of action of teratogens and to determine whether different teratogens that produce similar malformations can be counteracted by the same supplemental treatment, thereby providing evidence for similar or dissimilar modes of teratogenic action.

By withholding all food for 24 hours on the ninth day of pregnancy Runner and Miller ( 1956) produced abnormalities in 22 per cent of the embryos of strain 129/Rr. The defects consisted of cranioschisis and vertebral and rib deformities. Certain supplements fed by stomach tube, such a glucose, casein, amino acids, corn oil, and acetoacetate, reduced the frequency to between 2 and 11 per cent ( Runner, 1959).

The offspring of mice injected with 6-amino-nicotinamide displayed a variety of malformations in the axial and appendicular skeletons ( Pinsky and Fraser, 1959). In subsequent tests, matings were made between strains C57BL/6 and A/J and pregnant females were given a single intramuscular injection of 6-aminonicotinamide at approximately 9 ½ or 11 ½ days after the vaginal plug was observed ( Pinsky and Fraser, 1960). If a standard dose of nicotinamide (depending upon the dam's body weight) was given simultaneously with 6-aminonicotinamide, there were no malformations and no increase in number of resorptions. However, when the same dose was given 2 hours after the teratogen, there was a marked increase in frequency of both malformations and resorptions. If twice the standard dose of nicotinamide was given after 2 hours, the frequency of malformations was reduced for treatments at 9 ½ days but not for those at 11 ½ days after mating. On the basis of these and other investigations, it was concluded that the teratogenicity of 6-aminonicotinamide arises from its ability to form a nicotinamide adenine dinucleotide (NAD) analogue that is inactive in some NAD-dependent enzymatic reactions. The differential effects at 9 ½ and 10 ½ days appeared to indicate that the requirements of the maternal-fetal system vary during embryogenesis.

Thymidine, in appropriate amounts, protected strain 129/Rr embryos against fluorodeoxyuridine ( Dagg and Kallio, 1962). At 10 days after mating, 20 mg/kg of fluorodeoxyuridine caused 92 per cent of the fetuses to have malformed tails, 39 per cent malformed hind feet, and 18 per cent cleft palates. Thymidine was given immediately after the teratogen in a series of doses ranging from 2.5 to 640 mg/kg. Tail abnormalities decreased in frequency with 5 to 640 mg/kg of thymidine. The frequency of malformed hind feet dropped to a low level of 4 per cent at 20 mg/kg of thymidine and underwent a reversal, climbing to 88 per cent at 160 mg/kg. The palate was not protected at any dose; instead, in the range of 80 to 640 mg/kg of thymidine, the proportion of embryos with cleft palate was greater than in the controls. The results cannot be assuming a single mode of action of either the drug or the supplement. Fluorodeoxyuridine is partially transformed to fluorouracil and therefore could inhibit both DNA and RNA synthesis in embryonic tissues. The synthesis of one or both of the nucleic acids may be the critical events in the affected organs, the level of teratogen required for effective inhibition varying between organs. Furthermore, low doses of thymidine may bypass the blockade of DNA synthesis created by fluorodeoxyuridine, and thereby protect some organs (tail and feet). At higher doses the amount of thymine formed from thymidine may be sufficient to block the degradation of the fluorouracil formed from fluorodeoxyuridine, thereby prolonging the maintenance of effective levels of the teratogen.

Peer et al. ( 1958) reported that folic acid, but not riboflavin, protected Swiss albino embryos against cortisone-induced cleft palate. Using a cross between C57BL/6J females and A/J males and giving vitamins in food rather than by injection, Kalter ( 1959b) was unable to confirm that folic acid protected against cortisone.

According to Runner ( 1964) X-radiation protected embryonic ribs against damage that would have resulted from fasting, hypoxia, or iodoacetate given alone. Since irradiation (130 R) by itself caused abnormal vertebrae, but practically no abnormal ribs in 27 per cent of the young, it was concluded that fasting, hypoxia, and iodoacetate did not protect against irradiation, but the reverse must be the case. Furthermore, the data showed that in the dual treatments (X-radiation with fasting and X-radiation with iodoacetate) there were additive and potentiated interactions in the production of vertebral abnormalities.

The teratogenic action of X-radiation is reduced or prevented altogether by a variety of supplementary treatments. As mentioned in the section on Interaction between Agents, Russell et al. ( 1951) and Russell and Russell ( 1954) observed a protective effect of hypoxia against radiation damage to the embryo. Woollam and Millen ( 1960b) administered cysteamine (β-mercaptoethylamine) 5 minutes before 300 R of X-radiation and noted a reduction of radiation-induced deformities of the brain and skull from 23 to 2 per cent. Cysteamine had but little protective effect if given 30 minutes after exposure. Rugh and Grupp ( 1960) studied 15 agents for effectiveness in protecting 8 ½ day CF1 mouse embryos against embryonic death and malformations caused by 200 R and found that only cysteamine, cystamine, and hypoxia were beneficial.

GENETIC FACTORS

Without question, genetic factors in both embryo and pregnant female play major roles in determining the type, frequency, and the severity of defects that develop spontaneously or through experimental intervention.

Spontaneous deformities

Whenever sufficiently large numbers of mice of either inbred or noninbred strains are examined, some individuals show such marked departures from morphological norms for their respective strains that they are classifiable as abnormal. Each strain of mice has characteristic, relatively uniform frequencies of one or more spontaneous defects. The contributors to Inbred Strains of Mice, No. 3 ( 1963) reported some of the characteristic spontaneous deviants and their frequencies. These include: 10 to 15 per cent cleft lip with or without cleft palate in sublines of the A strains, 8 to 20 per cent microphthalmia and anophthalmia in strains C57BL/6 and C57BL/10, 4 per cent open eyelids in A/JFr, 1 to 3 per cent hydrocephalus in C57BL/6J and C57BL/10J, and 5 per cent kinked or bent tails in LG/M1.

Spontaneous defects may occur more frequently in one sex than the other. According to Dickie (Inbred Strains of Mice, No. 3) 20 per cent of the females and 3 per cent of the males of C57BL/10J have microphthalmia and anophthalmia. Spontaneous cleft lip-cleft palate was found slightly more often in females (20 per cent) than in males (13 per cent) of strain A/HeJ (Dagg, unpublished).

A few of these deviants are the consequence of mutations, but most are environmental rather than genetic in origin. The absence of genetic differences between normal and aberrant animals of the same strain can be established from the lack of parent-offspring correlations when normal and deviant phenotypes are mated. In such cases, crosses between normal individuals produce as many phenodeviants as do crosses between phenodeviants. Although genetic differences are not responsible for most morphological variations between strains in frequencies and types of phenodeviants are strongly influenced by genetic factors. The evidence supports the interpretation that sporadic variants are a result of interaction between numerous genetic loci and environmental stresses. The combined effect of the several loci serves to position the population close to a threshold for development of a particular abnormality, and a less-than-optimal environment is a sufficient stimulus to complete the conditions for atypical development. The nature of the environmental stimulus or stimuli are not known, nor is it known whether the stimuli are identical for all similarly affected individuals in a single litter or in different litters. The effective environmental stimulus may be external to the females in some cases: for example, background radiation from all sources. Viral or bacterial infections and improperly balanced or otherwise inadequate diets may be sufficient stimuli. Inconstant poorly defined physiological states, for example changing physiology associated with aging of the female, may be important. The environmental stimulus may be localized within the uterus. Differences in blood supply to embryos, the effects of crowding, the proximity of dead and resorbing embryos, temperature, and many other factors may determine that some embryos will fail to conform to the strain-typical morphological pattern. Some of these localized uterine differences may be secondary consequences of the systematic physiological changes in the female.

The spontaneous cleft lip-cleft palate that develops with predictable regularity in various sublines of strain A mice is genetically determined but strongly influenced by environmental factors ( Grüneberg, 1952). From a study of inbred lines in which there was approximate homozygosis of genetic factors including those for harelip, Reed ( 1936b) concluded that harelip resulted either when a small number of cumulative genes were present in homozygous condition and when environmental conditions were favorable, or when a single gene in homozygous condition was present in association with several modifying genes and appropriate environmental factors. Among the factors affecting the development of harelip were sex of the individual, litter size, and age of the mother. As much as 76 per cent of the total variation was attributed to unknown environmental factors ( Reed, 1936a).

Trasler ( 1960) has found a possible clue to the nature of one of the uterine factors influencing development of spontaneous cleft lip in strain A/J. Embryos in the uterine site nearest the ovary developed cleft lip (with or without cleft palate) significantly more often than embryos in other positions in the uterus. The ovarian site was less favorable whether there were many or few embryos in the horn, and therefore the difference was probably due to an undetermined inherent quality of that area and not to crowding and competition for nutrients.

Woollam and Millen ( 1960a) reported a significant decrease in the incidence of cleft lip-cleft palate in the A/St strain following injection of the dam with thyroxine on days 11 and 12 of pregnancy. The average number of young per litter was not affected by the treatment. Cortisone injections, also, reduced the percentage of embryos with spontaneous cleft lip with cleft palate in the A/J strain ( Walker and Crain, 1959). The rate of embryonal mortality increased in treated litters, and therefore the lowered frequency of cleft lip-cleft palate embryos was interpreted as a consequence of a differential lethal effect of cortisone on defective embryos rather than a healing effect. Hypervitaminosis A which, like cortisone, produced cleft palate without cleft lip did not alter the incidence of cleft lip-cleft palate, although excessive vitamin A reduced the litter size about as much as did cortisone. Similarly, a riboflavin-deficient galactoflavin-containing diet caused a doubling of the normal resorption rate without changing the frequency of spontaneous cleft lip-cleft palate in strain A/J ( Kalter and Warkany, 1957). This diet did affect the incidence of four other deformities that occurred "spontaneously" in the controls. Increases were found in the frequencies of open eyes in strain A/J and cleft palate, miscellaneous eye defects, and brachygnathia in strain C57BL/6J. An increase in the percentage of harelip mice following treatment of the dam during the first half of pregnancy with an anterior pituitary preparation, Preloban, was reported by Steiniger ( 1940).

Hyperphalangy or polydactyly on the preaxial side of the hind foot occurs with a low incidence in all of the C57BL strains and sublines. In C57BL/10Dg the frequency is approximately 0.05 per cent and in C57BL/Ks it is 1.5 to 3 per cent ( Dagg, 1963). The difference between polydactylous and nonpolydactylous animals of strain C57BL/Ks is not genetic.

When 10-day exposures of strain C57BL/Ks were exposed to X-radiation, hyperphalangous hind feet were found in a large number of the fetuses ( Dagg, 1964). The dose-response curve over the range of 75 to 150 R did not show evidence of a threshold dose for polydactyly. Definite thresholds were obtained for two other strains: 129/Rr and BALB/cGn. These results were explained by assuming that genetically determined predispositions and uterine environmental factors interacted on the 10th day of gestation to produce spontaneous hyperphalangy in 1.5 per cent of the controls. Furthermore it was assumed that the entire population of embryos was distributed continuously with regard to this tendency toward abnormality, and therefore the normal control embryos as a group lay just below a threshold for spontaneous hyperphalangy. It was also assumed that X-radiation acted directly upon the foot plate and thereby, even at very low doses, caused an additional small number of embryos to cross the threshold of abnormality. As a teratogen, X-radiation may be unique in not showing clearly defined threshold doses under certain circumstances, because it can act directly upon the fetus. Chemical agents would be expected to show threshold because their actions are subject to inhibition by naturally occurring metabolites in the mother and fetus, and because they are subject to detoxification and excretion by the pregnant dam.

It follows from the assumptions on the mechanism of interaction of X-radiation and genetically determined tendencies toward spontaneous hyperphalangy that a crucial factor in the interaction is the relationship between the time of gene action on the affected organ and the critical period for the experimentally produced malformations of the same type. How closely these two events should correspond is not known. However, it would be expected that an increase in frequency of spontaneous deformities would not occur in some circumstances because the teratogen was not effective at the proper time. For some deformities the genetic and the teratogenic events need no take place within a short span of time, especially if those deformities can be induced over a relatively long period of development. An example is cleft palate.

Differences between strains

In virtually all teratogenic experiments in which different strains were compared, the frequencies of particular defects and syndromes of defect, that is, the total arrays of defects and their frequencies, were found to vary between strains. These differences between strains are generally regarded as genetic in origin with multiple genetic factors, rather than single genes, as the basis. However, the set of multiple factors responsible for differences between strains in response to one teratogen may not be identical to the set responsible for differences in response to another. Evidence for this conclusion is found in the observations that the relative order of strains in terms of frequencies of a particular malformation and interstrain differences in syndromes depend upon the nature of the teratogens. Furthermore, the set of multiple factors involved in the production of malformations in one organ may not be identical with those that induce maldevelopment in other organs of the same embryo.

The effects on the axial skeleton produced by fasting pregnant females throughout the ninth day of pregnancy are not strain-specific ( Miller, 1962). The same array of defects was found in strain 129/Rr, C57BL/6J, and LG/Rr. The frequency of malformations was higher in strain 129/Rr than in C57BL/6J, and the maximum frequency occurred in strain 129/Rr a day later than in C57BL/6J.

Smithberg and Runner ( 1963) determined the relative responses of three strains to the hypoglycemia-producing sulfonamide, tolbutamide. The axial skeleton was more susceptible in strain 129/Rr than in BALB/cRr, which in turn was more susceptible than C57BL/6J. Insulin was much more teratogenic in strain 129/Rr, with 62 per cent of the fetuses malformed, than in strain BALB/cRr, with 3 per cent malformed.

Dagg ( 1963) treated strains 129/Rr, C57BL/6J, and BALB/cRr with 5-fluorouracil. When strains were compared with respect to the frequency of malformed hind feet, strain C57BL/6J was the most susceptible, BALB/cRr embryos were the most resistant, and strain 129/Rr was intermediate. Thus, the relative order of susceptibility of strains 129/Rr and C57BL/6J was the same for fasting as for tolbutamide, but the relative order of the three strains — 129/Rr, C57BL/6J, and BALB/cRr — was not the same for tolbutamide and fluorouracil.

Ingalls et al. ( 1953) compared responses of several strains to temporary hypoxia on the ninth day of gestation and found the relative order of sensitivity to the induction of malformed vertebrae and ribs to be, from most to least susceptible: A/J (75 per cent), C57BL/6J, DBA/1J, C57BR/cdJ, and BALB/cSc (17 per cent). Sternal deformities were in a different order: DBA/1J (74 per cent), C57BR/cdJ, BALB/cSc, A/J, and C57BL/6J (23 per cent).

Kalter and Warkany ( 1957) subjected strains A/J, DBA/1J, 129/J, and C57BL/6J to a riboflavin-deficient galactoflavin-containing diet for 4 days starting and 9 1/3 or 10 1/3 days after mating. Skeletal malformations were found in the skull, vertebrae, ribs, and appendages of fetuses. At a low dose of galactoflavin the DBA/1J and 129/J fetuses displayed the entire syndrome of abnormalities. At the same dose level C57Bl/6J fetuses showed no deformities and the A/J fetuses only a few. A higher dose level did not produce more malformations in DBA/1J or 129/J, but C57BL/6J fetuses showed moderate frequencies of certain abnormalities, and the A/J fetuses developed the entire syndrome. Each strain presented its own pattern of defects. For example, strain 129/J had the highest frequency of digital defects, A/J had the highest incidence of open eye, and DBA/1J had proportionately more fetuses with cleft palate and micromelia.

Kalter and Warkany ( 1957) compared the frequencies of cleft palate produced by riboflavin deficiency with those produced by cortisone in the same strains ( Kalter, 1954). Cortisone caused cleft palate in 100 per cent of the A/J fetuses, in 75 per cent of DBA/1J, and in 20 per cent of C57BL/6J. Riboflavin deficiency caused cleft palate in 3 per cent of A/J fetuses, in 41 per cent of DBA/1J, and in 13 per cent of C57BL/6J. The palates of strain A/J embryos were also more susceptible than those of C57BL to excessive vitamin A ( Walker and Crain, 1960). Ninety-six per cent of the A/J fetuses and 57 per cent of C57BL fetuses from treated dams had cleft palate. Furthermore, according to a system of rating the stages of palatal closure ( Walker and Fraser, 1956), the palates of A/J fetuses were the more seriously affected.

Since the incidence of a particular malformations is dependent upon the developmental age of the embryo when exposed to a teratogen, some interstrain differences in frequencies of malformations may be a consequence of dissimilarities in developmental ages in contrast to chronological (postconceptual) ages at the time of treatment. With some teratogens a change in time of treatment by 12 hours produces marked changes in frequencies of particular malformations (Dagg, 1963, 1964).

Walker and Crain ( 1960) devised a system of grading embryos on the basis of morphological criteria. Developmental stages of forefeet, hind feet, ears, hair follicles, and eyes are assigned numerical values. The morphological rating for an embryo is calculated by adding the numerical values for each of the five developmental stages. Photographs to assist in classifying embryos according to these developmental features have been published by Trasler ( 1964).

The crown-rump length of C57BL/Ks, 129/Rr, and BALB/cGn embryos has been used as a measure of relative developmental ages (Dagg, 1963, 1964). In these embryos, as in the series described by Grüneberg ( 1943), the crown-rump length is correlated with gross changes in external features of the embryos, such as foot pads, branchial arches, and length of tail. The C57BL/Ks embryos were found to be about 12 ½ hours more advanced that the BALB/cGn and 129/Rr embryos. The (C57BL/Ks female x BALB/cGn male) hybrid embryos were equal in length to homozygous C57BL/Ks embryos, whereas the (BALB/cGn female x C57BL/Ks male) hybrid embryos were intermediate in size relative to parental strain embryos.

There is considerable variation between and within litters in crown-rump lengths ( Dagg, 1964). The largest embryos of strain BALB/cGn examined at 10 days after conception were equal in size to the smallest 11-day embryos. If the difference between average crown-rump lengths of 10- and 11-day embryos was used as an estimate of 1 day's development, then the largest and the smallest 10-day embryos differed by 1 day of development, and in one litter the differences in crown-rump lengths corresponded to 1 day's development.

Inheritance of susceptibility

The first informative and conclusive investigation of the inheritance of interstrain differences in response to a teratogen was made by Kalter. The results have been described in detail (Kalter, 1954, 1957, 1964) and will be summarized briefly. Four daily intramuscular injections of cortisone beginning about the 11th day of pregnancy produced cleft palate in 100 per cent of A/J embryos, in 19 per cent of C57BL, and in 12 per cent of CBA. Reciprocal crosses were made between CBA and A/J. The F₁ embryos, identical in genotype except for the sex chromosomes, had nearly equal frequencies of cleft palates which were intermediate relative to the parental frequencies. In the reciprocal crosses between C57BL and A/J the F₁ embryos had different frequencies of deformity: 44 per cent in the A/J mothers and 3 per cent in the C57BL mothers. Since, again, the embryos were genetically uniform, the differences between the two types of F₁ embryos must have been influenced by maternal factors. To determine whether this maternal effect was nuclear or cytoplasmic in origin, F₁ mothers from the mating of (A/J female x C57BL male) or the reciprocal cross were mated to A/J males. The nuclear genotypes of these two types of F₁ mothers were identical, but the egg cytoplasm came exclusively from one or the other strain. These backcross embryos had equal responses in both types of mothers and, therefore, the maternal effect observed in the F₁ embryos was probably not die to cytoplasmic factors transmitted in the egg. Since the (A female x A male) embryos and the (A female x C57BL male) embryos were growing in the same type of mothers and since these embryos responded differently to cortisone, genetic factors operating in the embryo must also have played a role in determining the response to teratogens.

A study of strain differences in reaction to cortisone ( Loevy, 1963) also demonstrated genetic influences on the response to a teratogen and, more interestingly, provided evidence of a patroclinous effect; that is, the responses of reciprocal F₁ embryos were different and each type of F₁ embryo tended to be more like the paternal rather than the maternal strain. Cortisone treatment induced cleft palate in all of the embryos of the A/St strain and 36 per cent of the C3H embryos. The hybrid embryos produced by mating C3H females with A/St males had a higher frequency of cleft palates than did the embryos of the reciprocal cross. Figures published by Kalter ( 1964), who used the same strains but a higher dose of cortisone, corroborate these findings.

Fluorouracil, given at 10 days after mating, caused hind foot deformities, primarily hyperphalangy, more often in strain 129/Rr than in BALB/cGn ( Dagg, 1963). The amount of drug injected was based on maternal weight at the time of treatment. The frequency of malformed hind feet was significantly higher in (BALB/c x 129) F₁ embryos developing in BALB/c mothers than in those developing in strain 129/Rr mothers. by itself, this finding could have been interpreted as evidence of a patroclinous effect, but an inspection of the data revealed that the BALB/c females were heavier than 129/Rr females, and therefore, received proportionately more of the drug. When comparisons were made between groups with the same ranges in weight, the patroclinous effect disappeared.

The work of Goldstein et al. ( 1963) supports the idea that embryonal genes influencing responses to teratogens may act by producing variations in response of specific organ systems rather than, or in addition to, altering the embryo's general susceptibility to a given teratogen. Vertebral fusions produced by 6-aminonicotinamide occurred more often in A/J embryos (89 per cent) than in C57BL/6J (56 per cent). The F₁ embryos sired by A/J fathers had a higher incidence of vertebral fusions than those sired by C57BL/6J (67 per cent vs. 45 per cent), a patroclinous reciprocal cross difference. Several alternative explanations for the patroclinous difference were offered: (1) random variation; (2) a factor for resistance transmitted in the sperm but not in the egg; (3) a maternal uterine or cytoplasmic factor for resistance transmitted in the A/J strain, interacting with an embryonal genetic factor for resistance transmitted in the C57BL/6J, thereby resulting in F₁ embryos which are genetically intermediate but, because of maternal factors for resistance, the embryos developing in A/J mothers had a lower frequency of vertebral fusions than did embryos developing in C57BL/6J mothers; and (4) differences in rates of development of the two types of hybrid, so that F₁ embryos in C57BL/6J mothers were treated at the point of maximum vertebral sensitivity, whereas embryos in A/J mothers had developed faster and were treated at a point of lower sensitivity. In the latter case the rate of development would be maternally rather than paternally determined.

The frequency of cleft palate produced by 6-aminonicotinamide was higher in strain A/J embryos (76 per cent) than in C57Bl/6J (11 per cent). In the reciprocal crosses the response was greater in F₁ embryos developing in C57BL/6J mothers (4 per cent), a matroclinous reciprocal cross difference. Since the reciprocal cross differences were in opposite directions for vertebral fusions and cleft palates, the strain differences were regarded as organ-specific. To determine whether the maternal effect was a reflection of maternally determined differences in the embryos' environments or was due to differences in factors transmitted through the egg cytoplasm, the two types of F₁ mothers (A/J female x C57BL/6 male and C57BL/6 female x A/J male) were backcrossed to A/J males and treated with 6-aminonicotinamide. The frequency of deformed embryos in (A/J female x C57BL/6J male) F₁ mothers was higher than the frequency in F₁ mothers produced by the reciprocal cross (24 per cent vs. 6 per cent). These results were interpreted as suggesting that the teratogenic response to 6-aminonicotinamide is influenced by factors transmitted through the egg cytoplasm.

Hybrid embryos treated with X-radiation tend to resemble the more resistant parental strain in frequency of abnormalities. Rugh et al. ( 1961) irradiated embryos of strains C57BL/6, CF1, and the (CF1 female x C57BL/6 male) F₁ hybrids. The percentages of total implantations that developed into normal fetuses were 5 for C57BL/6, 38 for CF1, and 44 for the hybrids. The exposure of 10-day embryos to 150 R of X-rays caused polydactylous hind feet in 38 per cent of strain C57BL/Ks and 5 to 8 per cent in strain BALB/cGn ( Dagg, 1964). The F₁ embryos from reciprocal crosses were like the resistant strain, showing abnormalities in 3 to 9 per cent.

A provisional estimate has been made of the minimum number of gene pairs involved in differences between strains 129/Rr and BALB/cGn in response to fluorouracil, with the conclusion that at least two pairs of genes would be required to account for the observed differences ( Dagg, 1963). The estimate was made from F₁ embryos and from two successive backcrosses to the BALB/c strain and was based on several assumptions: (1) Alleles determining susceptibility were dominant to those for resistance; (2) if more than one pair of genes were involved, they segregated independently and had nearly equal effects; and (3) the genes were effective only in the fetus or entire conceptus and the maternal genotype was of no consequence. It should be noted that the assumption of dominance by susceptibility-determining genes was based on a similar frequency of malformed hind feet in F₁ embryos and homozygous 129/Rr embryos. This assumption may not be valid because the F₁ embryos appeared to be more advanced in development than 129/Rr embryos, and because 129/Rr embryos were more susceptible at more advanced stages of development (10 ½ vs. 10 days of gestation). If dominance is absent and if the alleles have additive effects, the minimum number of loci would be greater than two.

Since responses to teratogens are at least partially determined by the genotype, as shown by the investigations of interstrain differences in susceptibility to teratogens, it would be expected that under certain circumstances teratogens would affect the phenotypic expression of single genes which by themselves produce morphogenetic changes.

A reduction in penetrance of a mutant gene following maternal treatment with a teratogenic agent has been reported by Watney and Miller ( 1964). The mutation, identified as a recurrence of lid gap ( lg) in C3H/M1, caused open eyelids at birth in 77 per cent of the offspring from lg/ lg matings. Normally the eyelids close on approximately the 17th day of gestation and do not reopen until 2 weeks after birth. A single low dose of cortisone administered intramuscularly on the 15th day of gestation to lg/ lg females, mated to lg/ lg males, completely inhibited the penetrance of the gene so that all newborn mice had closed eyes. This low dose of cortisone did not increase neonatal mortality nor did it cause cleft palates, whereas higher doses or multiple doses produced both of these effects.

Barber ( 1957) reported an effect of trypan blue on embryos heterozygous for the eyeless ( ey) genes. The number of gene pairs, one or two, involved in the inheritance of anophthalmia appears to be undecided ( Chapter 8). Barber mated anophthalmic mice of one colony to normal-eyed mice of another and treated the pregnant dams with trypan blue on the seventh, eighth, and ninth days of gestation. Anophthalmia was found in 15 of 40 fetuses and newborn animals from these mothers. The same treatment regimen had no effect on fetuses from matings of normal-eyed parents, and none of the fetuses from untreated dams in matings between anophthalmic and normal mice had anophthalmia. It was concluded that maternal treatment with trypan blue resulted in the expression of a recessive congenital trait that would otherwise be suppressed. Subsequently, Barber et al. ( 1959) reported that injections of cortisone into pregnant dams on days 7, 8, and 9 of gestation had no effect on the size of the eyes in normal embryos nor in embryos heterozygous for the eyeless genes.

Beck ( 1963) confirmed the findings made by Barber ( 1957) on the effect of trypan blue on embryos carrying the eyeless gene and suggested that a single dose of the mutant gene acted to bring the embryo closer to a threshold for reduced eye size that is the case in a homozygous normal individual. Beck ( 1964) demonstrated that genes other than the eyeless gene also influence the frequency of anophthalmia in the offspring of mothers treated with trypan blue. In the untreated control groups, 2.5 per cent of the fetuses from C57BL/10 females were microcephalic or anophthalmic, and 7 per cent of the fetuses from C57BL/6 females mated to ZRDCT ey/ ey males had these eye defects. The difference in incidence of defect in the two types of mothers was not significant, When C57BL/10 females, mated to ZRDCT ey/ ey males, were treated with trypan blue, 5 per cent of fetuses had abnormally small eyes, indicating no increase in frequency over the control level. In contrast, 36 per cent of the fetuses from C57BL/6 females mated to ZRDCT ey/ ey males and treated with the dye had anophthalmia or microphthalmia.

Teratogenic agents may increase the penetrance and expressivity of mutant genes that cause gross skeletal malformations of the limbs. The autosomal genes luxate ( lx) and luxoid ( lu) produce a variety of skeletal changes including preaxial hyperphalangy of the hind feet in heterozygotes and tibial hemimelia in homozygotes ( Carter, 1951; Green, 1955). In strain C57BL/10 the penetrance of lx/+ is 0.68, and for lu/+ the penetrance is 0.85 ( Forsthoefel, 1958). In the same strain, 0.25 mg of 5-fluorouracil injected into pregnant females 10 days after conception produced hyperphalangous hind feet in 42 per cent of the fetuses ( Dagg, 1965). At a higher dose, 0.50 mg, the incidence of hyperphalangy increased and some of the fetuses showed tibial hemimelia. To test the combined effects of the mutants in the teratogen, C57BL/10- lu/+ and - lx/+ males were mated to C57BL/10-+/+ females and the females were injected with 0.25 mg of fluorouracil. The pregnant dams, therefore, were carrying either +/+ and lu/+ embryos or +/+ and lx/+ embryos. Hyperphalangous hind feet were found in 68 per cent of the fetuses sired by lu/+ males and in 77 per cent of those sired by lx/+ males. In addition, many fetuses had tibial hemimelia, a characteristic of homozygotes or embryos treated with the higher dosage. The results were interpreted as demonstrating that fluorouracil increased the penetrance of lu and lx in heterozygotes and, furthermore, caused some of the heterozygotes to manifest the homozygous phenotype.

Kobozieff and Pomriaskinsky-Kobozieff ( 1953, 1962) have described a mutant resembling luxoid ( lu) in almost all essential features. The heterozygotes are polydactylous and the homozygotes show tibial hemimelia and luxation. The penetrance varies on different genetic backgrounds, for example, in crosses between strain MO females and heterozygous polydactylous males only 7 per cent of the offspring are polydactylous instead of the expected 50 per cent. The effect of trypan blue on the penetrance and expressivity of this mutant was studied by injecting the dye on the eighth day of gestation into MO females that had been mated to MO males or heterozygous polydactylous males ( Kobozieff et al., 1959). Neither the penetrance nor the expressivity was changed by the treatment. Other malformations, including exencephaly, occurred with equal frequencies in both types of matings.

Pennycuik ( 1965) investigated the effects of acute exposure to high temperatures on prenatal development of mice carrying the sex-linked gene tabby ( Ta). Normal females (+/+) of an outbred stock called TS were mated to males carrying tabby ( Ta/Y). Pregnant females, therefore, contained tabby ( Ta/+) female fetuses and normal (+/Y) male fetuses. At different gestational ages, from 1 to 18 days, the dams were exposed to 43°C for 1 hour. Foot malformations were produced by treatment at 11 days, but mice with deformed feet were distributed almost equally between the sexes, so that it was unlikely that the tabby gene influenced the sensitivity of the feet to heat damage. One of the effects of the tabby gene is to reduce the number of secondary vibrissae. The effect of heat on vibrissa development was greater in mice carrying the tabby gene than in normal sibs, and the effects of heat and the tabby gene were found to be additive.

SUMMARY

The production of developmental abnormalities by teratogenic agents is a complex event involving the interplay of multiple environmental and genetic factors. The most important experimental variables are the nature of the agent, the developmental stage at which it is applied, and the dose or intensity of treatment. Some of the factors that may modify the embryonal response are diet, season of the year, other teratogens, protective agents, maternal weight and age, position of the embryo in the uterus, and the physiological condition of the mother. In a few cases, cytoplasmic factors passed through the egg may exert an effect on the response to a teratogen. More commonly, genes acting in the embryo and the mother determine and modify the responses to teratogens. Differences between strains of mice, whether they are reflections of maternal or embryonal genotypes or both, are polygenic in origin. Different sets of genes may be implicated in responses to different types of teratogens, and different sets of genes may be operating in the responses of different embryonal organs. in a few experimentally devised situations, the interaction of single genes and teratogens has been studied.

¹The writing of this chapter was supported in part by Public Health Service Grant HD 00473 from the National Institute of Child Health and Human Development.

LITERATURE CITED

Barber, A.N. 1957. The effects of maternal hypoxia on inheritance of recessive blindness in mice. Amer. J. Ophthalmol. 44: 94-101.
See also MGI.

Barber, A.N., C. Afeman, and J. Willis. 1959. Inheritance of congenital anophthalmia in mice. II. Effects of cortisone and maternal immunization with brain. Amer. J. Ophthalmol. 48: 763-769.
See also PubMed.

Beck, S.L. 1963. Frequencies of teratologies among homozygous normal mice compared with those heterozygous for anophthalmia. Nature 200: 810-811.
See also MGI.

Beck, S.L. 1964. Sub-line differences among C57 black mice in response to trypan blue and outcross. Nature 204: 403-404.
See also MGI.

Carter, T.C. 1951. The genetics of luxate mice. I. Morphological abnormalities of heterozygotes and homozygotes. J. Genet. 50: 277-299.
See also MGI.

Dagg, C.P. 1960. Sensitive stages for the production of developmental abnormalities in mice with 5-fluorouracil. Amer. J. Anat. 106: 89-96.
See also PubMed.

Dagg, C.P. 1963. The interaction of environmental stimuli and inherited susceptibility to congenital deformity. Amer Zool. 3: 223-233.
See also MGI.

Dagg, C.P. 1964. Some effects of X-irradiation on the development of inbred and hybrid mouse embryos, p. 91-102. In W.D. Carlson and F.X. Gassner [ed.] Effects of Ionizing Radiation on the Reproductive System. Pergamon Press, New York.

Dagg, C.P. 1965. Effects of fluorouracil on the penetrance of two skeletal mutants in mice. Anat. Rec. 151: 341. (Abstr.)

Dagg, C.P., and E. Kallio. 1962. Teratogenic interaction of fluorodeoxyuridine and thymidine. Anat. Rec. 142: 301-302. (Abstr.)

DiPaolo, J.A. 1963. Congenital malformations in strain A mice. J. Amer. Med. Ass. 183: 139-141.
See also PubMed.

DiPaolo, J.A. 1964. Polydactylism in the offspring of mice injected with 5-bromodeoxyuridine. Science 145: 501-502.

European Society for the Study of Drug Toxicity. 1963. Effects of Drugs on the Foetus. Proc. Vol. I. Excerpta Medica Foundation, Amsterdam. 58 p.

Fishbein, M. [ed.] 1960. First International Conference on Congenital Malformations. Lippincott, Philadelphia. 314 p.

Fishbein, M. [ed.] 1963. Second International Conference on Congenital Malformations. International Medical Congresses, New York. 442 p.

Forsthoefel, P.F. 1958. The skeletal effects of the luxoid gene in the mouse, including its interactions with the luxate gene. J. Morphol. 102: 247-288.
See also MGI.

Fraser, F.C., and T.D. Fainstat. 1951. Production of congenital defects in the offspring of mice treated with cortisone; progress report. Pediatrics 8: 527-533.
See also PubMed.

Giroud, A., and M. Martinet. 1960. Action tératogène de l'hypervitaminose A chez la souris en fonction du stade embryonnaire. Compt. Rend. Soc. Biol. 154: 1353-1355.
See also PubMed.

Giroud, A., and H. Tuchmann-Duplessis. 1962. Malformations congénitales. Roles des facteurs exogènes. Pathol.-Biol. 10: 119-151.
See also PubMed.

Goldstein, M., M.F. Pinsky, and F.C. Fraser. 1963. Genetically determined organ specific responses to the teratogenic action of 6-aminonicotinamide in the mouse. Genet. Res. 4: 258-265.

Green, M.C. 1955. Luxoid — a new hereditary leg and foot abnormality in the house mouse. J. Hered. 46: 91-99.
See also MGI.

Grüneberg, H. 1943. The development of some external features in mouse embryos. J. Hered. 34: 89-92.

Grüneberg, H. 1952. The Genetics of the Mouse, 2nd ed. Nijhoff, The Hague. 650 p.

Hamburgh, M. 1952. Malformations in mouse embryos induced by trypan blue. Nature 169: 27.
See also PubMed.

Heiberg, K., H. Kalter, and F.C. Fraser. 1959. Production of cleft palates in the offspring of mice treated with ACTH during pregnancy. Biol. Neonat. 1: 33-37.
See also PubMed.

Heitz, F., and M. Martinet. 1961. Dualité des stades sensibles dans le développement du palais chez la souris mise en évidence par les rayons X. Compt. Rend. Soc. Biol. 155: 707-709.
See also PubMed.

Inbred Strains of Mice, No. 3. 1963. The Jackson Laboratory, Bar Harbor, Maine. 104 p.

Ingalls, T.H., F.R. Avis, F.J. Curley, and H.M. Temin. 1953. Genetic determinants of hypoxia-induced congenital abnormalities. J. Hered. 44: 185-194.
See also MGI.

Ingalls, T.H., and F.J. Curley. 1957. The relation of hydrocortisone injections to cleft palate in mice. New Engl. J. Med. 256: 1035-1039.
See also PubMed.

Ingalls, T.H., F.J. Curley, and R.A. Prindle. 1952. Experimental production of congenital anomalies; timing and degree of anoxia as factors causing fetal deaths and congenital anomalies in the mouse. New Engl. J. Med. 247: 758-768.
See also PubMed.

Ingalls, T.H., E.F. Ingento, and F.J. Curley. 1964. Acquired chromosomal anomalies induced in mice by a known teratogen. J. Amer. Med. Ass. 187: 836-838.
See also PubMed.

Isaacson, R.J., and A.P. Chaudhry. 1962. Cleft palate induction in strain A mice with cortisone. Anat. Rec. 142: 479-484.

Kageyama, M. 1961a. Differences in susceptibility to induction of congenital malformations by ethylurethan among various strains of mice. Acta Anat. Nippon. 36: 1-9.

Kageyama, M. 1961b. Disturbances in the skeletal development of mouse embryos induced by injection of triethylene melamine (TEM) during pregnancy. Acta Anat. Japon. 36: 246-255.

Kageyama, M., and H. Nishimura. 1961. Developmental anomalies in mouse embryos induced by triethylene melamine (TEM). Acta Scholae Med. Univ. Kioto 37: 318-327.
See also PubMed.

Kalter, H. 1954. The inheritance of susceptibility to the teratogenic action of cortisone in mice. Genetics 39: 185-196.
See also PubMed.

Kalter, H. 1956. Modification of teratogenic action of cortisone in mice by maternal age, maternal weight, and litter size. Amer. J. Physiol. 185: 65-68.
See also PubMed.

Kalter, H. 1957. Factors influencing the frequency of cortisone induced cleft palate in mice. J. Exp. Zool. 134: 449-467.
See also PubMed.

Kalter, H. 1959a. Seasonal variation in frequency of cortisone-induced cleft palate in mice. Genetics 44: 518-519. (Abstr.)

Kalter, H. 1959b. Attempts to modify the frequency of cortisone-induced cleft palate in mice by vitamin, carbohydrate, and protein supplementation. Plast. Reconstr. Surg. 24: 498-504.

Kalter, H. 1960. Teratogenic action of a hypocaloric diet and small doses of cortisone. Proc. Soc. Exp. Biol. Med. 104: 518-520.
See also PubMed.

Kalter, H. 1963. Congenital malformations of the central nervous system. Amer. J. Clin. Nutrit. 12: 264-274.
See also PubMed.

Kalter, H. 1964. Interplay of intrinsic and extrinsic factors, p. 57-80. In J.G. Wilson and J. Warkany [ed.] Teratology: Principles and Techniques. The University of Chicago Press, Chicago.

Kalter, H., and J. Warkany. 1957. Congenital malformations in inbred strains of mice induced by riboflavin-deficient, galactoflavin-containing diets. J. Exp. Zool. 136: 531-566.
See also PubMed.

Kalter, H., and J. Warkany. 1959. Experimental production of congenital malformations in mammals by metabolic procedures. Physiol. Rev. 39: 69-115.
See also PubMed.

Kalter, H., and J. Warkany. 1961. Experimental production of congenital malformations in strains of inbred mice by maternal treatments with hypervitaminosis A. Amer. J. Pathol. 38: 1-21.
See also PubMed.

Kobozieff, N., and N.-A. Pomriaskinsky-Kobozieff. 1953. Recherches sur la constitution génotypique des souris luxées et polydactyles. Compt. Rend. Soc. Biol. 147: 196-199.
See also PubMed.

Kobozieff, N., and N.A. Pomriaskinsky-Kobozieff. 1962. Hémimélie chez la souris. II.Étude morphologique des homozygotes atteints de différentes anomalies du squelette. C.-membres posterieurs: polydactylie intégral et ceinture pelvienne. Rec. Méd. Vét. 138: 485-505.
See also MGI.

Kobozieff, N., H. Tuchmann-Duplessis, L. Mercier-Parot, and N.-A. Pomriaskinsky-Kobozieff. 1959. Influence du bleu trypan sur la fréquence d'apparition et la gravité des malformations chez des souris présentant une polydactylie héréditaire. Rec. Méd. Vé. 135: 317-324.

Loevy, H. 1963. Genetic influences on induced cleft palate in different strains of mice. Anat. Rec. 145: 117-122.
See also PubMed.

Miller, J.R. 1962. A strain difference in response to the teratogenic effect of maternal fasting in the house mouse. Can. J. Genet. Cytol. 4: 69-78.
See also PubMed.

Murakami, U., and Y. Kameyama. 1963. Vertebral malformation in the mouse foetus caused by maternal hypoxia during eraly stages of pregnancy. J. Embryol. Exp. Morphol. 11: 107-118.

Murakami, U., Y. Kameyama, A. Majima, and T. Sakurai. 1961. Patterns of radiation malformations in the mouse fetus and subjected stage of development. Annu. Rep. Res. Inst. Environ. Med., Nagoya Univ. 9: 71-81.
See also PubMed.

Nishimura, H. 1964. Chemistry and Prevention of Congenital Anomalies. Charles C. Thomas, Springfield, Ill. 119 p.

Nishimura, H., M. Kageyama, and K. Hayashi. 1962. Teratogenic effect of the methionine derivatives upon the mouse embryos. Acta Scholae Med. Univ. Kioto 38: 193-197.
See also PubMed.

Nishimura, H., and M. Kuginuki. 1958. Congenital malformations induced by ethyl-urethan in mouse embryos. Okajimas Folia Anat. Japon. 31: 1-13.
See also PubMed.

Nishimura, H., and K. Nakai. 1958. Developmental anomalies in offspring of pregnant mice treated with nicotine. Science 127: 877-878.
See also PubMed.

Nishimura, H., and K. Nakai. 1960. Congenital malformations in offspring of mice treated with caffeine. Proc. Soc. Exp. Biol. Med. 104: 140-142.
See also PubMed.

Nishimura, H. and H. Nimura. 1958. Congenital malformations in mouse embryos induced by 8-azaguanine. J. Embryol. Exp. Morphol. 6: 593-596.
See also PubMed.

Nishimura, H., and S. Takagaki. 1959a. Developmental anomalies in mice induced by 2,3-dimercaptopropanol (BAL). Anat. Rec. 135: 261-267.
See also PubMed.

Nishimura, H., and S. Takagaki. 1959b. Congenital malformations in mice induced by nitrogen mustard. Acta Scholae Med. Univ. Kioto 36: 20-26.
See also PubMed.

Peer, L.A., W.H. Bryan, L.P. Strean, J.C. Walker, W.C. Bernhard, and G.C. Peck. 1958. Introduction of cleft palate in mice by cortisone and its reduction by vitamins. J. Int. Coll. Surg. 30: 249-254.
See also PubMed.

Pennycuik, P.R. 1965. The effects of acute exposure to high temperatures on prenatal development in the mouse with particular reference to secondary vibrissae. Austral. J. Biol. Sci. 18: 97-113.
See also PubMed.

Pinsky, L., and A.M. DiGeorge. 1965. Cleft palate in the mouse: a teratogenic index of glucocorticoid potency. Science 147: 402-403.
See also PubMed.

Pinsky, L., and F.C. Fraser. 1959. Production of skeletal malformations in the offspring of pregnant mice treated with 6-aminonicotinamide. Biol. Neonat. 1: 106-112.
See also PubMed.

Pinsky, L., and F.C. Fraser. 1960. Congenital malformations following a two-hour inactivation of nicotinamide by its analogue, 6-aminonicotinamide, in pregnant mice. Brit. Med. J. 2: 195-197.
See also PubMed.

Poulson, E., J.M. Robson, and F.M. Sullivan. 1963. Teratogenic effect of 5-hydroxytryptamine in mice. Science 141: 717-718.
See also PubMed.

Reed, S.C. 1936a. Harelip in the house mouse. I. Effects of the external and internal environments. Genetics 21: 337-360.
See also PubMed.

Reed, S.C. 1936b. Harelip in the house mouse. II. Mendelian units concerned with harelip and application of the data to the human harelip problem. Genetics 21: 361-374.
See also PubMed.

Rugh, R., and E. Grupp. 1960. Protection of the embryo against the congenital and lethal effects of X-irradiation (Part I and Part II). Atompraxis 6: 209-217.

Rugh, R., E. Grupp, and M. Wohlfromm. 1961. Evidence of prenatal heterosis relating to X-ray induced congenital abnormalities. Proc. Soc. Exp. Biol. Med. 106: 219-221.
See also PubMed.

Runner, M.N. 1954. Inheritance of susceptibility to congenital deformity — embryonic instability. J. Nat. Cancer Inst. 151: 637-649.
See also PubMed.

Runner, M.N. 1959. Inheritance of susceptibility to congenital deformity. Metabolic clues provided by experiments with teratogenic agents. Pediatrics 23: 245-251.
See also PubMed.

Runner, M.N. 1964. General mechanisms of teratogenesis, p. 95-103. In J.G. Wilson and J. Warkany [ed.] Teratology: Principles and Techniques. The University of Chicago Press, Chicago.

Runner, M.N., and C.P. Dagg. 1960. Metabolic mechanisms of teratogenic agents during morphogenesis. Nat. Cancer Inst. Monogr. 2: 41-54.
See also PubMed.

Runner, M.N., and J. R. Miller. 1956. Congenital deformity in the mouse as a consequence of fasting. Anat. Rec. 124: 437-438.

Russell, L.B. 1954. The effects of radiation on mammalian prenatal development, p. 861-918. In A. Hollaender [ed.] Radiation Biology. Vol I, part II. McGraw-Hill, New York.

Russell, L.B., and W.L. Russell. 1954. An analysis of the changing radiation response of the developing mouse embryo. J. Cell. Comp. Physiol. 43 (Suppl. 1): 103-147.
See also PubMed.

Russell, L.B., W.L. Russell, and M.H. Major. 1951. The effect of hypoxia on the radiation induction of developmental abnormalities in the mouse. Anat. Rec. 111: 455. (Abstr.)

Savkur, L.D., B.K. Batra, and B.N. Sridharan. 1961. Effect of 20-methylcholanthrene on mouse embryos. II. Strain C3H (Jax). J. Reprod. Fertil. 2: 374-380.
See also PubMed.

Sinclair, J.G. 1950. A specific transplacental effect of urethane in mice. Texas Rep. Biol. Med. 8: 623-632.
See also PubMed.

Smithberg, M., and M.N. Runner. 1963. Teratogenic effects of hypoglycemic treatments in inbred strains of mice. Amer. J. Anat. 113: 479-489.
See also PubMed.

Steiniger, F. 1940. Über die experimentelle Beeinflussing der Ausbildung erblicher Hasenscharten bei der Maus. Z. Menschl. Vererb. Konst. 24: 1-12.

Sugiyama, T., H. Nishimura, and K. Fukui. 1960. Abnormalities in mouse embryos induced by several aminoazobenzene derivatives. Okajimas Folia Anat. Japon. 36: 195-205.

Thalhammer, O., and E. Heller-Szöllösy. 1955. Exogene Bildungsfehler ("Miszbildungen") durch Lostinjecktion bei der graviden Maus. Z. Kinderheilk. 76: 351-365.
See also PubMed.

Trasler, D.G. 1960. Influence of uterine site on occurrence of spontaneous cleft lip in mice. Science 132: 420-421.
See also PubMed.

Trasler, D.G. 1964. Strain differences in susceptibility to teratogenesis: survey of spontaneously occurring malformation in mice, p. 38-56. In J.G. Wilson and J. Warnkany [ed.] Teratology: Principles and Techniques. The University of Chicago Press, Chicago.

Trasler, D.G., B.E. Walker, and F.C. Fraser. 1956. Congenital malformations produced by amniotic-sac puncture. Science 124: 439.
See also PubMed.

Waddington, C.H., and T.C. Carter. 1952. Malformations in mouse embryos induced by trypan blue. Nature 169: 27-28.
See also PubMed.

Walker, B.E., and B. Crain, Jr. 1959. The lethal effect of spontaneous cleft lip-cleft palate. Texas Rep. Biol. Med. 17: 637-644.

Walker, B.E., and B. Crain, Jr. 1960. Effects of hypervitaminosis A on palate development in two strains of mice. Amer. J. Anat. 107; 49-58.
See also PubMed.

Walker, B.E., and F.C. Fraser. 1956. Closure of the secondary palate in three strains of mice. J. Embryol. Exp. Morphol. 4: 176-189.

Warburton, D., D.G. Trasler, A. Naylor, J.R. Miller, and F.C. Fraser. 1962. Pitfalls in tests for teratogenicity. Lancet 2: 1116-1117.

Watanabe, G., and T.H. Ingalls. 1963. Congenital malformations in offspring of alloxan-diabetic mice. Diabetes 12: 66-72.
See also PubMed.

Watney, M.J., and J.R. Miller. 1964. Prevention of a genetically determined eye anomaly in the mouse by the administration of cortisone during pregnancy. Nature 202: 1029-1031.
See also MGI.

Wilson, J.G., and J. Warkany [ed.] 1964. Teratology: Principles and Techniques. The University of Chicago Press, Chicago. 279 p.

Wolstenholme, G.E.W., and C.M. O'Conner [ed.] 1960. Ciba Foundation Symposium on Congenital Malformations. Little, Brown, Boston. 308 p.

Woollam, D.H.M., and J.W. Millen. 1960a. Influence of thyroxine on the incidence of harelip in the "Strong A" line of mice. Brit. Med. J. 1: 1253-1254.
See also PubMed.

Woollam, D.H.M., and J.W. Millen. 1960b. The modification of the activity of certain agents exerting a deleterious effect on the development of the mammalian embryo, p. 158-177. In G.E.W. Wolstenholme and C.M. O'Conner [ed.] Ciba Foundation Symposium on Congenital Malformations. Little, Brown, Boston.