INTRODUCTION

Being a matter of great interest from the standpoints of developmental biology, genetics and evolution, the wing color patterns of butterflies have been investigated by many researchers. The patterns have been examined and compared in many kinds of species and also in various kinds of geographical, genetical and seasonal variations. In addition, perturbation experiments such as micro-cautery, micro-excision, and low or high temperature treatments have been made, as reviewed in detail by Nijhout (1991). More recently, new experimental techniques have been used to estimate the enzyme activities in presumptive pattern regions (Koch et al., 1998, 2000) and to analyze the appearance of wing patterns using molecular biological methods (Carroll et al., 1994; Beldade et al., 2002).

Various pigments in the wing of butterflies have also been studied (Needham, 1978; Umebachi, 2000). In this way, melanin, pterins, and ommochromes were found in the wings and identified. Umebachi (1985) isolated papiliochromes, new yellow pigments from the wings of a swallowtail butterfly, Papilio xuthus. One of them, papiliochrome II, was characterized and proved to be a combination of N-β-alanyldopamine and L-kynurenine.

The present paper is the first to show that external administration of an acid carboxypeptidase to early pupae produces various blackened or yellow wing-types without any modification to the shape or veins of wings. The occur-rence and type of such perturbed wing color patterns depend upon the developmental stage of the treated pupae. These results should thus help to elucidate the mechanism of color pattern formation.

MATERIALS AND METHODS

Injection

After Toughmac-E and Molsin were solubilized to water in concentrations of 4 mg per ml and 1.5-3 mg per ml, respectively, one of them was administered by hypodermal injection to one prepupa or pupa. Likewise, L-dopa solution (1.3 mg per ml) or dopamine solution (3 mg per ml) was used. As to Inosea, only the central part of the tablet was used, and the water solution was prepared in the same way as in Toughmac-E. The injection of 50 μl of each prepared solution was given into the body cavity under the left white patch of the third abdominal segment of the prepupa (Fig. 1a) or under the outer margin of the left wing of the 0- to 4-day-old pupa (Fig. 1b).

Fig. 1

Injection site of prepupa (a) or pupa (b).

RESULTS

I. Administration of Toughmac-E and Molsin by hypodermal injection

A. Administration to prepupa

The greater region of the normal wing pattern of adult P. xuthus emerged from untreated prepupae (pharatepupae) and pupae was made up of mainly black and yellow colors (Fig. 2). The yellow tone of the underside was a little stronger than that of the upperside. The male was a paler yellow than the female.

Fig. 2

Normal wing color patterns of P. xuthus. Left wings of the (a) male and (b) female show the undersides, and right wings the upper-sides. The size of the wings is reduced to 75% of the original.

Compared with the normal pattern, there were no variations in color pattern throughout the entire adult wing of individuals administered Toughmac or Molsin as prepupae (data not shown).

B. Administration to 0-day-old pupae

At the time when Toguhmac-E or Molsin was injected into pupae just after pupation, no essential difference of the effects on the wing color pattern was found between these two substances. The upperside of the forewing (forewing recto, f.w.r.) appears normal, but the submarginal, pale yellow band of the underside (forewing verso, f.w.v.) broadens inwards (Fig. 3). A submarginal, pale yellow fleck line on the hindwing upperside (h.w.r.) develops well and distinctly to form a broad band. This tendency is especially clear with a submarginal, yellow band on the hindwing underside (h.w.v.): the inner black band either separates (Fig. 3, a and c) or is substantially reduced (Fig. 3b), causing the appearance of blue flecks and orange areas which were originally covered with the black band in the normal pattern. Such wing color patterns are designated Y (yellow) -type.

Fig. 3

Yellow-type (Y-type) wings obtained by injection into 0-day-old pupae: (a) male from the Molsin-injected pupa; (b) and (c) males from Toughmac-E-injected pupae. Left wings show the undersides, and right wings the uppersides. The size of the wings is reduced to 75% of the original.

C. Administration to 1-day-old pupae

At the time when Toughmac-E or Molsin was injected into 1-day-old pupae, two types of adult butterflies with abnormal wing color patterns, B-Y (black-yellow) -type (Fig. 4) and B (black) -type (Fig. 5b-5d) were obtained. C-1. Color patterns of B-Y type

Fig. 4

Black and yellow-type (B-Y type) wings obtained by injection into 1-day-old pupae: (a) Molsin, and (b) Toughmac-E were injected into a male and a female, respectively. Left wings show the undersides, and right wings the uppersides. The size of the wings is reduced to 75% of the original.

Fig. 5

Blackened wings (B-type) obtained by injection into young pupae. Sex, species of enzyme injected and age are: (a) male, Molsin into 2-day-old pupa; (b) male, Toughmac-E into 1-day-old pupa; (c) female, Molsin into 1-day-old pupa; and (d) female, Toughamac-E into 1-dayold pupa. Left wings show the undersides, and right wings the uppersides. The size of the wings is reduced to 75% of the original.

On the f.w.r. of both sexes, the pale yellow region was broadly enlarged except for black lines on most veins and around the wing margin, and a large, triangular black marking appears above the inner margin (Fig. 4a and 4b). A similar tendency was also observed in the f.w.v., but the yellow region broadens more strongly than that of the f.w.r., while the triangular black marking is substantially reduced.

On the h.w.r. of the male, the black region is well developed and occupies the majority part (Fig. 4a). In this black region, the submarginal, pale yellow fleck line has a tendency to broaden inwards and separates into two lines: a distinct, original submarginal line and another newly formed, obscure discal line, on which there are faint blue scales. On the female, a perturbation pattern fundamentally similar to that of the male is seen (Fig. 4b), but the enlargement of melanization is much less and its intensity is much weaker. The broadened black region does not reach the prediscal, postbasal or basal parts. On the h.w.v. of the male, not only is the yellow region enlarged, but the black region is more markedly reduced than that of the h.w.r. A postdiscal black band of the normal wing shifts more inwards, and an irregular, discal blue fleck line appears on the black region. The perturbation pattern of the female (Fig. 4b) seems to be similar to that of the male (Fig. 4a), but the yellow region broadens more strongly, whereas the black region has a tendency to be smaller and thinner.

Melanization develops markedly in the whole wing, particularly in the f.w.r. (Fig. 5d) and h.w.r. (Fig. 5b and 5c).

D. Administration to 2-day-old pupae

At the time when Molsin was injected into 2-day-old male pupae, the blackend region developed markedly in both sides of the forewing as well as the h.w.r. (Fig. 5a). In contrast, the yellow region developed in the h.w.v., where the pattern was rather similar to that of the normal.

E. Administration to 3- and 4-day-old pupae

Injection of Toughmac-E or Molsin into 3- or 4-day-old pupae did not have any influence on the normal wing color pattern (data not shown). This is the same in injection into prepupae, as shown in experiment A. We therefore concluded] that these substances probably do not perturb the color pattern when injected into pupae of 5-days or older.

DISCUSSION

The black and pale yellow markings in the wing of P. xuthus result from deposit of the corresponding pigments, melanin and papiliochromes, in scales, whereas the blue is a structural color originating from Thyndall blue in black scales (Nijhout, 1981). Melanin is probably derived from dopamine (Koch et al., 1998; 2000), while papiliochrome II consists of L-kynurenine and N-β-alanyldopamine (Rembold and Umebachi, 1984; Umebachi, 1985, 1993). Therefore, in Papilio butterflies, both black and yellow pigments necessitate dopamine as the common precursor, so that a balance between its supply and utilization in scales must influence the color pattern. L-Kynurenine is derived from tryptophan, and N-β-alanyldopamine is from dopamine and β-alanine formed from uracil (Osanai et al., 1994), respectively.

In addition, various enzymes are included in the formation of pigments and their precursors in scales, as shown in the synthesis of papiliochrome II (Yago, 1989; Soul and Sugumaran, 1991). At the formation of the wing color pattern of another papilionid butterfly, P. glaucus, the included enzyme species as well as the timing of the appearance of their activity are different between the presumptive black and yellow regions in wings (Koch et al., 1998, 2000). Even in the same enzyme species, the period of the appearance of the activity varies between these two presumptive regions. Moreover, the growth rate of black scales is also different from that of yellow scales, probably due to the different time program in the presumptive black scales. These internal, physiological conditions of enzymes probably also rule the formation mechanism of the color pattern of P. xuthus.

Molsin, an acid carboxypeptidase (EC 3.4.23.18) of Aspergillus saitoi has a capacity to liberate tyrosine and phenylalanine at the C-terminal residues of protein and peptide at low pH (Yoshida and Nagasawa, 1956). Therefore, the pattern perturbation induced by Molsin should be attributed not only to the quantitative changes in melanin and papiliochromes but also to the varied degree of formation of colored wing scales. It is possible that a similar, endogenous enzyme, the action of which is the same as that of Molsin, probably another acid carboxypeptidase, functions in P. xuthus for supply of dopamine, the precursor indispensable for formation of melanin and papiliochromes. Since the effective period for the pattern perturbation induced by Molsin was strictly limited from day 0 to 2 of pupa, the activity of the endogenous acid carboxypeptidase of this butterfly must appear only during this period. Overhigh activity based on the summation of these endogenous and exogenous acid carboxypeptidases probably causes the abnormal changes in the wing color pattern.

In the Y-type, the pattern of the f.w.r. and f.w.v. obtained from the 0-day-old pupae treated with Molsin showed almost no variation from the normal pattern, such that this exogenous enzyme did not yet sufficiently influence the pattern changes. In contrast, the enhancement of the yellow region of both sides of the hindwing suggests that dopamine derived from tyrosine or phenylalanine freed by the acid carboxypeptidase may be utilized for the synthesis of papiliochrome much more strongly. In the types B-Y (Fig. 4) and B (Fig. 5b–5d) which were obtaind from 1-day-old pupae treated with Molsin, the balance between the formation of the two pigments was not stable: sometimes melanin-rich and papiliochrome-poor pattern occurred, and sometimes the reverse pattern occurred. In the B-type (Fig. 5a) obtained from the 2-day-old pupa treated with Molsin, compared with the B-type from 1-day-old pupae, the pattern perturbation exhibited a tendency to return to normal. Namely, at this stage, the exogenous enzyme did not as strongly affect the pattern perturbation, in particular in the h.w.r. and h.w.v. Probably, the summed overhigh activity was decreased with the reduced activity of endogenous acid carboxypeptidase, because the dose of enzyme exogenously given to pupae was always the same in all the stages used. It is, therefore, concluded that in the normal butterfly of P. xuthus, the activity of the endogenous acid carboxypeptidase must appear in the early pupal stage. This activity peaks in the 1-day-old pupa, and is lower at the earlier or later pupal stage. The higher the activity of endogenous acid carboxypeptidase in the pupa is, the greater the quantities of tyrosine and phenylalanine that are liberated.

Unlike vertebrates, insects are unable to easily catabolize aromatic amino acids. For example, histidine is almost not metabolized during the pupal stage of Bombyx mori (Osanai and Kikuta, 1981). In contrast, phenylalanine, tyrosine and tryptophan could be metabolized and polymerized to various pigments (Koga and Osanai, 1967; Neehdam, 1978; Kayser, 1985; Umebachi, 2000). Pigments accumulated in the wing scales of butterflies, which are related to the excretion of nitrogenous substances, are regarded as storage excretory materials in “kidneys of accumulation” (Cochran, 1985).

The substrate of the endogenous acid carboxypeptidase is probably the reserve protein synthesized in the larva as a source of amino acids for the de novo syntheses of various enzymes and structural proteins in the pupa and adult (Osanai and Kikuta, 1981). By removal of aromatic amino acids from the C-terminal residues of the original reserve protein in the pupa, this protein is reconstructed to a more utilizable protein. It has been reported that an Arg-C endopeptidase causes the second maturation of spermatozoa of the silkworm, together with the energy-yielding system from arginine (Osanai et al., 1987; Osanai, 1992) by cooperation with an Arg-carboxypeptidase (Aigaki et al., 1988). Like these proteases, the endogenous acid carboxypeptidase of P. xuthus thus plays another important role in the regulation, utilization and excretion of amino acids.

Last but not least, it is interesting that the administration of L-dopa or dopamine did not cause any perturbation of wing color pattern. The difference of action between Molsin and dopa or dopamine is a subject for future research.