An increase in the level of vitamin E in the diet of saccate, amictic female Asplanchna modifies the development of their parthenogenetic female embryos developing in utero. The offspring develop prominent outgrowths of the body wall (BWO response), and some mature as mictic females which produce male progeny by haploid parthenogenesis. We have followed the fate of [3H] α-tocopheroI fed to immature females. In each generation, females lose about 50% of their α-tocopherol; the remaining material, which is 76–100 % undegraded, is transmitted almost entirely to their male or female offspring. The α-tocopherol content of a female is proportional to the degree of her BWO response. The results support the hypotheses that vitamin E acts directly on embryos to control development, i.e. that it serves as both an intrinsic and extrinsic control signal.

Organisms make extensive use of extrinsic (environmental) signals, as well as intrinsic signals, for the modification and control of development. The response to extrinsic signals is nowhere more strikingly seen than in the seasonal changes in morphology and mode of reproduction shown by many groups of freshwater planktonic invertebrates. Seasonal changes in morphology (cyclomorphosis) are especially common among the rotifers and Cladocera, where they are often sufficiently drastic to create taxonomic confusion. Seasonal transitions between asexual (parthenogenetic) and sexual reproduction are also characteristic of these groups. These changes in morphology, physiology, and mode of reproduction, which we designate collectively as developmental polymorphism, have in many instances been shown to be controlled by environmental stimuli (see review by Hutchinson, 1967). We have been using the monogonont rotifer Asplanchna for a detailed study of this phenomenon. In particular, we are concerned with the nature of the effective environmental stimuli (Birky, 1964, 1969; Kiechle & Buchner, 1966; Gilbert, 1967,1968,1969; Gilbert & Thompson, 1968; Buchner, Kiechle & Tiefenbacher, 1969), the mode of action of these stimuli at the cellular level (Birky, 1968), and the adaptive significance of the response to these extrinsic signals (Gilbert & Thompson, 1968; Birky, 1969).

Natural populations of A. brightwelli and A. sieboldi, and laboratory populations grown on a diet of paramecia, consist mainly or entirely of females which are saccate (sack-shaped) and amictic (producing clones of genetically identical daughters by diploid parthenogenesis). In natural populations at various times during the summer, the saccate females (especially of A. sieboldi) produce daughters with body-wall outgrowths (BWO’s, or humps; see Birky & Power (1969) for diagrams of various morphotypes). These daughters are often mictic females, which produce males by haploid parthenogenesis or, if fertilized, diploid resting eggs which hatch as amictic females. In the laboratory, the production of daughters with BWO’s and of mictic daughters may be induced at will by the addition of vitamin E (α-tocopherol) to the maternal diet (Gilbert & Thompson, 1968). This treatment also increases the number of nuclei in certain organs and decreases the longevity and fecundity of females (Birky & Power, 1969 ; Birky, 1969). The level of vitamin E in the maternal diet constitutes an extrinsic signal which, directly or indirectly, acts on embryos in utero to control morphogenetic growth of the body wall (BWO response), the mitosis/meiosis alternative in the oocytes (mictic female response), and the number of mitotic divisions in certain embryonic organ Anlagen (Birky, 1968).

In this paper, we ask whether or not vitamin E acts directly on the embryos to control development, i.e. whether or not it also serves as an intrinsic signal or inducer. A direct test, using embryos reared in vitro, proved impractical. However, Birky & Power (1969) have shown that, when amictic females are fed vitamin E during their immature period only (A1 generation), the BWO response is seen not only in their daughters (A2 generation), but also in at least two subsequent generations (A3 and A4) when the initial stimulus is sufficiently strong. The magnitude of the response typically remains constant or increases in the A3 females, then declines in the A4 and A5 generations. If vitamin E itself is the intrinsic inducer, then it must be transmitted intact with high efficiency from generation to generation, with a gradual dilution of effective concentration. We have therefore examined the pattern of uptake, transmission, and loss of labelled α-tocopherol; our data show that it is indeed transmitted intact from parent to offspring with high efficiency, and that the α-tocopherol content of a female is at least roughly proportional to the magnitude of her BWO response. Our results also are consistent with the hypothesis that vitamin E is required in these rotifers for male fertility. This hypothesis provides an evolutionary rationale for the use of dietary vitamin E as an extrinsic signal for sexual reproduction.

Asplanchna brightwelli, inbred clone 5B4S76 (Birky, 1967), was used in one experiment. All other work used A. sieboldi of clone 12C1, which is highly sensitive to dietary vitamin E (Birky, 1969). The animals were kept in plastic dishes or depression slides at 25 °C in the dark. The culture medium (GCF) is a dilute infusion of Perth grass containing Paramecium aurelia (with some Aerobacter aerogenes and miscellaneous contaminating bacteria) (Gilbert, 1968). Penicillin was added in some cases to reduce contamination. Sterile GCF for washing animals was obtained by filtering GCF through paper and subsequently through a Millipore filter. The population density was generally 10 females/ml, with all animals being transferred to fresh medium every one or two days. Under these conditions, the parthenogenetic generation time is about days.

Labelling experiments

Initial labelling experiments used [14C]d-α-tocopheryl succinate (Distillation Products Industries, sp.act. 1·0616 mCi/mM); subsequently we used [3H]dl-α-tocopherol from Amersham/Serle, with a specific activity of 2500 or 2720 mCi/mM and an advertised purity of 97–98 %. Both molecules were labelled in the 5-methyl group; specific activities are those quoted by the manufacturers. The [3H] α-tocopherol was stored in benzene: ethanol under N2 at 4 °C; aliquots were dried and taken up in ethanol as needed for experiments. The [14C] α-tocopherol was stored in ethanol under air at 4 °C and may have undergone some oxidation before use. Labelled and unlabelled α-tocopherol in ethanol were injected with a syringe directly into the culture fluid to form a fine emulsion at a final concentration of 10−6 M (occasionally, 10−7 M). This emulsion is ingested by paramecia and subsequently by the rotifers.

For scintillation counting, labelled animals were washed with sterile GCF to remove all unbound radioactivity, rinsed in distilled water, and transferred with a minimum of fluid to Bray’s solution. Radioactivity is completely extracted in Bray’s ; treatment with Biosolve (Eastman Organic), sonication, homogenization, and/or extraction with methanol alone or with N-hexane released no significant additional amount of radioactivity. Sample vials contained from 1 to 500 rotifers and were counted for from 10 to 100 min as needed to obtain significant counts, in a Nuclear Chicago 6801 scintillation counter using quenched standards and the channels ratio method to determine efficiencies. The results are expressed in net dpm/female and in moles/female (using manufacturers’ stated activities) to correct for variations in counting efficiency, but our conclusions are unchanged if calculations are made on the basis of net cpm/female. Background counts (generally about 27 cpm with 5 % S.D. for 14C and about 31 cpm with 6 % S.D. for 3H) were obtained from matching vials containing rotifers treated identically to the experimental animals, but with unlabelled dl-α-tocopherol. Where appropriate, background counts from a set of control vials were averaged and the mean was used to correct a number of different experimental vials. All counts were significantly above background unless noted otherwise.

The design of a typical pulse-labelling experiment is shown in Fig. 1. Partheno-genetic generations are called A1, A2, etc. Immature A1 females aged 0–5 h are collected from mothers (A0) reared in the absence of added vitamin E. They are allowed to mature in GCF plus α-tocopherol for about 20 h, at which time their first offspring are nearly ready for birth. The A1 females are washed free of unbound radioactivity in sterile GCF ; some are then placed in Bray’s solution for counting and some are put in GCF to obtain the next generation. Between 12 and 24 h later, these animals are separated from their A2 offspring; the old (A1) mothers are either put in Bray’s or are kept in GCF to obtain their later A2 offspring. Of the immature A2 females, which are generally of parity (birth order) I, II, and III, some may be placed in GCF to obtain the A3 generation and some are put in Bray’s for counting.

Fig. 1.

Flow-sheet diagram of a typical experiment to determine the uptake, loss, and transmission patterns of α-tocopherol in Asplanchna. See text for explanation.

Fig. 1.

Flow-sheet diagram of a typical experiment to determine the uptake, loss, and transmission patterns of α-tocopherol in Asplanchna. See text for explanation.

In some experiments, the response of the animals to α-tocopherol was measured using the morphotypic score system of Birky & Power (1969). A sample of animals is killed in 30 % ethanol and expanded in 70 % ethanol, and the development of the body-wall outgrowths (BWO’s or humps) in each animal is scored on an ordinal scale ranging from 1 (saccate) to 5 (fully humped). The mean score of a group of experimental animals is compared to that of the corresponding group of controls reared in GCF without added vitamin E.

Thin-layer chromatography of extracts

In initial experiments, we found that pure [3H] α-tocopherol at the very low concentrations used in these experiments, as well as labelled material extracted from females, underwent rapid oxidation during chromatography, with as much as 96 % of the α-tocopherol being converted to α-tocopheryl quinone and other compounds. This problem could be overcome partially by minimizing exposure to light and air and especially by the addition of large amounts of cold carrier α-tocopherol to the extraction mixture. The final procedure adopted began with extraction of females in 1 ml re-distilled acetone containing 0·1 mg each of dl-α-tocopherol and α-tocopheryl quinone, under N2 at 4 °C for 1 h. The acetone was removed and the animals, after being rinsed in 1 ml fresh acetone, were dried in a vial for counting. From the pooled extracts, an 0·2 ml sample was taken and also dried for counting; the remaining extract was dried under a stream of N2, taken up in chloroform, and spotted on plates with 250 μm layers of silica gel G. Adjacent channels on each plate were used for the extracts of labelled females, extracts of unlabelled control females (for background counts), and extracts of unlabelled females to which had been added, at the beginning of extraction, a sample of [3H] α-tocopherol of approximately the same activity as the labelled females (to determine the amount of oxidation occurring during the procedure).

All plates were developed in chloroform under N2 in dim light, and then sprayed with I2 vapor for approximate localization of the positions of α-tocopherol and α-tocopheryl quinone. Each channel was divided into 16 to 18 segments of about 1 cm each, which were then scraped off into individual vials. Scrapings, extract aliquots, and extracted females were counted in toluene-PPO-POPOP fluid. Counts of dried animals and of extract aliquots were used to calculate the efficiency of extraction and of recovery of material from the plates. Extraction of label ranged from 2 to 95 %, with most values 89 % or greater; the low values are almost certainly in error due to difficulties in handling the animals and extracts. Calculated recovery ranged from 50 to 181 %. Neither extraction efficiency nor recovery showed any correlation with the calculated percentages of α-tocopherol in the animals, and probably introduced no significant bias into the results.

The dl-α-tocopherol and α-tocopheryl quinone used in these experiments were purchased from Pierce Chemical and Nutritional Biochemicals Corporation, respectively. Three dimeric oxidation products of α-tocopherol (the spiro dimer; 5,5′-Bi-α-tocopherol; and 5,5′-methylene bis-(γ-tocopherol)), and toco-pheronolacteone were donated by Distillation Products Industries through the kindness of Dr David C. Herting.

Autoradiography of labelled females

Following conventional fixation schedules in osmic acid or glutaraldehyde, labelled α-tocopherol is rapidly extracted from tissues during dehydration in . ethanol before embedding in methacrylate, or during embedding in the watersoluble medium ‘Polyamph’ (Polysciences, Inc.). This difficulty was overcome, at least for some blocks, by fixing washed animals in 2 % osmic acid for 17 h at 4 °C, washing, and post-fixing in 0·5 % KMnO4 for 15 min. Fixatives were made in 50 % Gilbert’s (1963) saline. After careful rinsing, the animals were dehydrated rapidly in ethanol and embedded in methacrylate. Two-micron sections were covered with Kodak NTB 3 emulsion and exposed for 2| months. Only about 3 % of the total label was extracted during dehydration.

Pulse and long-term labelling of A1 females

The results of the first labelling experiments are shown in Table 1. Females from a culture exposed to [14C] α-tocopherol continuously for 6 days, including animals from five or six successive amictic generations, contained about 5·4 × 10−13 moles of α-tocopherol per female. (This assumes that all label is in α-tocopherol; see below.) A1 females labelled during their immature period only (20–24 h) contained from 1·1 to 4·8 × 10−13 moles/female. These results with [14C] α-tocopherol are consistent with those obtained with [3H] α-tocopherol (Tables 2–4).

Table 1.

Incorporation of [14C] α-tocopherol by Asplanchna sieboldi females

Incorporation of [14C] α-tocopherol by Asplanchna sieboldi females
Incorporation of [14C] α-tocopherol by Asplanchna sieboldi females
Table 2.

Incorporation and transmission of [3H]dl-α-tocopherol by Asplanchna* females. A1 females labelled 18–21·5 h in 10−6 or 10−7 M α -tocopherol

Incorporation and transmission of [3H]dl-α-tocopherol by Asplanchna* females. A1 females labelled 18–21·5 h in 10−6 or 10−7 M α -tocopherol
Incorporation and transmission of [3H]dl-α-tocopherol by Asplanchna* females. A1 females labelled 18–21·5 h in 10−6 or 10−7 M α -tocopherol

The incorporation by A1 females varied greatly between experiments. The range was 11·6–753 × 10−15 moles/female, or, excluding the unusually low value in Exp. 10, 106–753 × 10−15 moles/female. A statistical comparison of counts of individual A1 females from Exps. 9 and 11 showed that the difference in mean Ax uptake was highly significant, and cannot be accounted for by sampling error or by differences in background counts. We conclude that these differences represent real variability in uptake by females. Such variability is not surprising, as it is completely consonant with biological variability in Asplanchna measured by other means, e.g. (1) mictic female production induced by α-tocopherol (Gilbert & Thompson, 1968); (2) reproductive parameters (C. W. Birky & J. J. Gilbert, unpublished); and (3) sensitivity to streptomycin (C. W. Birky & A. E. Brodie, unpublished).

Label in females exposed to [14C] α-tocopherol continuously for 6 days (Table 1, Exp. 1) fell well within the range of label in females exposed to 20-to 24-h pulses of [14C] or [3H] α-tocopherol. This result agrees with previous data on the morphological response to vitamin E (Birky & Power, 1969) and suggests that the animals may be essentially ‘saturated’ withα-tocopherol within 24 h or less. It was expected that at least a portion of this label might be adsorbed on the surface of the animals rather than ingested. This was verified by killing females with KCN and placing them in GCF plus [14C] α-tocopherol for 20 h. The corpses contained approximately one-third as much radioactivity as did living females. We have not attempted to use this value to correct our A1 data for surface adsorption, because it is unlikely that living animals have precisely the same adsorption properties as corpses.

In a later experiment (Exp. 9) using [3H] α-tocopherol of very high specific activity, individual Ax females were counted to obtain information about the individual variation in uptake. Counts of ten females labelled for 20 h with 10−6 M [3H] α-tocopherol showed a range of 2·84–11·32 × 10−13 moles/female, with a mean of 7·53 ×10−13 moles/female. Individual A1 females thus show a fourfold variation in uptake of vitamin E; it is not known how much of this variability is in actual ingestion and how much is in adsorption.

Fate of ingested α-tocopherol: patterns of loss and transmission

Table 2 shows the data from seven experiments concerning the loss of labelled α-tocopherol and its transmission from generation to generation. It should be noted first that, in contrast to the almost 70-fold variation in A1 uptake, the percentage of label which is transferred from generation to generation is relatively consistent from experiment to experiment. The following pattern emerges for A. sieboldi (Exps. 7, 9, 10, 11 and 12):

(1) A2 females transfer 14·5–28·7 % of their total label to their first-born offspring (A3I females). Succeeding A3 offspring (parities II–IV) receive successively smaller portions of the total parental label. A3 males (Exp. 7) each receive smaller proportions of parental label (10·2 %) than do females.

(2) A1 females transfer smaller proportions (3 · 4 – 16 · 3 %) of their total label to their A2I offspring; again, subsequent offspring receive successively smaller portions of the parental label. The interpretation of the A1 – A2 data is complicated by the probability that some of the A1 label represents material adsorbed on the surface of the females. This material might well contribute to the loss of label (discussed below), by being washed off as the animals swim or by being oxidized by the bacteria which also adsorb to the animals’ surfaces, but it could not be transferred to their daughters. It may be that, in Exps. 7 and 11, which show a markedly lower transfer from A1 to A2 than from A2 to A3, the A1 females had a relatively high proportion of surface-bound label. In Exp. 9a, counts were made of 45 individual A2 females, of known parity, from nine A1 parents. A2 females of parity I contained 8 · 27 – 15 · 7 × 10−14 moles/female, thus showing a nearly twofold individual variation. When contrasted with the fourfold individual variation in label content in their A1 parents (see above),.these data provide further evidence that much of the A1 variation is due to surfacebound label which cannot be transmitted. In each of the nine sibships studied, with few exceptions, A2 females of increasing parity contained decreasing amounts of label.

(3) For purposes of comparison, a single experiment (Ab) was done with the closely-related species A. brightwelli (inbred clone 5B4S76; Birky, 1967). This species produces mictic females in response to vitamin E more readily than does A. sieboldi, but forms only weakly developed body-wall outgrowths, and those only after a lag of several generations (Birky (1964) and unpublished; see also Kiechle & Buchner (1966), where this species is erroneously identified as A. sieboldi). The transfer and loss patterns for this species appear to be very similar to those for A. sieboldi. Note especially that in both species, A3 males receive a smaller portion of their parents’ label than do A3 females; this is not due to smaller amounts of label in their parents, for the amount of label is similar in mictic and amictic females of the A2 generation.

The loss of label from rotifers to their environment can be estimated by comparing the total label in a group of females plus their offspring with the label in a sample of the same group of females taken within 24 h after their birth and before their offspring are born. The results of these calculations are shown in Tables 3 and 4. In general, an A1 or A2 female loses 40 – 50 % of her total label to the environment in the course of producing four to six female offspring, which is the usual maximum under our conditions. A loss of more than 60 % is seen at low population densities (Exp. 12LD, discussed below), in the parity IV A2 females of Exp. 11 (where the value is uncertain due to low counts in their A3 progeny), and in amictic A2 females of A. brightwelli. K large portion of this loss of label may occur early in life, coincident with the birth of the first one or two offspring (see A1 in Exp. 12, A2 in Exp. 96, and Exp. Ab).

Table 3.

α -Tocopherol in Asplanchna: percent loss by A1 females, and percentage of available A1 label received by A2 females (calculated excluding label lost or already transmitted)

α -Tocopherol in Asplanchna: percent loss by A1 females, and percentage of available A1 label received by A2 females (calculated excluding label lost or already transmitted)
α -Tocopherol in Asplanchna: percent loss by A1 females, and percentage of available A1 label received by A2 females (calculated excluding label lost or already transmitted)
Table 4.

α -Tocopherol in Asplanchna: percentage loss by A2 females and percent of available A2 label received by A3 females or males (calculated excluding label lost or already transmitted)

α -Tocopherol in Asplanchna: percentage loss by A2 females and percent of available A2 label received by A3 females or males (calculated excluding label lost or already transmitted)
α -Tocopherol in Asplanchna: percentage loss by A2 females and percent of available A2 label received by A3 females or males (calculated excluding label lost or already transmitted)

Tables 3 and 4 also show the percentage of available label transferred from mother to offspring; this is calculated excluding both the label which is lost and the label which has been transferred to offspring already born. A comparison with Table 2 shows that these percentages are relatively constant for offspring of different parity, at least through parity V. For the transfer from A2 to A3 females, the results show an average transfer of about one-third of the available parental label.

For the purposes of a later discussion, the following generalizations suffice to describe the results: (1) in each generation, about one-half (rarely two-thirds) of the total α -tocopherol in a female is lost to the environment; (2) of the remaining α -tocopherol, nearly all is transferred to offspring: (3) approximately one-third of the available α -tocopherol in a female at any given time is transferred to her next female offspring; and (4) each male offspring of a mictic female receives about one-tenth of its parent’s available label.

Correlation of vitamin E content with response

In some experiments, the BWO response of the animals to vitamin E was measured by determining their morphotypic scores. A positive correlation is apparent between morphotypic scores and α -tocopherol content. This can be seen in Exp. 9 (Table 2), where the A2 females of increasing parity show decreasing label content and decreasing mean scores. In Exp. 11, the decrease in morphotypic score with parity in the A2 and A3 generations is less striking, but it is clear that there is an overall decline in response between generations. In addition, A3 females from A2 mothers of parity IV have both a lower response and a lower label content than those from parity IA2 mothers. We have applied the Spearman rank correlation test to these data; the correlation between α -tocopherol content and mean morph score is highly significant in Exp. 9 (P ⪡ 0 · 01) and significant in Exp. 11 (0 · 01 < P < 0 · 05).

Birky (1969) has shown that females raised at higher population densities are more sensitive to α -tocopherol. This phenomenon was used in Exp. 12 (Tables 2 and 3) as an additional test for correlations between α -tocopherol content and the BWO response. A1 females labelled for 18 h at a density of 10 females/ml (12H) incorporated more α -tocopherol and transferred a greater proportion of their total label to their first-parity A2 daughters than did females labelled at a density of 1 female/ml (12L). The more effective transmission was due to a lower loss of label in the high-density females. This effect may be quite independent of the soluble sensitizing factor(s) demonstrated by Birky (1969); it is possible, for example, that label which is lost is promptly re-ingested, more rapidly at higher population densities. Whatever the mechanism, it is clear that the higher label content of the high-density A2 females was correlated with higher morphotypic scores. Taken together, the results of our experiments indicate a clear correlation between the α -tocopherol content of the rotifers and the degree of their BWO response.

Localization of label in embryos and body-cavity fluids

In order to directly demonstrate the presence of α -tocopherol in embryos and to determine the approximate stage of development at which it is transferred from parent to daughter, A1 females were labelled for 20 h, washed, and then drawn into a fine-bore pipet in order to force out their embryos, along with the maternal uterine fluid and much of the pseudocoel cavity fluid. The embryos were collected in water, pooled into three groups representing different developmental stages, and counted in Bray’s. The remaining maternal tissues were counted separately. The results are shown in Table 5. All stages of embryos were heavily labelled; the ‘mature’ embryos, which were in the final developmental stages of swelling due to uptake of fluid and were nearly ready for birth, contained more than twice as much α -tocopherol as earlier stages. About 41 % of the label initially present in the females could not be accounted for by the sum of the label in the embryos and in the maternal tissues after the embryos and fluid were removed. The bulk of this radioactivity is believed to represent α -tocopherol in the maternal uterine fluid and/or pseudocoel cavity fluid, which was lost during removal of the embryos.

Table 5.

[3H]dl-α -tocopherol in embryonic and adult tissues of Asplanchna sieboldi*

[3H]dl-α -tocopherol in embryonic and adult tissues of Asplanchna sieboldi*
[3H]dl-α -tocopherol in embryonic and adult tissues of Asplanchna sieboldi*

Autoradiographic localization of labelled α -tocopherol

Additional evidence for a localization of α -tocopherol in the maternal uterine fluid was obtained by autoradiography. Most autoradiographs of A1 and A2 females showed no obvious localization of label in any tissues except for the expected light concentration in stomach cells ; embryos were also weakly labelled. A1 females in two blocks, however, differed from others in showing a dense, osmium-stainable material in the uterus. This material showed a marked accumulation of label (Fig. 2). We believe that the dense material represents the uterine fluid or some component thereof, in which labelled α -tocopherol is concentrated. Probably, for unknown reasons, the material was not well fixed in most blocks and, together with the label, was extracted during dehydration and/or embedding.

Fig. 2.

Autoradiograph of A1 female A. sieboldi fed 10−6M [3H] α -tocopherol for 20 h after birth. Note intensive label in uterine cavity (ut), and absence of label in pseudocoel (pc) and stomach cavity (stc). There is some label in the stomach wall (stw), obscured by heavily stained lipid droplets. Focus is a compromise between the levels of the tissue and the emulsion. ( × 700.)

Fig. 2.

Autoradiograph of A1 female A. sieboldi fed 10−6M [3H] α -tocopherol for 20 h after birth. Note intensive label in uterine cavity (ut), and absence of label in pseudocoel (pc) and stomach cavity (stc). There is some label in the stomach wall (stw), obscured by heavily stained lipid droplets. Focus is a compromise between the levels of the tissue and the emulsion. ( × 700.)

Identification of labelled material by thin-layer chromatography

The results of the most successful thin-layer chromatographs of extracts of labelled females are summarized in Table 6. Examination of the data for control runs of highly (advertised as 97 – 98 %) pure [H3] α -tocopherol indicates that some oxidation of the tocopherol is occurring on the plate, in spite of the precautions described under Materials and Methods. It is therefore necessary to correct the percentage of α -tocopherol found in the extracts for this oxidation. When this is done, it is found that between 76 and 100 % of the labelled material present in A1, A2 and A3 females migrates as α-tocopherol. This material is therefore probably undegraded α-tocopherol. Variable amounts of what is probably α -tocopheryl quinone or perhaps tocopheronolactone were also found in all runs (authentic samples of these two compounds have similar Rf values under our conditions). No marked accumulations of radioactivity were found with Rf values of dimeric oxidation products (the spiro dimer and 5,5′-Bi-α -toco-pherol, structures I and III respectively of Nelan & Robeson, 1962, and 5,5′- methylene bis-( γ -tocopherol)).

Table 6.

Thin-layer chromatography of acetone extracts of female Asplanchna sieboldin

Thin-layer chromatography of acetone extracts of female Asplanchna sieboldin
Thin-layer chromatography of acetone extracts of female Asplanchna sieboldin

Four conclusions emerge directly from our experimental data.

(1) Of the α -tocopherol consumed by female Asplanchna, about half is lost to the environment in each generation; the remainder is largely present in apparently undegraded condition

This stability of α -tocopherol in vivo is in marked contrast to its extreme susceptibility to oxidation in vitro ; it indicates the existence of a mechanism for the protection of the molecule from metabolism (including extensive use as an antioxidant) and its preservation for other purposes. The fate of the lost material is unknown. It may be metabolized before it is lost, but if so, the retention time of derivatives of the labelled methyl group, at least, must be quite short. Nor do we know the mechanism of loss. Since a large part of the labelled α -tocopherol seems to be localized in a uterine fluid, it is possible that it leaks out during birth; this would explain why much of the loss occurs at about the time of birth of the first offspring.

(2) The remaining α-tocopherol present in each generation is almost entirely transmitted to offspring

In the A1 generation, this means that vitamin E must be efficiently transferred from the stomach to the pseudocoel cavity fluid and hence to the embryo via the uterus and possibly also via the vitellarium. The maternal vitellarium, which functions analogously to nurse cells in other organisms, cannot be the only source of vitamin E for embryos, since vitamin E fed to mature females can modify the development of embryos already detached from the vitellarium (Gilbert, 1968; Birky, 1968). This is confirmed by the autoradiographic localization of labelled α -tocopherol in the uterine cavity, and by the scintillation counting experiment which indicated that at least 40 % of the total label in A1 females is in the pseudocoel or uterine fluids, or both.

Krishnamurthy & Bieri (1963) have studied the stability and retention of d- α -tocopherol-5-methyl-14C, administered orally to rats and chicks partially depleted of vitamin E. In rats, about one-third of the total dose was excreted or could not be recovered after 24 h ; about 1 °/0 was excreted each day for the next 20 days. The bulk of the excreted and retained label was in the form of α -tocopherol. Retention was somewhat lower in the chicks, and a considerable amount of the α -tocopherol was degraded after 24 h. No attempt was made to detect radioactivity in the offspring.

By extrapolation from our autoradiographs, it seems likely that vitamin E in Asplanchna would also be localized in body fluids in the A2 and subsequent generations, even though these animals have obtained their vitamin E from their parents rather than by ingestion. This has been verified by Oliver (1970), who has obtained the combined pseudocoel and uterine fluids by dissecting females under oil, and shown that this fluid can induce the BWO response in the progeny of females reared from birth in the fluid. The response was obtained with the fluid from females whose parents had been fed the water-soluble derivative, α -tocopheryl polyethylene glycol 1000 succinate (TPGS), and also from control females taken from cultures with no added vitamin E. The fluids were also fractionated by thin-layer chromatography and the fractions tested separately; the major inducing component in the fluid from induced females migrated with the Rf of α-tocopheryl succinate (presumably derived from TPGS), while the major inducer in the fluid control females was identified by its Rf as α-tocopherol - presumably obtained from the traces identified in Scottish grass infusion by Gilbert & Thompson (1968), and responsible for the weak BWO responses often observed in control cultures. In addition, Oliver (1970) has shown that embryos taken from control females can be induced to form weakly developed BWO’s by rearing them in vitro in the combined pseudocoel and uterine fluids taken from induced females.

(3) Successive offspring receive smaller amounts of their parental α-tocopherol, but approximately constant amounts of that remaining after the birth of their older sibs; males receive less than females

In rotifers, one oocyte is matured at a time. After maturation, which requires about 6 h, the oocyte is set free from the vitellarium and undergoes complete development in the uterus (about 20 h) while the next oocyte begins maturation. Mature amictic females thus contain four to five female offspring in different stages of development; mictic females produce more (male) offspring and generally carry more embryos at one time. These observations suggest a possible model for the transmission of vitamin E from generation to generation. We suppose that vitamin E is specifically localized in the uterine fluid, and that most of this localization occurs before reproductive maturity. The molecules are absorbed by maturing embryos from the uterine fluid; as successive embryos mature and are born, the concentration of vitamin E and consequently the amount available for absorption declines. Male embryos, competing for vitamin E with a larger number of siblings in utero, would obtain smaller amounts of label than females.

(4) There is a positive correlation between the α-tocopherol content of females and their BWO response

This generalization cannot, of course, apply to A1 females, which cannot respond because they were not exposed to the α-tocopherol as embryos. Other possible exceptions will be considered below.

Our experiments were begun in order to provide at least an indirect test of two hypotheses about the role of vitamin E in developmental polymorphism. The first of these hypotheses is that vitamin E itself acts directly upon embryos in utero to control development, i.e. that it is the intrinsic as well as the extrinsic inducer. Alternative hypotheses would be that the intrinsic inducer is a metabolic derivative of α -tocopherol, or another molecular species or a metabolic state induced in the females by α -tocopherol. If α-tocopherol is the intrinsic inducer, it would have to be transmitted intact in reasonable quantity from a pulse-fed A1 female to her progeny through as many as four generations in order to explain the short-term inheritance of the BWO response. It might also be possible to show at least a rough correspondence between endogenous α - tocopherol levels and the magnitude of the BWO response. Both of these conditions have been met in the experiments reported here. While this does not directly prove that α-tocopherol is an intrinsic inducer, it argues strongly for the hypothesis.

In this connexion, two observations of Birky & Power (1969) posed some difficulty for the identification of α-tocopherol as the intrinsic inducer, and require special comment. In many experiments where A1 females were fed vitamin E during their first day of life, it was found that (1) their first three or four A2 progeny showed progressively increasing morphotypic scores, whereas our present data indicate that they should have contained decreasing amounts of α-tocopherol, and (2) mean morphotypic scores remained constant or rose between the A2 and A3 generations, and then declined approximately linearly, while α-tocopherol content should decline logarithmically in every generation. Our data suggest two probable explanations. First, it is possible that α - tocopherol in A1 females is localized in the uterine fluid rather slowly, so that parity IA2 embryos absorb most of their α-tocopherol too late in development to influence morphogenesis (cf. Table 5). Thus, although their content of a-tocopherol would be greater than that of embryos of later parity and of the next (A3) generation, their response would be lower.

The second explanation is suggested by the data of Gilbert & Birky (1971), who determined the BWO response in A2 females when their A4 parents were fed concentrations of α-tocopherol ranging from 10−11 M to 5 × 10−7 M. Their data indicate a maximal response at about 5 × 10−8 M; the response declines steeply below this, while higher concentrations may actually be slightly inhibitory. The experimental conditions used by Gilbert & Birky (1971) are not strictly comparable to those used in the present study. However, we have used our present data on uptake and transmission to estimate the actual α-tocopherol content of the A2 females in the previous study, and to plot a rough curve of concentration against response. The results show a peak response at 3 × 10−14 moles/female, with higher contents being inhibitory. From an examination of the data in Table 2, it seems likely that the first-born A2 females of Birky & Power (1969) actually contained inhibitory amounts, and the A3 and later A2 females contained optimal amounts, of α-tocopherol. Our calculations from the data of Gilbert & Birky (1971) also indicate that their A2 females showed a positive BWO response when they contained about 5 × 10−16 moles of α - tocopherol, and no response at about 1 × 10−16 moles. This estimate of the minimum amount of α-tocopherol required to induce a BWO response in A2 females agrees well with the data in the current paper. In Exp. 11 (Table 2), A3 females showed a positive response at about 9 × 10−16 moles and no response at 5 × 10−16 moles. In Exp. 7, we can estimate that the A4 and A5 females contained about 9 × 10−16 and 3 × 10−16 moles/female respectively (by extrapolation from the A2 and A3 data in Table 2); the A4 females showed a significant BWO response, while the A5 did not. Our data thus suggest that, in any generation, about 5 · 10 × 10−16 moles, or 3 · 6 × 108 molecules, of α-tocopherol are required to induce a recognizable BWO response. This, of course, is in addition to the basal content of α-tocopherol found in all females and obtained from the Scottish grass infusion.

Our second hypothesis about vitamin E concerns the adaptive significance of the use of this molecule as a necessary, if not sufficient, stimulus for the production of mictic females. We have suggested (Gilbert & Thompson, 1968; Birky, 1969; and especially Gilbert, 1971) that high levels of vitamin E may be essential for the fertility of male, but not female, Asplanchna. If so, it would be entirely reasonable for rotifer populations to have evolved mechanisms which would require the availability of high levels of vitamin E before sexual reproduction is initiated. Since male rotifers do not feed, they can only obtain vitamin Efrom their parents; if it is required for fertility, a mechanism would also have to be evolved to ensure that dietary vitamin E is protected from degradation and efficiently transmitted from parent to offspring. Our present data show that this is indeed the case. It seems to us extremely unlikely that the remarkable behavior of vitamin E, which sets it apart from the bulk of dietary macromolecules, is fortuitous; it probably indicates a special role for the molecule. Our hypothesis was originally suggested by the necessity of vitamin E for male, but not female, fertility in certain other organisms, and is consistent with the autoradiographic localization of α-tocopherol in the male testis (Gilbert, 1971).

We thank Dr George Malacinski for helpful comments on the manuscript and Mrs Maxine Bean for expert and enthusiastic assistance. The work reported in this paper was supported by National Science Foundation Research Grant GB-7717 to J. J. G. and U.S. Public Health Service Research Fellowship 1 F03 GM 43071-01 to C. W. B., Jr.

Birky
,
C. W.
Jr
. (
1964
).
Studies on the physiology and genetics of the rotifer, Asplanchna. I. Methods and physiology
.
J. exp. Zool.
155
,
273
292
.
Birky
,
C. W.
Jr
. (
1967
).
Studies on the physiology and genetics of the rotifer, Asplanchna. III. Results of outcrossing, selfing, and selection
.
J. exp. Zool.
164
,
105
116
.
Birky
,
C. W.
Jr
. (
1968
).
The developmental genetics of polymorph ism in the rotifer Asplanchna. I. Dietary vitamin E control of mitosis and morphogenesis in embryos
.
J. exp. Zool.
169
,
205
210
.
Birky
,
C. W.
Jr.
(
1969
).
The developmental genetics of polymorphism in the rotifer Asplanchna. III. Quantitative modification of developmental responses to vitamin E, by the genome, physiological state, and population density of responding females
.
J. exp. Zool.
170
,
437
448
.
Birky
,
C. W.
Jr.
&
Power
,
J. A.
(
1969
).
The developmental genetics of polymorphism in the rotifer Asplanchna. II. A method for quantitative analysis of changes in morphogenesis induced by vitamin E and the short-term inheritance of the effects of vitamin E
.
J. exp. Zool.
170
,
157
168
.
Buchner
,
H.
,
Kiechle
,
H.
&
Tiefenbacher
,
L.
(
1969
).
Untersuchungen fiber die Bedingungen der heterogonen Fortpflanzungsarten bei den Radertieren I. Die miktische Reaktion, ihre Beziehungen zur Populationsdynamik und ihre Abhângigkeit vom Milieu
.
Zool. Jb. Physiol.
74
,
329
426
.
Gilbert
,
J. J.
(
1963
).
Mictic female production in the rotifer Brachionus calyciflorus.
J. exp. Zool.
153
,
113
124
.
Gilbert
,
J. J.
(
1967
).
Control of sexuality in the rotifer Asplanchna brightwelli by dietary lipids of plant origin
.
Proc. natn. Acad. Sci. U.S.A.
57
,
1218
1225
.
Gilbert
,
J. J.
(
1968
).
Dietary control of sexuality in the rotifer Asplanchna brightwelli Gosse
.
Physiol. Zoöl.
41
,
14
43
.
Gilbert
,
J. J.
(
1969
).
Control of polymorphism in Asplanchna by two compounds in dried grass
.
Am. Zool.
9
,
620
(abstr.).
Gilbert
,
J. J.
(
1971
).
a-Tocopherol in males of the rotifer Asplanchna sieboldi : its metabolism and its distribution in the testis and rudimentary gut
.
J. exp. Zool. (In the Press
.)
Gilbert
,
J. J.
&
Birky
,
C. W.
Jr.
(
1971
).
The sensitivity and specificity of the Asplanchna response to dietary α-tocopherol
.
J. Nutr.
101
,
113
126
.
Gilbert
,
J. J.
&
Thompson
,
G. A.
Jr.
(
1968
).
Alpha tocopherol control of sexuality and polymorphism in the rotifer Asplanchna.
Science, N. Y.
159
,
734
738
.
Hutchinson
,
G. E.
(
1967
).
A Treatise on Limnology,
vol.
2
.
New York
:
John Wiley and Sons, Inc
.
Kiechle
,
H.
&
Buchner
,
H.
(
1966
).
Untersuchungen fiber die Variabilitat der Râdertiere: V. Dimorphismus und Bisexualitât bei Asplanchna
.
Ann. Soc. Suisse Zool.
73
,
283
300
.
Krishnamurthy
,
S.
&
Bieri
,
J. G.
(
1963
).
The adsorption, storage, and metabolism of α-tocopherol-C14 in the rat and chicken
.
J. Lipid Res.
4
,
330
336
.
Nelan
,
D. R.
&
Robeson
,
C. D.
(
1962
).
The oxidation product from α-tocopherol and potassium ferricyanide and its reaction with ascorbic and hydrochloric acids
.
J. Am. chem. Soc.
84
,
2963
2965
.
Oliver
,
I.
(
1970
).
Control of morphogenesis by vitamin E in the Rotifer Asplanchna. Master of Arts Thesis, University of California, Berkeley
.