Convergent evolution of neuroendocrine control of phenotypic plasticity in pupal colour in butterflies

Phenotypic plasticity in pupal colour occurs in three families of butterflies (the Nymphalidae, Papilionidae and Pieridae), typically in species whose pupation sites vary unpredictably in colour. In all species studied to date, larvae ready for pupation respond to environmental cues associated with the colour of their pupation sites and moult into cryptic light (yellow–green) or dark (brown–black) pupae. In nymphalids and pierids, pupal colour is controlled by a neuroendocrine factor, pupal melanization-reducing factor (PMRF), the release of which inhibits the melanization of the pupal cuticle resulting in light pupae. In contrast, the neuroendocrine factor controlling pupal colour in papilionid butterflies results in the production of brown pupae. PMRF was extracted from the ventral nerve chains of the peacock butterfly Inachis io (Nymphalidae) and black swallowtail butterfly Papilio polyxenes (Papilionidae). When injected into pre-pupae, the extracts resulted in yellow pupae in I. io but brown pupae in P. polyxenes. These results suggest that the same neuroendocrine factor controls the plasticity in pupal colour, but that plasticity in pupal colour in these species has evolved independently (convergently).     

Pupation site preference and environmentally cued pupal colour dimorphism in the swallowtail butterfly Papilio polyxenes Fabr. (Lepidoptera: Papilionidae)

Environmentally cued polymorphisms are hypothesized to evolve when the environment is coarsegrained and different genotypes are unable to choose the habitats in which they are most fit. In Papilio polyxenes , which has an environmentally cued pupal colour dimorphism, there is genetic variation in both tendency to produce brown or green pupae and preference for green- or brown-inducing pupation sites, but the two traits are not correlated.     

The evolution of environmentally-cued pupal colour in swallowtail butterflies: natural selection for pupation site and pupal colour

1. Environmentally-cued pupal colour in swallowtail butterflies has been hypothesized to evolve as a consequence of (a) the evolution of a preference for pupation sites above the ground that vary in colour and (b) natural selection for crypsis on such sites.
2. This hypothesis was tested by comparing the field survival of green and brown Papilio polyxenes Fabr. pupae placed on green or brown pupation sites that were either above the ground on near the ground.
3. Green pupae on green sites above the ground had a significantly higher probability of survival than did all other pupal colour and pupation site combinations.
4. Pupae on sites above the ground were more likely to be preyed upon during the day, whereas those on sites near the ground were more likely to be preyed upon during the night, suggesting that variation in nocturnal and diurnal predation influences the evolution of pupation site preference.
5. To the extent that diurnal predators use colour vision to locate prey, diurnal predation should favour environmentally-cued pupal colour.     

Natural pupation sites of swallowtail butterflies (Lepidoptera: Papilioninae): Papilio polyxenes Fabr., P.glaucus L. and Battus philenor (L.)

Abstract. 1. Natural pupation sites have been found in Papilio polyxenes and P.glaucus by releasing prepupal larvae marked with UV-fluorescent paint and locating them at night with a UV lamp, and in Battus philenor by searching a forest habitat where the larval foodplant is abundant.
2. P.polyxenes , a species of weedy habitats, pupates off the ground on a variety of substrates including grasses, weed stalks, posts, etc. The pupae may be green or brown, resembling the substrate.
3. P.glaucus , a species of forest habitats, pupates very close to the ground in the litter and has monomorphic brown pupae.
4. B.philenor , also a forest species, pupates on exposed surfaces (chiefly tree-trunks or cliffs) well off the ground. Its pupae may be brown or green, but the latter were found only on the slenderest twigs.
5. The results for polyxenes and glaucus support the generalization of Clarke & Sheppard (1972) that species of stable habitats are likely to have monomorphic pupae, while those of habitats in which available sites may not be so similar from one generation to the next will be dimorphic.
6. B.philenor is more problematical, but its tendency towards pupal monomorphism (brown) is logical in relation to its common pupation sites.     

Environmental control of pupal colour in swallowtail butterflies (Lepidoptera: Papilioninae): Battus philenor (L.) and Papilio polyxenes Fabr.

Abstract. 1. Some swallowtail butterflies produce both green and brown pupae. The phenotypes result from the joint action of genotype and environment and usually make the pupae cryptic in their habitats.
2. The major environmental cues influencing pupal colour in two swallowtail species were determined to be textural and optical.
3. Differences in the usage of these kinds of cues in the two species are thought to have evolved because of major differences in the pupation habitats. P.polyxenes , which usually pupates on slender stems amidst vegetation, responds more strongly to optical cues. B.philenor , which usually pupates on exposed surfaces of tree trunks and cliffs, responds more strongly to textural cues.
4. Differences in the overall tendency to produce brown pupae ('sensitivity': Hazel, 1977) are thought to be related to the frequency of brown pupation sites utilized by these two species: high average sensitivity in philenor , which often uses brown sites, and lower average sensitivity in polyxenes , which often uses green sites.     

Two different sensory mechanisms for the control of pupal protective coloration in butterflies

The butterflies Graphium sarpedon nipponum Fruhstorfer and Papilio xuthus Linné show pupal protective color polymorphism, but the two species appear to have different sensory mechanisms for determining pupal coloration. When light was of sufficient illumination, the larvae of Graphium sarpedon became bright yellowish green pupae on white pupation boards and reddish brown pupae on black pupation boards. The pupal coloration thus strongly depended on the brightness of the pupation site. In addition, larvae became bright yellowish green pupae in complete darkness. From these results, measurement of the illumination suggested that pupal color is determined by the illuminant difference between incidence light from the dorsal direction and ventral light from a paper board; i.e., the sum of the reflected light of the board plus the penetrated light passing through the board. The illuminant difference required for reddish brown coloration was 40 lux or more. The optical signals received through the stemmata during a critical period before formation of the thorax garter (band string) were important for coloration. By contrast, in Papilio xuthus, successive tactile signals from a rough surfaced pupation site during a critical period before and after formation of the garter were important for determining brown pupal coloration.     

Interactions of environmental factors influencing pupal coloration in swallowtail butterfly Papilio xuthus

The swallowtail butterfly Papilio xuthus Linné [Lepidoptera: Papilionidae] exhibits pupal protective color polyphenism. Interactions of various environmental factors on pupal coloration were analyzed in non-diapausing individuals. Under sufficient light (200lux), most pupating larvae became green pupae when the surface of the pupation site was smooth, while they became brown when the surface was rough. Tactile signals are the positive environmental factors causing induction of the brown pupal coloration. In dark boxes, the induction of the brown pupal coloration was easily induced even on a smooth surface, suggesting that light suppresses induction of brown coloration. Different colors of pupation sites did not affect pupal coloration under sufficient light. Environmental factors received during a critical period both before girdling and after girdling affected pupal coloration. When tactile signals received from rough surfaces reach threshold levels during pupation, brown pupal coloration is determined. Larvae reared under a daily periodicity of natural light formed a girdle at midnight, subsequently, the prepupae received strong daylight the following day. Under natural light most larvae produced brown pupae on rough surfaces and green pupae on smooth surfaces.     

Extraction and partial purification of the pupal melanization reducing factor (PMRF) from Inachis io (Lepidoptera)

Pupae of Inachis io show a morphological color adaptation. The pigmentation is controlled by the background color during a sensitive period preceding pupation. From ligation experiments it has been concluded that the pigmentation is controlled by a factor from the head region, which gradually reduces the melanization.An extract was prepared from prepupal heads and prothoraces that reduces the normally strong melanization of pupae on a dark background. The active factor is preliminarily called a pupal melanization reducing factor (PMRF). The factor was extracted with water from isolated head and prothorax fragments of prepupae. It was partly purified by precipitations and chromatographic methods including HPLC. A biotest for PMRF was established based on the degree of pupal melanization. The nature of the PMRF activity seems to be a peptide, as shown by inactivation by the proteolytic enzymes pepsin and pronase. The molecular weight, as indicated by gel filtraction, is between 1000 and 5000.     

Factors affecting equal catchability in two swallowtail butterflies, Papilio polyxenes and P.glaucus

Abstract. 1. Recapture probabilities were analysed for individually marked black swallowtails, Papilio polyxenes F., and tiger swallowtails, P.glaucus L. (Lepidoptera: Papilionidae).
2. Recapture rates differed with sex and behaviour at time of capture for P.polyxenes , but not with age. For P.glaucus , only differences related to behaviour at time of capture were significant.
3. Black swallowtail males that were not physically restrained in identification had a recapture probability (73%) that was significantly higher than the 45% for males that were captured and handled. The higher rate was also exhibited by released, laboratory-reared males.
4. The capture effect was due to increased dispersal out of the areas and not to avoidance of capture or the capture site.     

Hormonal control of lutein incorporation into pupal cuticle of the butterfly Inachis and the pupal melanization reducing factor

Abstract. . Morphological colour adaptation of pupae of the butterfly Inachis io L. (Lepidoptera: Nymphalidae) is controlled by a factor which reduces cuticular melanization (Biickmann & Maisch, 1987). This so-called pupal melanization reducing factor (PMRF) is located throughout the entire central nervous system of prepupae (Stamecker et al. , 1994).
Extracts of abdominal ganglia also stimulated dose-dependently lutein incorporation into pupal cuticle. In the bioassay higher doses were required to increase cuticular lutein content than to reduce melanization. Ligatures during the prepupal stage demonstrated two different critical periods for these pigmentation effects: an early one for melanization reduction and a late one for lutein incorporation.
An initial chromatographic purification yielded only two adjacent fractions which contained both the PMRF and the stimulation of lutein incorporation activity. Therefore it is assumed that only one hormone with a dual function may be responsible for pupal pigmentation.
Lutein content was found in gut, fat body, epidermis and haemolymph of I.io. Lutein incorporation into cuticle occurred within 1.5 days of the pupal moult when the cuticle was not yet fully sclerotized. Lutein content is significantly higher in cuticle of yellow pupae than of black ones.     

Temporal occurrence of the pupal melanization reducing factor during development of the butterfly, Inachis io

Abstract. . In prepupae of Inachis io L. (Lepidoptera: Nymphalidae), a pupal melanization reducing factor (PMRF) which controls morphological colour adaptation (Bückmann & Maisch, 1987) is located in the brain, suboesophageal ganglion, thoracic ganglia, and all abdominal ganglia and their closely associated neurohaemal organs (Stamecker et al , 1994)
In animals adapted to a yellow background, PMRF content decreased in all these ganglia complexes during the prepupal stage which may be due to a release of the hormone at the critical period of the melanization reducing effect. The release of PMRF apparently occurs in a slow, but continuous, manner and may be superimposed by an incessant PMRF production at the same time recognizable by reincreasing melanization scores towards the end of prepupal and beginning of pupal stage. Therefore PMRF content in ganglia were not completely exhausted. When animals were kept on a black background, such a decline of PMRF content did not occur in both posterior ganglia complexes, whereas values from brain-suboesophageal ganglion complexes were too variable.
The target cells seem to be sensitive to PMRF treatment over a wide time range of nearly 20 h from the early stage of spinning a silk mat to 13-h-old prepupae for the melanization reducing effect.
PMRF activity was also detected in first-instar larvae and in the nervous system of third-instar larvae as well as in pupae which had completed their pigmentation. Furthermore, all three parts of the adult body still contained PMRF. Possibly PMRF may have functions in larval and adult stages in addition to its effect on morphological colour adaptation.     

Regulatory mechanisms in phenotypic plasticity of diapause and nondiapause pupal colouration of the swallowtail butterfly Papilio machaon

Nondiapause pupae of Papilio machaon L. exhibit pupal colour diphenism comprising green–yellow and brown–white types. To understand the regulatory mechanism underlying the control of pupal colouration in P. machaon, the effect of environmental cues on diapause and nondiapause pupal colouration is investigated. When larvae reared under short‐day and long‐day conditions are allowed to pupate in sites with a smooth surface and a yellow background colour, all diapause pupae exhibit a brown–white type and 89.5% of nondiapause pupae exhibit a green–yellow type, respectively. With rough‐surface pupation sites, all diapause pupae exhibit brown–white and intermediate types, whereas a large proportion of nondiapause pupae exhibit brown–white and intermediate types, although some exhibit a green–yellow type. When extracts prepared from the head‐thoracic and thoracic‐abdominal regions of larval central nervous systems are injected into the ligated abdomens of P. machaon short‐day pharate pupae, all recipients exhibit a brown–white colouration. Furthermore, when each extract is injected into the ligated abdomen of Papilio xuthus L. short‐day pharate pupae with orange‐pupa‐inducing factor activity, recipients injected with the head‐thoracic extract exhibit the brown type, whereas those injected with the thoracic‐abdominal extract exhibit an orange colour. The results indicate that the response to the environmental cues of pupation site in P. machaon changes according to the photoperiodic conditions experienced during larval stages, and that at least two hormonal factors producing brown–white pupae are located in the larval central nervous system, with the secretion of these factors being regulated by the recognition of environmental cues in long‐day larvae.     

Hot summer temperatures may stop movement of Papilio canadensis butterflies and genetic introgression south of the hybrid zone in the North American Great Lakes Region

The reasons that the northern tiger swallowtail butterfly Papilio canadensis does not move south of the Great Lakes hybrid zone, where it meets P. glaucus , may be largely due to natural temperature-induced stress on diapausing pupae. Temperatures of 36°C for only four days killed all P. canadensis (from northern Michigan) and most of the Papilio troilus , a species that lives south of the hybrid zone (sympatrically with P. glaucus ). In contrast, interspecific hybrids ( P. glaucus mother× P. canadensis father) had significant adult eclosion or pupal survival at both 30° and 36°C. All surviving hybrid pupae were only females (the heterogametic sex, which are known to express the prolonged diapause due to Haldane effects).
The southern species, P. troilus , had almost all (81%) normal (non-deformed) adults eclose at 30°C, whereas only 20% of the P. canadensis females and ca 30% of the P. canadensis males eclosed to produce non-deformed adults. Unlike the case with P. troilus and P. canadensis , no hybrid females eclosed at 30°C (only males did). Unlike P. troilus and P. canadensis pupae, hybrid (female) pupae remain viable, some of which have already successfully emerged after the chamber experiments.
A follow-up study using P. glaucus , P. canadensis (from Vermont), and their hybrids with more normal lower thermal regimes included (27°, 30°, 33°, and 36°C) again showed higher hybrid survival as uneclosed (living) pupae at 36°C. In addition, P. glaucus and P. canadensis showed high mortality and wing deformity of eclosing adults at 36°C, suggesting that geographic source of the P. canadensis may reflect differential tolerances of the extreme 36°C temperature.     

Diapause pupal color diphenism induced by temperature and humidity conditions in Byasa alcinous (Lepidoptera: Papilionidae)

We investigated whether diapause pupae of Byasa alcinous exhibit pupal color diphenism (or polyphenism) similar to the diapause pupal color polyphenism shown by Papilio xuthus. All diapause pupae of B. alcinous observed in the field during winter showed pupal coloration of a dark-brown type. When larvae were reared and allowed to reach pupation under short-day conditions at 18 °C under a 60 ± 5% relative humidity, diapause pupae exhibited pupal color types of brown (33%), light-brown (25%), yellowish-brown (21%), diapause light-yellow (14%) and diapause yellow (7%). When mature larvae reared at 18 °C were transferred and allowed to reach pupation at 10 °C and 25 °C under a 60 ± 5% relative humidity after a gut purge, the developmental ratio of brown and light-brown, yellowish-brown, and diapause light-yellow and diapause yellow types was 91.2, 8.8 and 0.0% at 10 °C, and 12.2, 48.8 and 39.0% at 25 °C, respectively. On the other hand, when mature larvae reared at 18 °C were transferred and allowed to reach pupation at 10 °C, 18 °C and 25 °C under an over 90% relative humidity after a gut purge, the developmental ratio of brown and light-brown, yellowish-brown, and diapause light-yellow and diapause yellow types was 79.8, 16.9 and 3.3% at 10 °C, 14.5, 26.9 and 58.6% at 18 °C, and 8.3, 21.2 and 70.5% at 25 °C, respectively. These results indicate that diapause pupae of brown types are induced by lower temperature and humidity conditions, whereas yellow types are induced by higher temperature and humidity conditions. The findings of this study show that diapause pupae of B. alcinous exhibit pupal color diphenism comprising brown and diapause yellow types, and suggest that temperature and humidity experienced after a gut purge are the main factors that affect the diapause pupal coloration of B. alcinous as environmental cues.     

The effect of larval photoperiod on pupal colour and diapause in swallowtail butterflies

ABSTRACT.

• 1 There are significant differences in the effects of larval photo-period on diapause and pupal colour among the species Papilio polyxenes Fabr., P.troilus L., Battus philenor (L.) and Eurytides marcellus (Cramer).
• 2 Diapause and pupal colour in P.polyxenes and P.troilus are strongly influenced by larval photoperiod, short photophase eliciting brown diapausing pupae. Photoperiods of 15L:9D permit the expression of the green and brown pupal colour alternatives.
• 3 Pupal colour in B.philenor and E.marcellus is not affected by larval photoperiod, but short photophase induces diapause in these species.
• 4 All species except B.philenor show an association between brown pupal colour and diapause: Emarcellus when reared on long (midsummer) photophase, P.polyxenes and P.troilus when reared on short (autumnal) photophase.
• 5 In P.polyxenes, short photophase can affect pupal colour responses directly, whether the individual enters diapause or not.
• 6 Differences among the species are related to differences in the ecology of their natural pupation sites.

     

Intraspecific variation in wing and pupal melanization in copper butterflies (Lepidoptera: Lycaenidae)

Melanin is a widespread pigment causing variation in skin darkness, with darker phenotypes typically reaching higher equilibrium temperatures than lighter ones. Therefore, selection is expected to favour darker phenotypes in colder environments. In the present study, we show intraspecific variation in pupal (and wing) melanization along an altitudinal gradient in two species of copper butterflies. Both, pupal and wing melanization increased with increasing altitude. Consistent with the thermal melanism hypothesis, darker (high-altitude) pupae reached higher equilibrium temperatures than paler (low-altitude) ones. However, as temperature differences were rather small despite pronounced differences in melanization, we cannot rule out that factors (e.g. ultraviolet protection, disease resistance) other than temperature comprise the principal selective agents. Mechanistically, variation in melanization might be related to variation in hormone titres, as demonstrated by low-altitude pupae showing higher ecdysteroid and juvenile hormone titres compared to high-altitude ones. Furthermore, we report sex differences in wing melanization, with males being darker than females, which is potentially related to a higher flight activity of males.  © 2009 The Linnean Society of London, Biological Journal of the Linnean Society , 2009, 98 , 301–312.     

Photoperiodic induction of diapause in normal and allatectomized precocious pupae of Papilio machaon

Diapause in the swallowtail, Papilio machaon occurs at the pupal stage in response to short days during larval development. The precocious pupae obtained by allatectomy also entered diapause in response to short days. The photoperiodic induction of diapause in precocious pupae was very similar to that in normal pupae in spite of the abbreviation of the last-larval instar. The critical photoperiod for diapause in the precocious pupae was 14.5 h, being about the same as that in normal pupae. Occurrence of diapause in the swallowtail is closely associated with the pupal stage and the induction process may be triggered by the cessation of juvenile hormone secretion.     

Differences in pupal cold hardiness and larval food consumption between overwintering and non‐overwintering generations of the common yellow swallowtail,Papilio machaon (Lepidoptera: Papilionidae), from the Osaka population

To clarify differences in pupal cold hardiness and larval food consumption between overwintering and non‐overwintering generations of the common yellow swallowtail, Papilio machaon, we reared larvae from the Osaka population under photoperiods of 16 h light : 8 h dark (LD 16:8) (long day) or LD 12:12 (short day) at 20°C. We examined the relationship between food consumption and weight during the final larval stadium and pupae, and measured the pupal supercooling point (SCP). Although the ratio of assimilation to consumption did not differ significantly between photoperiods, the ratio of assimilation to pupal weight differed significantly between individuals reared under long and short days. All diapausing pupae were brown, whereas 56% of non‐diapausing pupae were green with the remainder brown. The mean pupal body length (L), dorsal width (W₁) and lateral width (W₂) were larger in non‐diapausing than in diapausing pupae, and the W₁/L and W₁/W₂ ratios differed significantly between non‐diapausing and diapausing pupae. SCP was approximately –20°C and did not differ among pupae 5, 15 and 30 days after pupation under long‐day conditions. However, under short‐day conditions, mean SCP gradually decreased, stabilizing at approximately –24 to –25°C by 30 days after pupation. After freezing, some diapausing pupae emerged as adults, whereas all non‐diapausing pupae died. Both egestion and assimilation were greater under long‐day conditions. The results revealed that pupae of this papilionid exhibit seasonal polyphenism in physiological and morphological traits. Energy from food appears to be expended on increasing cold hardiness in the overwintering generation and on reproduction in the non‐overwintering generation.     

Pupal colour dimorphism in a desert swallowtail (Lepidoptera: Papilionidae) is driven by changes in food availability,not photoperiod

1. The swallowtail butterfly Battus polydamas archidamas Boisduval, 1936, exhibits polyphenism for pupal coloration (green and brown). It is distributed across arid regions with winter rains and is monophagous on Aristolochia plants, which emerge after the winter rains and dry out the during summer. Thus, day length does not covary positively with host plant productivity. It was hypothesised that pupal colour was driven by food availability, not photoperiod. The benefits of pupal coloration matching the colour of pupation sites in terms of field survival were also investigated to evaluate the adaptive value of pupa colour. 2. Larvae were reared under a factorial array of two photoperiods (LD 10:14 h and LD 14:10 h) and two food availability regimes (leaves ad libitum and available every other day) to assess the frequency of green and brown pupae. Field survival of green and brown pupae was quantified in three commonly used habitats that differ in background coloration (cacti, rocks and shrubs). 3. Food availability determined pupal colour. Larvae in the ad libitum regime resulted mostly in green pupae, while those with restricted food were mostly brown. In contrast, photoperiod did not influence pupal colour. Survival probability of pupae placed on cacti was higher than those placed on rocks and shrubs, and the lowest predation risk across habitats was for green pupae on cacti. 4. Food availability plays a major role in the seasonal polyphenism for pupal colour of specialist butterflies inhabiting arid environments with winter rains.     

Allochronic isolation and incipient hybrid speciation in tiger swallowtail butterflies

Hybridization leading to reproductively isolated, novel genotypes is poorly understood as a means of speciation and few empirical examples have been studied. In 1999, a previously non-existent delayed flight of what appeared to be the Canadian tiger swallowtail butterfly, Papilio canadensis, was observed in the Battenkill River Valley, USA. Allozyme frequencies and morphology suggest that this delayed flight was the product of hybridization between Papilio canadensis and its sibling species Papilio glaucus. The mitochondrial DNA (mtDNA) restriction fragment length polymorphisms presented here indicate that only P. canadensis-like mtDNA occurs in this population, suggesting that introgression likely occurred from hybrid males mating with P. canadensis females. Preliminary studies of this population indicated that delayed post-diapause pupal emergence in this hybrid genotype was the root cause behind the observed delayed flight, which suggests a potential empirical example of a mechanism leading to reproductive isolation. Here we provide further evidence of the role of adult pupal emergence as a reproductive barrier likely leading to reproductive isolation. In particular, we present results from pupal emergence studies using four different spring and two different winter temperature treatments. The results indicate a clear separation of adult emergences between the hybrid population and both parental species. However, our results indicate that exceptionally hot springs are likely to lead to greater potential for overlap between the local parental species, P. canadensis, and this delayed population with hybrid origins. Conversely, our results also show that warmer winters are likely to increase the temporal separation of the hybrid population and the parental species. Finally, we report recently collected evidence that this hybrid population remains morphologically distinct.