Organismic and Molecular Approaches to the Study of Adaptations

Zoology

[441] One form of molecular approach to adaptations consists in identifying biochemical aspects of organisms that relate to physical or biological aspects external to those organisms—that is, adaptive aspects encountered in studies starting at the molecular scale.

A genuine molecular approach to adaptation must, however, begin by considering it at the level of the organism, where natural selection has its impact, and then proceed to analyse that adaptation down to the molecular level. Such is the proper nature of the molecular approach. The present work brings together an overview of the results of such an analysis concerning the spinning of a cocoon by the silkworm.

Animals do not always spend their entire existence in the same environment or in the same mode of relation to it. In many cases, changes in organism–environment relations are accompanied by morphological changes in the structures of the organism: this is what is called metamorphosis. A frog tadpole has a tail and no legs; it has gills and no lungs. As soon as it reaches metamorphosis, legs appear while the tail and gills disappear. Remodelling has produced modified anatomical structures adapted to the new environment, and the change has occurred before the animal enters [442] that environment. Remodelling may also involve new biochemical adaptations suited to the new environment. The ureotelism that appears in the adult frog is an adaptation to the aerial environment, helping to counteract the effects of dehydration through evaporation, and results from an increase in the content of ureogenesis enzymes in the cells of the hepatic parenchyma (Brown and Cohen, 1958).^[T1] During metamorphosis, tadpole haemoglobin is exchanged for a new haemoglobin with a lower affinity for oxygen and displaying a Bohr effect. The higher affinity of tadpole haemoglobin for oxygen appears as an adaptation to frequent variations in oxygenation of the aquatic environment, whereas the appearance of the Bohr effect and its reciprocal, the Haldane effect, may be regarded as a physiological adaptation linked to CO₂ exchange in the lungs. Bennett and Frieden (1962) surveyed a series of biochemical adaptations linked to the passage of amphibians from the tadpole to the adult stage. It is becoming increasingly apparent that metamorphosis is not merely a morphological modification, but involves a series of biochemical changes constituting transitions from adaptation to one environment to adaptation to a new environment. These biochemical adaptations linked to ecological relations confirm the notion brought out by Snodgrass that, for example, the larvae of holometabolous insects cannot be regarded as representing aspects of embryonic development or aspects of phylogenetic recapitulation. The adaptations operating during the larval and pupal stages of lepidopteran development, for example, are adaptive in nature and bear witness to specialisation relative to the adult stage. Since the different developmental stages of a holometabolous insect possess the same genotype, it appears that different sequences of purine and pyrimidine bases are brought into play at different stages by triggering the action of different operators, under the influence of controllers acting at different times. A very precise biological clock regulates the development of Bombyx mori. The silk glands, which are modified [443] salivary glands, become active on the fourth day of the fifth instar in the Europe strain.^[T2] Once the cocoon has been spun, proteolytic enzymes are released within the glands and catalyse their lysis. Between these two events, the glands synthesise the silk of the cocoon that the animal spins and in which the larva encloses itself to pupate and pass through the pupal stage under its protection. The importance of such protection for the survival of wild forms possessing it is obvious.

The present work is devoted to a very detailed study of the metabolic origins of the proteins of the silkworm cocoon. The cocoon represents an adaptive radiation of protein synthesis. The objective of the study reported here is to situate this biochemical adaptation within the biochemistry of the silkworm, in order to define the biochemical aspects underlying the adaptation and thereby provide an aetiological definition of it extending down to the molecular scale.^[T3]

Research into hormonal actions has led to the introduction of experimental surgery in insects. The methods described below were adopted for a different objective: the study of metabolism. Four moults separate the five instars of the larval life of Bombyx mori. Before each moult, the silkworms stop eating mulberry leaves and remain motionless. Their body surface takes on a waxy appearance. A triangular mark appears at the posterior part of the head, beneath the dorsal integument of the prothoracic segment. At the end of the “triangle” period, ecdysis occurs. Since the fifth instar varies considerably in duration, comparable reference points must be found for metabolic studies. From the nutritional point of view, three periods can be distinguished during the fifth instar. If larvae are deprived of food during the period from the first to the fifth or sixth day, they die before reaching the pupal moult. This portion of the fifth instar will therefore be called the period of essential feeding. If, after being fed from the first to the fifth or sixth day, the silkworms are left without [444] food during a second period, from the fifth or sixth to the tenth or eleventh day, they undergo the pupal moult normally, but without spinning a cocoon, laying down only a small number of threads. During the fourth instar, a small portion of the chains of ingested protein foods is directed towards the silk glands, whereas during the fifth instar the relative proportions of food used by the tissues and taken up by the gland are progressively modified, so that during the second period, that of optional feeding, the greater part of the food material is directed towards the silk glands. This explains why the silkworm can be deprived of food during the second part of the fifth instar, with the consequence that it will spin no cocoon or only a rudimentary one. If, on the other hand, the silk gland is surgically excised—most easily done on the third or fourth day after the third larval moult—most silkworms pass normally through metamorphosis and reach the adult and reproductive stage.

The last period of the fifth instar is a fasting period. A reliable reference point is provided by what is called the purge, or final defaecation: the evacuation of the digestive tract as a brownish liquid rich in phosphates. After the final defaecation, the course of metabolic events is much less variable than during the preceding period. In the silkworm, the day of the final defaecation can be determined precisely. As soon as the silkworms stop feeding, each larva is placed in a paper cone. The cones are examined daily, and the day on which the yellowish purge stain appears is noted on the paper. The silkworms spin their cocoons at the tips of their cones, and four days after the purge the completed cocoon can be removed. One end can be cut off, like the top of a boiled egg, to follow the shedding of the integument that precedes the pupal moult. What happens in silkworms deprived of their glands during the pupal moult (Jeuniaux and Florkin, 1958)?

Weight loss is observed in both normal silkworms and glandectomised ones. This results from evaporation occurring in both groups during a [445] fasting period. But there is a difference between the controls and the glandectomised animals, which obviously spin no cocoon. Comparing the dry-residue weights of the prepupae reveals a clear difference: that of the glandectomised animals is 1.4 times that of the controls. If, on the other hand, both glandectomised animals and controls are put on a fast at the end of the essential-feeding period, weight loss begins at that point, and there is no longer any difference between the two groups.

These experimental results agree with the notion that food taken during the optional-feeding period is used to form the cocoon, partly directly and partly through the intermediary of the tissues, which in the controls supply the material for silk formation after fasting begins. If this final silk formation does not occur, pupal weight becomes greater, since the weight-regulating effect of the last phase of silk production is absent.

There is no doubt that the most important scientific contribution of studies on silk lies in the results of the X-ray diffraction patterns obtained by Nishikawa and Ono in 1913. It was during their research on silk fibroin that these authors proposed the concept of the extended polypeptide chain. These brilliant studies by the Japanese authors somewhat overshadowed the biological interest of the cocoon as mechanical protection for the prepupa during the pupal moult and for the pupa itself. In arthropods, silk is most often secreted by labial glands opening in the head region. This adaptive radiation of protein synthesis, resulting in structures whose mechanical properties are brought into play in the organism–environment relation during prepupal and pupal life, may operate in other aspects of that relation. Examples are the production, this time by adult insects, of the silk forming the egg-protecting ootheca, or the production of a spider’s web. The term fibroin is generally used to designate scleroproteins very rich in glycine, alanine and serine. The fibroin of the silkworm cocoon remains [446] after the sericin coating the fibroin thread has been dissolved, for example in boiling water. The thread is the perfect example of an extended polypeptide chain. Its mechanical strength is due to the presence of the sequence

Gly–Ala–Gly–Ala–Gly–{Ser–Gly–(Ala–Gly)ₙ}₈–Ser–Gly–Ala–Ala–Gly–Tyr

which may be repeated several times, the fragments constituted by this sequence being separated by regions containing all the protein amino acids.^[T4]

Rudall (see Rudall, 1962) proposes delimiting the category of fibroins on the basis of X-ray patterns and contrasting them in this respect with keratins or collagens. This classification also agrees with the chemical classification of fibrous proteins: the fibroin type appears when approximately 50 per cent of the total amino acids consists of glycine, alanine and serine, whereas the keratin type appears when small amino acids constitute less than 50 per cent of the total and the cystine content is high.

The collagen configuration appears when a higher proline and hydroxyproline content is added to the high proportion of alanine–glycine–serine. These categories of fibrous proteins correspond to different adaptive radiations of fibrous-protein biosynthesis, and their relations to phylogeny are illustrated by several examples. Keratins are found only in vertebrates. The use of collagen as an intercellular constituent is particularly developed in the deuterostome branch. Fibroins occur in insects and spiders. The structure of Bombyx mori fibroin is one of the bases for defining the parallel-β type, in which extended polypeptide chains are aligned parallel to the longitudinal axis of the fibre. Rudall discovered another type, with extended chains running transversely to the fibre axis. This form was demonstrated in the silk of the egg stalk of Chrysopa flava.

Astbury and Street (1931) had suggested the existence of a [447] compact rather than extended form of silk (the α form), and Rudall found this form, but only in one group of Hymenoptera, the Aculeata—for example, the honeybee. These are fibroins spun by larvae, as in lepidopteran cocoons. Another interesting adaptation occurs in the cocoon of the gooseberry sawfly Nematus ribesii, whose larval silk glands are almost full of liquid procollagen. This collagen contains proline but no hydroxyproline.

One interest of studying fibroins lies in their constituting a transitional form between the keratin–epidermin–myosin–fibrin group (K–M–E–F) and the collagen group. As Rudall (1962) stresses, incorporating proline into a fibroin will yield collagen, whereas incorporating long-molecule amino acids into a fibroin will yield a fibrous protein of the K–M–E–F group. Better knowledge of amino-acid sequences will undoubtedly soon provide suggestive evidence on these aspects of the biochemical evolution of fibrous proteins. More detailed consideration of scleroproteins such as molluscan conchiolins, octocorallian gorgonins and hexacorallian antipathins is of undeniable interest concerning possible protein modification along phylogenetic branches.

Fibroins vary very widely in composition, but, as we have said, they are accepted as containing high proportions of the sum glycine–alanine–serine, as collagen also does, whereas, unlike collagen, fibroins contain little proline. According to Lucas, Shaw and Smith (1960), the sum of glycine, alanine and serine, determined by Moore and Stein’s chromatographic method in many different fibroins, ranges—in terms of amino nitrogen per 100 g of protein nitrogen—from 42.6 per cent (Arctia caja) to 94.6 per cent (Anaphe moloneyi). Examination of their Table 2 shows that including serine in this large component is very arbitrary, since the serine nitrogen contents of the fibroins considered in the same table range from 0.7 (Anaphe [448] venata) to 40.7 per cent (Chrysopa flava). Lucas, Shaw and Smith’s Table 2 (1960) shows that the fibroins of the Saturniinae species they studied have a higher alanine than glycine content. According to the same table, the fibroins of these Saturniinae display high values for the sum glycine–alanine–serine, without large divergences for each of these amino acids. As Duchâteau, Florkin and Leclercq (1960) pointed out, the sum of glycine and alanine is particularly high in species of the genus Anaphe, which is confirmed by Lucas, Shaw and Smith’s Table 2 (1960). For the glutamic-acid content of lepidopteran fibroins, the values obtained by chromatographic and microbiological methods agree. For the various species they range from 0.5 to 5.0 per cent (Schroeder and Kay, 1955; Lucas, Shaw and Smith, 1957; Kirimura, 1957; Duchâteau, Florkin and Leclercq, 1960; Lucas, Shaw and Smith, 1960). By contrast, the fibroins of spiders of the genus Nephila, studied by Lucas, Shaw and Smith (1960), have a higher glutamic-acid content.

Florkin and Bricteux-Grégoire (1961) also applied the chromatographic method to the study of the fibroin of Psenulus concolor (Hymenoptera; Sphecidae). This fibroin forms the lid with which Psenulus closes the nest it builds in hollow bramble and elder stems. It should be noted that this is a fibroin secreted by an adult hymenopteran, whose secretory organ is unknown. It is the first analysis of a hymenopteran fibroin by Moore and Stein’s chromatographic method. This fibroin contains a high proportion of the three amino acids glycine–alanine–serine, resembling lepidopteran fibroins in that respect. It nevertheless has a higher glutamic-acid content than the latter, a feature also found in the fibroin of spiders of the genus Nephila.

The physical aspects involved in the organism–environment relation, which differ according to the character of that relation in different zoological groups, obviously originate in the particular metabolism of each group. The definition of the product of synthesis, with its particular characteristics of [449] structure, mechanical strength, insolubility and so forth, depends on the one hand on synthesis at a template constituted by a ribonucleic acid and, on the other, on making amino acids available to the biosynthetic mechanism in cells differentiated towards the synthesis of a scleroprotein such as keratin or fibroin. In the silkworm, fibroin is specifically synthesised by the posterior portion of the silk gland, which receives the materials for synthesis through the haemolymph. Their supply partly determines fibroin production and its adaptive nature.

It is important to determine the correspondence that can be demonstrated between the nature of the organic compounds present in the haemolymph during the fifth instar and the different segments of the long thread forming the cocoon. This study was carried out by T. Fukuda and his Japanese collaborators (1959) for the hybrid of Chinese and Japanese strains used in Japan, and by Fukuda and Florkin (1959a) for the Europe strain.

As we shall see, alanine is not one of the haemolymph amino acids taken up in large quantities by the gland for fibroin synthesis. Nevertheless, since this synthesis incorporates a high proportion of alanine and small quantities of it are absorbed by the gland, the fibroin synthesised will be labelled if the haemolymph contained labelled alanine on the day of its synthesis.

Each day for twelve days from the beginning of the fifth instar, individuals from the same rearing batch received 0.5 μCi of L-alanine-¹⁴C through the digestive tract. The different silkworms receiving this alanine were fed mulberry leaves as usual and spun their respective cocoons. Each cocoon’s thread was unwound using a rotating drum of known circumference fitted with a revolution counter, and each thread was cut into ten equal lengths. Each section was degummed by repeated boiling in an M/50 solution of sodium carbonate,^[T9] washed with ethanol and ether, and dried. The ten sections thus freed of sericin were arranged side by side along a horizontal line on a sheet of card, [450] each line beginning with the anterior end of the thread—that is, the end spun earliest. The horizontal series thus obtained for the different cocoons were aligned vertically in the order of the days on which labelled alanine had been administered after the beginning of the fifth instar. A radiographic film was placed against the card and kept in contact with it for eighteen days.

From these autoradiographic investigations, it can be concluded that the origin of the carbon atoms in a portion of silk thread can be related to the carbon derivatives present in the haemolymph when that portion was formed. If so, in the as-yet-unsolidified fibroin (fibroinogen) accumulated in the reservoir, the macromolecules must be arranged in a definite spatial order corresponding to the chronology of their biosynthesis. Fukuda and Florkin (1959b) confirmed this notion by showing with glycine-¹⁴C that successively biosynthesised portions are deposited in the reservoir in the chronological order of biosynthesis.

Furthermore, under these conditions the successively synthesised portions of fibroin must move within the reservoir under pressure from new portions of the same protein. This is indeed what appears in studies of longitudinal or transverse sections of the reservoir and its contents at different times after introducing ¹⁴C on the 120th day after the fourth moult (Fukuda and Florkin, 1959c).^[T5] The portions of fibroinogen successively spun into solid fibroin therefore emerge in the temporal order of fibroinogen synthesis.

From the day on which the gland begins synthesising fibroin and drawing the materials for this synthesis from the haemolymph—the sixth day in the Europe strain, the third in the hybrid of Chinese and Japanese strains—a large proportion of the amino acids introduced with food is taken up directly by the gland. If these amino acids were uniformly distributed throughout all tissues and the materials for fibroin synthesis were endogenous in origin, the thread would be uniformly labelled. The silk gland’s avidity for amino acids introduced into the haemolymph is further illustrated by the fact that, just an hour and a half after glycine-¹⁴C is introduced into the haemolymph [451] of a silkworm during its silk-producing period, no trace of radioactivity is found in the haemolymph proteins, whereas the radioactivity of the posterior portion of the gland is high.

Table I.

Table II. (Duchâteau-Bosson et al., 1961b).

Table I references: (1) Duchâteau et al., 1959. (2) Bricteux-Grégoire et al., 1959d. (3) Duchâteau-Bosson et al., 1960. (4) Duchâteau-Bosson et al., 1961a.

[452] By comparing free amino-acid concentrations in normal silkworms and those whose silk gland had been excised (Tables I and II), it was found that the principal amino acids taken up by the gland from the haemolymph after the sixth day of the fifth instar are glycine, glutamic acid, serine and tyrosine. Phenylalanine is not taken up in large quantities, whereas tyrosine is. It is therefore important to determine whether phenylalanine can be converted into tyrosine during the silkworm’s fifth instar. This is shown by Table III, which presents results obtained in silkworms injected with L-phenylalanine labelled at C-1. The fibroin tyrosine was carefully isolated from the cocoon and its purity checked. Fukuda (1956) also obtained labelled tyrosine after injecting phenylalanine labelled at C-2. The same author claimed that fibroin alanine can derive from phenylalanine. This is not what we observed. After careful isolation of alanine, glycine or serine, no radioactivity is observed in animals that received phenylalanine labelled at C-1.

Table III. (Bricteux-Grégoire et al., 1959a). Specific activities. Counts/min/mole.

As already noted, alanine is not removed from the haemolymph in appreciable quantities by the gland. It nevertheless constitutes one of the most abundant amino acids in fibroin, which consists of 43 per cent glycine, 35 per cent alanine and 16 per cent serine. Experiments on glandectomised silkworms clearly show that glycine and serine are removed from the haemolymph in large quantities, as is tyrosine formed in the silkworm’s body from phenylalanine. This is not the case for [453] that other important constituent of fibroin, alanine. The latter is synthesised within the gland, as are some of the fibroin’s glycine and serine, but not its tyrosine.

To summarise, silk fibroin results from incorporation into the protein synthesis characteristic of the posterior portion of the silk glands of alanine synthesised in that posterior portion, glycine and serine partly drawn from the haemolymph and partly synthesised in the gland, and tyrosine entirely drawn from the haemolymph and synthesised from phenylalanine.

How do the biosynthetic processes contributing alanine, glycine and serine to fibroin synthesis operate in the gland? In rats and pigeons, as in microorganisms, glycine can be used for serine biosynthesis. The metabolic pathways through which pyruvate gives rise to alanine by amination in mammals are also known. If glycine labelled in the carboxyl group is given to a silkworm, and glycine, serine and alanine are isolated from the fibroin, they are found labelled only at C-1. If glycine-2-¹⁴C is given, very high activity is found at C-2 of glycine and serine, presumably through the addition of formate. As in the rat, glycine C-2 can be a precursor of serine C-3. This results from formate addition. C-1 of glycine, serine and alanine is also labelled following administration of glycine-2-¹⁴C, as a consequence of passage through pyruvate.

Table IV shows the activities found in glycine, serine and alanine after introducing glycine-1-¹⁴C and glycine-2-¹⁴C. This table shows that approximately one-quarter to one-half of the injected glycine is actually incorporated into fibroin in the forms of glycine, alanine and serine. In the case of alanine, biosynthesis occurs chiefly in the silk gland. Above, we postulated incorporation of formate as the source of serine C-3. Table V confirms this. The table shows some specific activity at serine C-1 and C-2. This is [454] readily explained: pyruvate derived from serine degradation is labelled at C-3, but during passages through the Krebs cycle it becomes uniformly labelled, and alanine derived from pyruvate will also be uniformly labelled. Serine will be obtained from this pyruvate through hydroxypyruvic acid.

Table IV. (Bricteux-Grégoire et al., 1959b). Specific activities of the isolated amino acids and their degradation products. Counts/min/mmole.

Table V. (Bricteux-Grégoire et al., 1959c). Percentage of formate used for the synthesis of amino acids of silk fibroin.

[455] Table VI. (Bricteux-Grégoire et al., 1960). Specific and total activities of amino acids isolated from fibroin. ¹ Counts/min/mmole. ² Counts/min per cocoon.

DL-Aspartic acid-3-¹⁴C: 755 μg (1.78 × 10⁹ counts/min/millimole). Total activity injected: 10 100 000.

DL-Glutamic acid-3,4-¹⁴C: 350 μg (4.57 × 10⁹ counts/min/millimole). Total activity injected: 10 900 000.

DL-Glutamic acid-2-¹⁴C: 1.27 mg (1.23 × 10⁹ counts/min/millimole). Total activity injected: 10 600 000.

[456] Table VII. (Bricteux-Grégoire et al., 1960). Distribution of activity in isolated alanine, serine and glycine (counts/min/millimole).

Injected amino acid: Aspartic acid-3-¹⁴C.

Injected amino acid: Glutamic acid-3,4-¹⁴C.

Injected amino acid: Glutamic acid-2-¹⁴C.

As Tables VI and VII show, the ¹⁴C of aspartate and glutamate is incorporated to varying degrees into the major amino acids of fibroin, with the exception of tyrosine. It appears that silk alanine synthesis derives to a significant extent from the dicarboxylic acids, which are the most abundant constituents of the mulberry leaf capable of supplying the gland, both as free amino acids [457] and as their amides, with the pyruvate and nitrogen feeding these syntheses (amino acids).

Alanine can be formed in the posterior portion of the silk gland by transamination onto pyruvate. This is demonstrated by the experiments of Fukuda (1957a, b) and Koide (1955). Bricteux-Grégoire et al. (1959e), using pyruvate labelled with ¹⁴C at various positions, showed that when pyruvate-1-¹⁴C is introduced orally into the silkworm’s body, ¹⁴C appears at C-1 in the alanine, serine and glycine of fibroin isolated from the cocoon. When pyruvate-2-¹⁴C or pyruvate-3-¹⁴C is given, activity appears at all positions in alanine, serine and glycine (Table VIII). Administration of various labelled forms of glucose also shows that the pyruvate used in the gland for biosynthesis of fibroin alanine, glycine and serine is drawn from a pyruvate pool originating not only in glycine, serine, glutamic acid and aspartic acid, but also in other metabolic sources (Table IX). Tyrosine, however, derives only from exogenous phenylalanine (Bricteux-Grégoire et al., 1949a).^[T6]

These results are confirmed by the study of balances established in reared silkworms receiving abundant protein food (Fukuda et al., 1961).

[458] Table VIII. (Bricteux-Grégoire et al., 1959e). Specific activities of isolated amino acids and their degradation products (counts/min/mole).

Ingested labelled compound: Pyruvate-1-¹⁴C.

Ingested labelled compound: Pyruvate-2-¹⁴C.

Ingested labelled compound: Pyruvate-3-¹⁴C.

[459] Table IX. (Bricteux-Grégoire and Florkin, 1962). Specific and total activities of amino acids isolated from fibroin. ¹ Disintegrations/min/mmole. ² Disintegrations/min per cocoon.

Injected compound: Glucose-1-¹⁴C. Total activity injected: 2.22 × 10⁷ disintegrations/min.

Injected compound: Glucose-2-¹⁴C. Total activity injected: 2.22 × 10⁷ disintegrations/min.

Injected compound: Glucose-6-¹⁴C. Total activity injected: 2.22 × 10⁷ disintegrations/min.

During studies on glandectomised silkworms, an increase in haemolymph content had been observed not only for the fibroin-forming amino acids mentioned above—glycine, serine, aspartic acid, glutamic acid and tyrosine—but also for proline and threonine (Duchâteau-Bosson et al., 1960). Silkworm genetics has not so far revealed mutants with an abnormal fibroin composition. One mutant (the Nd strain) is known that produces practically no fibroin—0.6 per cent of cocoon weight according to Watanabe (1959)—but synthesises sericin of normal composition and quantity. Cocoons spun by Nd silkworms are consequently lighter and more transparent than those of normal silkworms. As we have said, fibroin is secreted [460] by the posterior portion of the silk gland, whereas sericin is synthesised by the wall of the reservoir constituting the middle portion of the gland. In the Nd strain, the posterior portion appears not to function in taking up from the haemolymph the constituents used in fibroin biosynthesis. In Nd silkworms, however, although an increase in the concentration of fibroin-forming amino acids is observed, as in glandectomised normal animals, no increase in threonine or proline is observed. Threonine and proline are taken up from the haemolymph at the reservoir (Bricteux-Grégoire et al., 1964).

The formation of the silkworm cocoon appears to consist in the secretion of a protective envelope within which the prepupal moult and pupal life take place. Fibroin, the mechanically important constituent of the cocoon, derives from a specific food, mulberry leaves, rich in glutamic acid, glutamine, aspartic acid and asparagine. This highly specific food is consumed in great excess during the silk gland’s period of activity and supplies the gland with a series of amino acids that enter its cells through selective absorption, during which the posterior portion of the gland becomes enriched in glycine, serine, aspartic acid and asparagine, glutamic acid and glutamine. The gland also absorbs tyrosine resulting from the transformation of dietary phenylalanine. Fibroin’s particular structure, with its sections formed from small amino acids, depends on the genetic code releasing messenger RNAs that is peculiar to the cells of the posterior portion. The amino acids forming the greater part of fibroin—alanine, glycine, serine and tyrosine—are supplied from the haemolymph in the case of tyrosine and some glycine and serine. Alanine is synthesised in the gland cells, as are some glycine and serine. Glutamic and aspartic acids, like their amides, are among the important suppliers of the pyruvate and nitrogen required for these syntheses.

During the gland’s period of activity, a large part [461] of the dietary amino acids is directed straight to the silk gland. In the final period of the fifth instar, when the last portions of the fibroin thread are synthesised, biosynthesis is supplied endogenously at the expense of the tissues, and this process contributes to regulation of prepupal weight.

The adaptive nature of the cocoon is expressed not only in the mechanical properties of the fibroin forming the silk thread, but also in the properties of the sericin associated with that thread in the cocoon and secreted by the reservoir during spinning.

At emergence, another biochemical adaptation operates at the cocoon, allowing the moth to escape. This adaptation depends on the nature of sericin, the natural substrate of a proteolytic enzyme secreted by the moth that selectively dissolves it, allowing the fibroin threads to be parted and an opening to be made through which the adult can pass (Fafatos and Williams, 1964).

We have reported above the results of a very detailed examination of the molecular aspects of the adaptation represented by cocoon spinning in the silkworm.

The physical properties of fibroin underlying the cocoon’s protective function depend on the particular features of protein synthesis in the posterior portion of the silk gland. Fibroin’s structure appears to be explained by the particular nature of the messenger RNA released in the cells of the posterior portion. This is doubtless true, at least within our present conceptions, and may appear to account for the adaptation under consideration. The detailed analysis reported above nevertheless teaches us that, although the explanation of the cocoon’s nature indeed lies at the level of a particular gene, the cocoon’s existence results from many other factors: the nature of the silkworm’s food; the behaviours attracting it to this narrowly defined food and inducing it to chew and then swallow it; the behaviour triggering the remarkable appetite characteristic of the “great feeding” [grande frèze];^[T7] [462] the characteristics of amino-acid uptake by the gland from the haemolymph; regulation of pupal weight; and so forth. All these factors control the very existence of the cocoon, reveal the distinctly polygenic character of the adaptation under consideration, and can be brought to light only by a molecular approach—provided, however, that the investigation starts at the level of the organism.

Bibliography#

Astbury, W. T. and Street, A. Phil. Trans. Roy. Soc., 1931, A230, 75.

Bennett, T. P. and Frieden, E. In Comparative Biochemistry (eds. M. Florkin and H. S. Mason), New York, Academic Press, vol. 4, 1962, p. 483.

Bricteux-Grégoire, S., Dewandre, A. and Florkin, M. Biochem. Zeitschr., 1960, 333, 370.

Bricteux-Grégoire, S., Dewandre, A., Florkin, M. and Verly, W. G. Arch. Internat. Physiol. Bioch., 1959c, 67, 687.

Bricteux-Grégoire, S., Dewandre, A., Florkin, M. and Verly, W. G. Arch. Internat. Physiol. Bioch., 1959b, 69, 693.

Bricteux-Grégoire, S., Duchâteau, Gh., Florkin, M. and Jeuniaux, Ch. Arch. Internat. Physiol. Bioch., 1959d, 67, 586.

Bricteux-Grégoire, S., Duchâteau-Bosson, Gh., Jeuniaux, Ch. and Florkin, M. Arch. Internat. Physiol. Bioch., 1964, 72, 489.

Bricteux-Grégoire, S. and Florkin, M. Arch. Internat. Physiol. Bioch., 1962, 70, 711.

Bricteux-Grégoire, S., Fukuda, T., Dewandre, A. and Florkin, M. Arch. Internat. Physiol. Bioch., 1959e, 67, 545.

Bricteux-Grégoire, S., Verly, W. G. and Florkin, M. Arch. Internat. Physiol. Bioch., 1959a, 67, 563.

Brown, G. W. and Cohen, P. P. In A Symposium in the Chemical Basis of Development (eds. W. D. McElroy and B. Glass), Baltimore, Johns Hopkins, 1958, p. 495.

Duchâteau, Gh., Florkin, M. and Jeuniaux, Ch. Arch. Internat. Physiol. Bioch., 1959, 67, 173.

Duchâteau-Bosson, Gh., Florkin, M. and Jeuniaux, Ch. Arch. Internat. Physiol. Bioch., 1960, 69, 327.

Duchâteau, Gh., Florkin, M. and Leclercq, J. Arch. Internat. Physiol. Bioch., 1960, 68, 190.

[463] Duchâteau-Bosson, Gh., Jeuniaux, Ch. and Florkin, M. Arch. Internat. Physiol. Bioch., 1961a, 69, 369.

Duchâteau-Bosson, Gh., Jeuniaux, Ch. and Florkin, [M.]. Arch. Internat. Physiol. Bioch., 1961b, 69, 485.

Fafatos, F. C. and Williams, C. M. Science, 1961, 146, 538.^[T8]

Florkin, M. and Bricteux-Grégoire, S. Arch. Internat. Physiol. Bioch., 1961, 69, 46.

Fukuda, T. J. Bioch. (Jap.), 1956, 43, 137.

Fukuda, T. J. Bioch. (Jap.), 1957a, 44, 505.

Fukuda, T. Nature, 1957b, 180, 245.

Fukuda, T., Duchâteau-Bosson, Gh. and Florkin, M. Arch. Internat. Physiol. Bioch., 1961, 69, 701.

Fukuda, T. and Florkin, M. Arch. Internat. Physiol. Bioch., 1959a, 67, 185.

Fukuda, T. and Florkin, M. Arch. Internat. Physiol. Bioch., 1959b, 67, 190.

Fukuda, T. and Florkin, M. Arch. Internat. Physiol. Bioch., 1959c, 67, 214.

Fukuda, T., Kirimura, J., Matuda, M. and Suzuki, J. J. Bioch. (Jap.), 1955, 42, 341.

Jeuniaux, Ch. and Florkin, M. Arch. Internat. Physiol. Bioch., 1958, 66, 552.

Kirimura, J. Protein Chemistry, 1957, vol. 5, Tokyo, Kyoritsu (cited from Lucas et al., 1960).

Koide, F., Nagayama, H. and Shimura, K. Nippon Nogei-Kagaku Kaishi, 1955, 29, 987.

Lucas, F., Shaw, J. T. B. and Smith, S. G. Nature, 1957, 179, 906.

Lucas, F., Shaw, J. T. B. and Smith, S. G. J. Molecular Biol., 1960, 2, 339.

Nishikawa, S. and Ono, S. Proc. Tokyo Math. Phys. Soc., 1913, 27, 131.

Rudall, K. M. In Comparative Biochemistry (eds. M. Florkin and H. S. Mason), New York, Academic Press, vol. 4, 1962, p. 397.

Schroeder, W. A. and Kay, L. M. J. Am. Chem. Soc., 1955, 77, 3908.

Watanabe, T. J. Seric. Sci. Japan, 1959, 28, 251.