Another way to be green: pigmented liquid ink in butterfly wings – a novel form of colouration

doi:10.21203/rs.3.rs-1506845/v1

Download PDF

Article

Another way to be green: pigmented liquid ink in butterfly wings – a novel form of colouration

https://doi.org/10.21203/rs.3.rs-1506845/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

A novel mechanism underlying butterfly wing colouration was discovered in two distantly related butterfly species, Siproeta stelenes (Nympalinae) and Philaethria diatonica (Heliconiinae). These butterflies have bright green wings, but the colour is not derived from pigments or nano-structures of the scales or the colour of the cuticular membrane, but from an ink-like liquid retained in the wings. Wing structure differs between the green and non-green areas. In the non-green region, the upper and lower cuticular membranes are attached to each other, whereas in the green region there is a space of 5–10 microns where liquid ink is held and living cells were observed. The green ink was composed of pigments similar to those in the hemolymph, but enriched with carotenoid lutein. This novel discovery sheds a new light on our understanding of the evolution of wing colouration in butterflies.

The diverse colours of butterfly wings are not only fascinating to the human eye but have many biological functions, including sexual attraction, communication, mimicry, camouflage (crypsis), aposematism (warning signals), and thermoregulation. There are more than 17,000 butterfly species in the world¹, and the mechanisms responsible for wing colouration pattern have therefore been extensively studied.

The wing membrane of butterflies is usually colourless, and the variety of colours generally comes from the scales that cover the wings^2–4. A scale is a complex cuticular structure derived from a single scale cell, and scale colour can be produced by pigment deposition⁵ or by micro/nanoscopic structures that produce iridescent or black colours through the combined action of optical interference and diffraction^6–11. In general, wing colouration is a product of various factors, including scale pigment types, scale shape and size, scale nanostructures, scale arrangement, and their position^12–14. Living scale cells do not contribute to the wing colouration in mature adult wings because they usually disappear at the moment of adult eclosion¹⁵.

A few butterflies have coloured wing membranes, especially the swallowtail butterflies of the genus Graphium (Papilionidae: Papilioninae) and some brush-footed butterflies of the subfamilies Charaxinae, Danainae, and Satyrinae^16–19. In these cases, wing colour comes from the structural colour of the cuticle or a solid pigment deposited in the cuticle¹⁷.

Thus, until now it has been thought that wing coloration is derived from either the scales or the wing membrane itself. Here we describe and provide evidence for a previously undescribed mechanism of wing colouration, namely an ink-like liquid between the upper and lower cuticle of the wing membrane. We also identify the biochemical nature of this ink compound.

Existence of green ink in wings

We focused on the green colour of Siproeta stelenes (Nymphalidae: Nymphalinae) and Philaethria diatonica (Nymphalidae, Haliconiinae) (see General information of the studied butterflies in Materials and Methods, Supplementary text 1 for the discovery story and Supplementary Fig. 1 for photos of live and pinned butterflies). When wing scales of both dorsal and ventral surfaces of pinned S. stelenes were removed, the green part of the wing was not affected (Fig. 1a and Supplementary Fig. 2). This shows that the green colouration does not come from scales. By carefully checking the green areas, we discovered that the costal area of the wing occasionally contained air bubbles and had no green colour (Fig. 1a’), suggesting that the green colour was present in liquid form.

To check for the presence of green-coloured liquid, we inserted a dry, white, filter paper in the green area of a recently cut wing from a live butterfly. The paper absorbed the liquid and turned green and the formerly green colour of the wing became pale (Figs. 1b, 1b’). In order to confirm the fluidity of the green liquid, we gently pressed a rubber eraser tip against a wing. As a result, the colour disappeared immediately from where the rubber tip was pressed. The liquid gradually returned to its position in approximately 5 minutes (Fig. 1c and Supplementary Movie 1). This result clearly indicates that a viscous green liquid exists inside the living wing (Supplementary Fig. 3). We made similar observations on P. diatonica and obtained the same results (Supplementary Figs. 4a, 4b).

An additional experiment was conducted on P. diatonica wings. Figures 1d and 1d' show a P. diatonica wing which has been split in half between the upper and lower cuticular membranes to expose the green liquid and then pressed against a white paper. The green colour was completely transferred to the paper, behaving like liquid ink and leaving the green areas of the wing membrane whitish translucent (Figs. 1d, 1d’). This observation clearly shows that the green colour of P. diatonica is not produced by scales nor is it deposited within the wing cuticle, but rather the wing membrane holds a type of green liquid as in S. stelenes. Exposing the green liquid of these two butterflies by separating the upper and lower cuticular membranes caused the liquid to dry out and become solid in a couple of seconds under ambient air temperatures of 18–23℃ and relative humidity of 70–80% (n > 30) (Supplementary Figs. 5a–d). In other words, this liquid is highly volatile and/or reactive, and becomes thicker, i.e. more viscous, and then becomes completely dry within a few seconds (n > 30).

We also observed that green liquid was already present in developing wings of late-stage pupae of S. stelenes (Supplementary Figs. 6a–c). This suggests that the onset of the green colouration occurs prior to pigment deposition in scales^20–22.

Our observations and experiments reveal a new source of butterfly wing colouration, namely a green liquid in the wing membrane, which we refer to as ink.

Wing structure in the area with and without green ink

The green area in Siproeta stelenes and Philaethria diatonica covers a significant part of the wing membrane, 26.3% and 29.8%, respectively. We examined the scales and wing membrane of S. stelenes to see if there were any differences between the green (with ink) and brown (without ink) areas. The scales on dorsal wing membrane were different in shape, size, and arrangement between these areas, i.e. the green areas had sparse, vertically positioned, small, brown to gold, semi-translucent scales, whereas the dark area was densely covered (tapestry-like) with dark-pigmented scales (Fig. 2a and Supplementary Fig. 7; see Supplementary Table 1 for general wing structure and scale characteristics, including those of three other blue-green butterflies such as Graphium sarpedon (Linnaeus, 1758), Parantica sita (Kollar, 1844) and Ideopsis similis (L., 1758): Supplementary Fig. 8a–c). The surface structure of the wing membrane was clearly different between the green areas and the rest of the wing (Figs. 2b–d). The surface structure of the cuticle in the green areas was slightly wrinkled and appeared to be rough (nanopillar-like structures were present^23–25) (Fig. 2c and Supplementary Fig. 9), whereas that of the brown areas was smoother (Fig. 2d). This phenomenon of wrinkled versus smooth areas was also true for P. diatonica (Supplementary Fig. 10).

Figures 3a and 3b show a cross-section of the wing, including the veins, where 3a is the green area and 3b is the brown area (see Supplementary Fig. 11 for cross-section sites). The vein of the green area (Fig. 3a) is filled with irregularly-shaped, granular, dried ink. In contrast, the brown area had no filling (Fig. 3b). Figures 3c and 3d show close-up images of the wing membranes of Figs. 3a and 3b. In the brown part of the wing membrane, the upper and lower cuticular membranes are closely positioned to one another (Fig. 3d, and Supplementary Fig. 12b), that is, there is virtually no space between the two cuticular layers. In the green area, there is a notable amount of space (ca. 5–10 µm) between the two layers (Figs. 3c and Supplementary Fig. 12a) and it contains irregularly-shaped, granular, dried ink as in the wing veins of green areas (Fig. 3c and Supplementary Fig. 13). Similar irregularly-shaped, granular dried ink was observed between the upper and lower wing cuticular membranes of P. diatonica green wing areas (Supplementary Fig. 14).

Existence of living cells in the wing membrane

Cell nuclei were observed with a Hoechst 33342 stain in each area (Fig. 4). In Siproeta stelenes many nuclei were uniformly present in the green areas (Fig. 4a1), while there were sparse nuclei arranged in a linear pattern in the brown areas (Fig. 4a2). We employed the same analysis on a species of Siproeta that does not have green wings, S. epaphus (Latreille, [1813]) (Fig. 4b and Supplementary Figs. 1g–i) and there were sparsely arranged nuclei in a linear pattern in both pale and dark areas (Figs. 4b1, 4b2). The uniformly distributed nuclei as seen in the green wing areas of S. stelenes were not observed in S. epaphus. The lineally arranged sparse nuclei (Figs. 4a2, 4b1, 4b2, 4c2) are thought to originate from socket cells, as has been observed in other butterflies^15,26. In addition, we investigated the nuclear arrangement pattern of Philaethria diatonica and as expected, we observed the homogeneous pattern of nuclei in the green areas (Fig. 4c1). We also examined the blue-green and brown wing cuticles in Graphium sarpedon, which has coloured wing membranes with no ink present, and in both areas there were regularly spaced socket cell nuclei (Supplementary Fig. 15). These results show that there is a relationship between the distribution of cell nuclei and the presence of ink. They also suggest that the epithelial cells of the green areas do not degenerate during or after wing development, even after extending the wings and flying, in contrast to the normal developmental process^27,28. These living cells in wing membranes are associated with ink; hereafter we call them green ink cells. Observing green ink cells of S. stelenes via F-actin staining revealed that they are hexagonally shaped (Fig. 5). The existence of living cells inside wing veins was also recorded (Supplementary Fig. 16). With the exception of androconial organs (male scent-producing cells, pads and patches, usually accompanied by modified scales)^29–33, living cells in wing membranes of butterflies are uncommon³⁴.

Identification of pigments responsible for the green colouration of the wings

In many phytophagous insects, including lepidopteran species, a green appearance usually results from a combination of yellowish carotenoid pigments, bluish bile pigments, and their binding proteins³⁵. From the green ink a yellow colour was efficiently extracted with non-polar solvents, such as chloroform, indicating the presence of carotenoid pigments. We therefore performed carotenoid composition analysis using LC-MS and confirmed that the yellow pigment included in the green liquid was lutein (Fig. 6a–c). The lutein was abundant in the green liquid, but less so in the brown (non-green) region of the wings (Fig. 6d). Because the green pigments suspended in phosphate-buffered saline (PBS) were concentrated in an ultrafiltration column (NMWL = 100kDa), they appeared to be associated with macromolecules. The absorbance spectrum of this green fraction had multimodal peaks around 380–480 nm and 650–675 nm, in addition to strong absorbance at 280 nm, which was presumably derived from proteins (Fig. 7a). When this macromolecular fraction was analyzed in a size-exclusion column, although the majority of the pigment was not eluted from the column, we obtained two separated chromatic components (peak I and II in Fig. 7b). The first component was eluted earlier than the upper marker (= 660 kDa) and therefore seems to have a larger molecular size. It had a characteristic absorbance maximum around 450 nm (Fig. 7d), which was apparently derived from the yellow carotenoid, lutein (Fig. 6b). We verified that this protein fraction contained a considerable amount of lutein. On the other hand, the second chromatic protein component (ca. 22 kDa) had absorbance maxima at 380 nm and 650 nm (Fig. 7d). This absorbance signature greatly resembles the absorbance spectra of blue bilin-binding proteins reported in several lepidopteran species³⁶. The absorbance spectrum of the original green ink can be fully elucidated by the combination of these two pigment-binding proteins (Fig. 7a vs 7d). In addition, the smaller blue protein (ca. 22kDa) was derived from > 100 kDa fractions, suggesting that it formed a complex with the larger yellow component in vivo. These results taken together suggest that the green pigment is a macromolecular assembly of yellow carotenoid-binding proteins and blue bilin-binding proteins.

Identification of green pigments in the hemolymph

We found that the hemolymph circulating in the body cavity showed a similar greenish colour, although the green ink in the wing was somewhat more yellowish. Spectrum comparison revealed that the green ink in the wing contains a relatively large amount of the carotenoid binding protein, which has absorbance maxima around 450 nm (Fig. 7a). Column separation of the green hemolymph consistently demonstrated that it has the same carotenoid-binding and bilin-binding proteins as the wing ink, but with a different ratio (Fig. 7b vs 7c). These results support the idea that the green pigments in the wing liquid were derived from those of the hemolymph and develop vivid yellow-green colouration by the preferential uptake of more carotenoid fractions.

Wing colouration based on an ink system in butterflies is an unexpected discovery and raises a number of interesting questions.

Why green ink?

According to Srygley and Chai (1990) the rufous-tailed jacamar is an important predator of Siproeta stelenes and Philaethria diatonica in Costa Rica³⁷. Both of these butterflies are active during the day in vegetation-rich, open environments where the green colour of their wings may act as camouflage and help them hide from predatory birds, especially while resting on the underside of leaves where they become leaf-like by transmitting and diffusing sunlight³⁸ (Supplementary Figs. 1b, 1e, 17). In postmortem conditions the green colour fades away and becomes a pale bluish-green over a short period of time³⁹ (Supplementary Figs. 1c, 1f, 17, Supplementary Movie 2). We found that the carotenoid lutein molecules are essential for the vivid green color of the wing of S. stelenes. It is known that the lutein pigment in insects is of plant origin (i.e., host plants consumed during the larval stage)^40–42, thus a continuous production and supply of lutein in adult butterflies is unlikely. As in carotenoids in general it has been reported that lutein is easily degraded by heat- and photo-oxidation⁴³. Therefore, preventing degradation of and maintaining the green colour is necessary. Additionally, the ‘fast-drying’ or ‘fast-coagulating’ character of the ink appears to prevent continuous oozing of green ink by sealing damaged parts of the wing (Supplementary Figs. 5b, 5c).

Where does the ink come from?

The pigment analysis revealed that the colour constituents of the green ink in the wings are the same as those in the body fluid. We found that the body fluid in the abdomen of Siproeta stelenes, S. epaphus and Philaethria diatonica is very similar in colour to the ink present in the wings of S. stelenes and P. diatonica (Supplementary Figs. 18a–d). However, it appears that there is no circulation of the observed green fluid within the wings of S. epaphus. When cutting between the base of the wing and the thorax of S. stelenes, greenish fluid seeped out from a section of the veins (Supplementary Fig. 13a), but a cross-section of the vein in the green area showed that the wing vein channel is filled with solid, granular, dried ink (Fig. 3a). On the other hand, in S. epaphus, which does not have green areas on its wings, there were no ink-like granules in the wing veins (Supplementary Fig. 19). Under a stereo-microscope, greenish ink was observed inside the veins of both S. stelenes (Supplementary Figs. 20a, 20b) and P. diatonica (Supplementary Figs. 20c, 20d). These results suggest that the ink most likely comes from the body via hemolymph circulation^34,44. It is also important to note that hemolymph is necessary for supplying nutrition to the green ink cells in the wing membranes. Additionally, the pre-eclosing S. stelenes adult inside the pupal shell showed a fully expanded abdomen with green fluid (Supplementary Fig. 21).

The tracer experiment showed that when coloured solution was injected into the central portion of the abdomen (dorsal vessel/heart chambers) of live S. stelenes, the dye reached the green areas of wing membranes within 30 mins (Supplementary Figs. 22a–d). The flow of the dye solution was also visible running through the wing base to the wing veins in less than a second after injection (Supplementary Fig. 22a). The flow of dye solution in the veins was more notable when using red dye (Supplementary Fig. 22d). We did not observe dye pigments in non-green areas. This suggests that there is a possible flow of green ink between the abdomen-thorax and the green areas of the wing membrane. We showed that both the wing green ink and hemolymph is composed of carotenoid- and bilin-binding proteins, which are prevalent in green larval integuments and hemolymph of lepidopteran species^35,45. This suggests that the unique wing colouring system found in S. stelenes and P. diatonica has evolved by diverting a pre-existing pigmentary mechanism to the green ink ‘pouches’ in the wings.

How common is the ink system?

It would be interesting to examine whether this novel wing colouring system occurs in other butterflies. We confirmed the presence of ink in the wings of just two species, Siproeta stelenes and Philaethria diatonica, which belong to distantly related nymphalid subfamilies, Nymphalinae and Heliconiinae, respectively (Supplementary Fig. 23). Thus, the acquisition of green colouration provided by ink in these two butterflies is probably a result of convergence, suggesting that wing colouration by ink may be more widespread. We have confirmed that S. epaphus, the sibling species of S. stelenes, does not have ink in the wings. In contrast, all species of the genus Philaethria are green when alive and the colour fades away in postmortem conditions³⁹, strongly suggesting that they all possess the ink. Thus, in Philaethria, the ink system may be a synapomorphic trait for the genus.

Non-structural green and blue colouration is very rare among butterflies⁴, and it remains to be established if other green ‘pigmented’ butterflies are exploiting the same mechanism as S. stelenes and Philaethria spp. If so, the wing ink-colouration system might be more widespread across the superfamily Papilionoidea. For example, the green-coloured swallowtails Graphium tynderaeus (Fabricius, 1793) from Africa, G. macleayanum (Leach, 1814) from Australia, and G. codrus (Cramer, 1777) from Southeast Asia and Oceania have a similar wrinkled surface structure on the wing membrane, suggesting possible presence of epithelial cells (Fig. 20 in¹⁸). It should be noted that such wrinkles were undetected in live and dried specimens of both S. stelenes and P. diatonica under a stereomicroscope with 64 x magnification. Wrinkles observed in SEM (Fig. 2c and Supplementary Fig. 10 green) are possibly a side effect of the dehydration processes involved in SEM sample preparation, although the wrinkle protrusions seen in the SEM images appear to coincide with the size and distribution of the green ink cells. Since the presence of wrinkles is not direct evidence of the existence of ink, future studies are required to determine whether ink is present in the wings of those species of Papilionoidea mentioned above.

There are undoubtedly many undiscovered, fascinating phenomena in the commonly seen and attractive organisms that surround us. Future research should include an examination of other insects having bright, non-metallic, green wings, such as cicadas.

General information of the studied butterflies

Siproeta stelenes biplagiata (Fruhstorfer, 1907)⁴⁶ (Nymphalidae: Nymphalinae: Victorinini) is commonly known as the Malachite. The wingspan is ca. 9 cm. The genus Siproeta Hübner, 1823 contains three species⁴⁷; however, only S. stelenes has the green colouration, which resembles and is frequently confused with species of the genus Philaethria. Siproeta stelenes is distributed from southern Texas to Brazil, and in Costa Rica it occurs from sea level up to ca. 1400 m⁴⁸. Individuals are commonly seen along open forest edges perched on lower vegetation, visiting flowers, decomposing materials and minerals on the ground, and resting on undersides of leaves in late afternoon.

Philaethria diatonica Billberg, 1820⁴⁷ (Nymphalidae: Heliconiinae: Heliconiini) is commonly known as the Green Longwing. The wingspan is ca. 11 cm. The genus Philaethria⁴⁷ contains 10 species, all of which have the green colouration. Philaethria diatonica is distributed from Mexico to Panama³⁹, and in Costa Rica it occurs from sea level up to about 1400 m⁴⁸. Individuals are infrequently observed and seen flying fast and high along wide trails, visiting flowers, or perched on undersides of leaves in late afternoon⁴⁷.

These two butterfly species are popular among the general public. Images and videos are posted on the internet, used as the cover of butterfly books⁴⁹, postal stamps (internationally), and even products such as notebooks, calendars, bookmarks, jigsaw puzzles, and smart phone cases. They are commonly reproduced in butterfly gardens in Costa Rica, and their pupae are exported to other countries while dried specimens are sold worldwide as souvenirs.

Butterfly specimen supply

Live adults of Siproeta stelenes (n > 50) were obtained in Costa Rica from the Monteverde Butterfly Gardens (Monteverde), the Pierella Ecological Garden (Horquetas de Sarapiquí), and wild specimens from the Estación Biológica Monteverde. Pupae were purchased from Mariposario Berakah (Aguas Zarcas) and the Butterfly Kingdom Costa Rica (Escazú) and were reared at Estación Biológica Monteverde. Pupae of Philaethria diatonica (n = 30) were also purchased from Mariposario Berakah and reared to mature adults. Siproeta epaphus (n = 12) were provided by the butterfly garden of the Escuela de Biología, Universidad de Costa Rica. Some specimens were placed in wax paper envelopes and dried or preserved in 75% EtOH and kept at 4°C. Some of these specimens were sent to the Laboratory of Pattern Formation of the Graduate School of Frontier Biosciences at Osaka University in Japan. Additionally, pupae of S. stelenes (n = 50) were exported from Costa Rica Entomological Supply (La Guácima) to the Laboratory at Osaka University.

Observation of specimens

In Costa Rica live butterflies were dissected and observed under a stereo-microscope, Zeiss Stemi DV4. Wing scales were removed using fine paint brushes. Wings of live butterflies were cut at the base using iris scissors. Fine-point forceps were inserted into the cross section of large veins to carefully split the wing membrane in half. To extract the green ink, a small, white, pointed filter paper was inserted between wing membranes. Photos and video clips were recorded with digital cameras OM-D E-M1 Mark II with a 60 mm macro lens, Canon EOS 7D with a 100 mm macro lens, iPhone 7 and Olympus Tough TG-5 for very close-up images. Images were processed and edited with Adobe Lightroom Classic and Photoshop software.

Measurement of the proportion of green area vs total area

The whole body with wings of Siproeta stelenes (n = 3) and Philaethria diatonica (n = 3) were spread and photographed. Using imageJ, the contours of the whole butterfly were extracted, and their areas were calculated. We also extracted the green areas using the segmentation command in imageJ and calculated their areas. From both areas we calculated the ratio of the area of the green region to the total area. The thresholds for both contours were set manually by comparing them with the actual images.

SEM analysis

The 75% ethanol-fixed samples were naturally dried or dehydrated with 90/100% ethanol and replaced with t-butanol. Cross sections were prepared using a razor. The samples in t-butanol were freeze-dried (FZ-2.5; Asahi Life Science, Japan) and coated with Au with DII-29010SCTR Smart Coater (JEOL Ltd., Tokyo, Japan), then observed with a scanning electron microscope JCM-6000 (JEOL Ltd., Japan).

Observation of cuticular autofluorescence in wing sections

Four percent PFA fixed samples were washed three times in phosphate-buffered saline (PBS), then dehydrated with 70 to 100% ethanol and QY-1 solution for over 2 hours in each step. The dehydrated samples were embedded with UV-resin. After the resin hardened, they were sectioned (20 µm) using a microtome and mounted on a glass slide. The samples were observed with confocal microscope LSM780 (Carl Zeiss, Germany). Autofluorescence was recorded with 488 nm excitation light.

Nuclei staining

Wing scales of butterfly specimens were carefully removed using fine, soft paint brushes. These samples were incubated in Hoechst 33342 solution (1:1000) for one hour, washed three times, and then mounted on a glass slide for observation with confocal microscope LSM780 (Carl Zeiss, Germany).

F-actin staining

The wing membrane of Siproeta stelenes was separated into upper and lower membranes with a pair of forceps. The samples were fixed with 4% PFA at 4℃ for 1 h. The fixed samples were incubated in PBS with 0.1% Tween® 20 Detergent (PBST) for 10 mins. They were then placed in Alexa Fluor™ 488 Phalloidin (Invitrogen) solution (1:100) for over 2 h, washed three times, and mounted on a glass slide. The samples were observed from the inside with a confocal microscope LSM780 (Carl Zeiss, Germany).

Tracer experiment

Two types of dye solutions were prepared: 45µL 0.1% Fast green FCS (Sigma-Aldrich) with 5µL Hoechst 33342 solution (Final concentration: 100µg/mL) and 45µL 1% congo red (NACALAI TESQUE) with 5µL Hoechst 33342 solution (Final concentration: 100µg/mL). Wing scales were partially removed with fine, soft paint brushes. Fifty µL of each dye solution was injected dorsally into centre portion of each Siproeta stelenes abdomen (n = 4), using a syringe with 30-gauge needle (BD) and at least 30 mins before examining under an All-in-One Fluorescence Microscope BZ-X 710 (Keyence, Japan).

Time-lapse fixed-point observation of wing colour change

A live Siproeta stelenes was frozen at -30℃ for 1 h and then brought to room temperature. All four wings were removed with iris scissors and placed in an isolated chamber under conditions of ca. 25℃, 50% humidity, externally lighted with 7.5 watts Leica L2 halogen lamp. The time-lapse feature of an Olympus Tough TG-5 compact digital camera was used (a single photo was taken every 15 mins for 5 days = 120 h), connected to an AC adapter-USB to maintain the camera power source.

Extraction, purification, and identification of green pigments

Green liquid was collected from Siproeta stelenes wings and abdomen of cryopreserved specimens. First, the wings and the body were separated in PBS and the wings were further processed only in the green region. The green liquid was extracted from the wings by applying pressure with forceps to the wings in the green region. The abdomen was cut in the middle and the green fluid from the abdomen was separated. The green pigment-containing solution was concentrated in an ultrafiltration column [Amicon Ultra, Merck, nominal molecular weight limit (NMWL) = 100 kDa] and washed with PBS three times in the same column. The concentrated green solution was then subjected to chromatographic separation using a size exclusion column (Superdex 200 Increase 10/300, Cytiva) in a liquid chromatography system (Prominence, Shimadzu) equipped with a photodiode array detector (SPD-20AC, Shimadzu). The separation was performed using a mobile phase [25mM tris (hydroxymethyl) aminomethane-hydrochloric acid buffer containing 150 mM NaCl, pH 7.5] at a flow rate of 0.75 mL/min. Some peak fractions exhibiting characteristic absorbance at 450 nm and 650 nm were collected for carotenoid analysis. These were mixed with water, methanol, and chloroform in a ratio of 2:2:1. After centrifugation, the upper polar phase was discarded and re-extracted with 2 parts of water. The organic phase was dried in vacuo, and stored at -80°C until analysis.

Carotenoid pigments were extracted with ethanol from the green and brown (non-green) regions of the wing, hemolymph from the adult abdomen, and the pupal wing bud. A 120 µL aliquot of the extracts was mixed with 50 µL of water and 50 µL of chloroform and mixed vigorously. Then, 200 µL of water was mixed to make a liquid-liquid phase separation (LLPS). After centrifugation, the upper aqueous phase was discarded and re-extracted with 200 µL of water. The yellow organic phase was dried in vacuo, and stored at -80°C until use.

Carotenoid compositions were analyzed using an LC-MS system composed of an ultra-performance liquid chromatograph (Acquity UPLC H-Class, Waters) and a high-resolution mass spectrometer equipped with an electrospray ionization source (Xevo-G2XS, Waters). The dried carotenoid samples were resuspended in methanol and separated in a CORTECS UPLC phenyl column (Waters, 2 mm i.d. x 150 mm length). The gradient elution program was started at 5% water containing 10mM ammonium formate, 15% methanol, and 80% acetonitrile, changed to 100% acetonitrile during 4 min, and kept for 2 mins at a flow rate of 0.3 mL/min. Carotenoid compounds were identified based on retention times, absorbance spectra, precursor m/z, and MS/MS product ion profiles. For quantification, we used peak areas of light absorbance at 450 nm.

Acknowledgments: We thank and express our deepest gratitude to the following people and institutions: Stephen Viquez Mora (Mariposario Berakah, Costa Rica), Bryna Belisle (Monteverde Butterfly Gardens, Costa Rica), Luis Ricardo Murillo (Mariposario de la Escuela de Biología, Universidad de Costa Rica), Hiroshi Nakamine and Satoshi Shimizu (Minoh Park Insectarium, Osaka, Japan), William and Kristal (Pierella Ecological Garden, Costa Rica), Miou Sasa (Akita-shi, Japan), Shuichi Ueno and Hiroyuki Kikukawa (Kagoshima-shi, Japan), Sergio Siles (Costa Rica Entomological Supply) for supplying live specimens; Mr. Shiraiwa and Mr. Ihara (Plant Protection Station, Japan), Blanca Kato (PROCOMER Costa Rica Exports, Japan) and Kenji Shiota (Nippon Express, Japan) for processing and helping us import live specimens; Shigehiko Shiyake (Osaka Museum of Natural History, Japan) for access to cicada specimens; Hajime Tsujimoto (Kashihara City Museum of Insects, Nara, Japan) for identification of the flowering plant in Supplementary Figure 5; Carla Penz (Department of Biological Sciences, University of New Orleans, USA) for providing information and literature, and revision and suggestions on the manuscript; William A. Haber (Monteverde, Costa Rica) for lending specimens of Philaethria diatonica and Siproeta superba, literature, and suggestions for the article title; Michael Braby (Australian National University) for comments and revision of the manuscript; Hiroki Gotoh (Shizuoka University, Japan) for suggestions, Atsuko Iwane and Takako Ichinose (RIKEN, Japan) for teaching some experimental procedures; Elizabeth Heffington (Law Library, Soledad, California) for reviewing and correction of the manuscript; Pamela Méndez (Monteverde, Costa Rica) for assisting in short video filming; La Selva Biological Research Station (Organization for Tropical Studies, Costa Rica), and Laura Chinchilla and Warner Masís (Zona Protectora El Rodeo/Universidad para la Paz, Costa Rica) for providing research locations; Lourdes Vargas Fallas, Javier Guevara Sequeira, Gustavo Induni Alfaro, Sandra Diaz, Fabricio Carbonel, Alexander León Campos, Isaac López Núñez, Celso Alvarado Murillo, Miguel Matarrita, Randall Zamora Castro, Angie Michelle Sanchez Nuñez, and Rafael Gutierrez Rojas (MINAE-SINAC, Costa Rica) for research/collecting and export permits. Research permit numbers are as follows: 056-2017-ACC-PI, R-SINAC-PNI-ACAHN-26-2019, and SINAC-ACC-PI-R-057-2019; and export/import permits: CUSBSE-861-2019, PE-CUSBSE-124-2020, CUSBSE-731-2021, Certificate No. 6999714 of Costa Rican Ministry of Agriculture and Livestock, 3K439 of Ministry of Agriculture, Forestry and Fisheries. Finally, we thank the Comisión Institucional de Biodiversidad, Universidad de Costa Rica for granting permission (CBio-3-2022) to access genetic and biochemical resources.

Notes

Author contributions

KN made the discovery. All authors except for PH conceived and designed the experiments. KN and HA carried out the experiments, except for pigment analysis done by MM. All authors wrote the manuscript. All authors discussed the results and commented on the manuscript.

Competing interests

The authors declare no competing interests.

Additional information

Supplementary files and information are available for this paper at https: ###.

Correspondence and requests for materials should be addressed to KN or HA.

Shields, O. World numbers of butterflies. J. Lepid. Soc. 43, 178–183 (1989).
Poulton, E. B. The Colours of Animals, Their Meaning and Use, Especially Considered in the Case of Insects. (Kegan Paul, Trench Tru ̈bner & Co. Ltd, 1890).
Nijhout, H. F. The color patterns of butterflies and moths. Sci. Am. 245, 140–153 (1981).
Nijhout, H. F. The Development and Evolution of Butterfly Wing Patterns. (Smithsonian Institution Scholarly Press, 1991).
Matsuoka, Y. & Monteiro, A. Melanin pathway genes regulate color and morphology of butterfly wing scales. Cell Rep. 24, 56–65 (2018).
Ghiradella, H. Light and color on the wing: structural colors in butterflies and moths. Appl. Opt. 30, 3492–3500 (1991).
Kinoshita, S., Yoshioka, S. & Kawagoe, K. Mechanisms of structural colour in the Morpho butterfly: cooperation of regularity and irregularity in an iridescent scale. Proc. Biol. Sci. 269, 1417–1421 (2002).
Kinoshita, S., Yoshioka, S., Fujii, Y. & Okamoto, N. Photophysics of structural color in the Morpho butterflies. Forma 17, 103–121 (2002).
Kinoshita, S., Yoshioka, S. & Miyazaki, J. Physics of structural colors. Rep. Prog. Phys. 71, 076401 (2008).
Yoshioka, S. & Kinoshita, S. Wavelength–selective and anisotropic light–diffusing scale on the wing of the Morpho butterfly. Proc. Biol. Sci. 271, 581–587 (2004).
Vukusic, P., Sambles, J. & Lawrence, C. Colour mixing in wing scales of a butterfly. Nature 404, 457–457 (2000).
Yoshioka, S. & Kinoshita, S. Single-scale spectroscopy of structurally colored butterflies: measurements of quantified reflectance and transmittance. J. Opt. Soc. Am. A Opt. Image Sci. Vis. 23, 134–141 (2006).
Stavenga, D. G., Giraldo, M. A. & Leertouwer, H. L. Butterfly wing colors: glass scales of Graphium sarpedon cause polarized iridescence and enhance blue/green pigment coloration of the wing membrane. J. Exp. Biol. 213, 1731–1739 (2010).
Stavenga, D. G., Matsushita, A. & Arikawa, K. Combined pigmentary and structural effects tune wing scale coloration to color vision in the swallowtail butterfly Papilio xuthus. Zoological Lett. 1, 1–10 (2015).
Dinwiddie, A. et al. Dynamics of F-actin prefigure the structure of butterfly wing scales. Dev. Biol. 392, 404–418 (2014).
Rothschild, M. & Mummery, R. Carotenoids and bile pigments in danaid and swallowtail butterflies. Biol. J. Lonn. Soc. 24, 1–14 (1985).
Kato, Y., Eba, J. & Yamada, H. Stimulation of blue bilin production by light in the wing of a papilionid butterfly Graphium sarpedon. Lepido. Sci. 61, 191–201 (2010).
Allyn, A., Rothschild, M. & Smith, D. Microstructure of blue/green and yellow pigmented wing membranes in Lepidoptera with remarks concerning the function of pterobilins. 1. Genus Graphium. Bull. Allyn Mus. 75, 1–20 (1982).
Sekimura, T. & Nijhout, H. F. Diversity and Evolution of Butterfly Wing Patterns: An Integrative Approach. (Springer Nature, 2017).
Iwata, M., Tsutsumi, M. & Otaki, J. M. Developmental dynamics of butterfly wings: real-time in vivo whole-wing imaging of twelve butterfly species. Sci. Rep. 8, 1–16 (2018).
Pomerantz, A. F. et al. Developmental, cellular and biochemical basis of transparency in clearwing butterflies. J. Exp. Biol. 224, jeb237917 (2021).
Connahs, H., Rhen, T. & Simmons, R. B. Transcriptome analysis of the painted lady butterfly, Vanessa cardui during wing color pattern development. BMC genomics 17, 1–16 (2016).
Siddique, R. H., Gomard, G. & Hölscher, H. The role of random nanostructures for the omnidirectional anti-reflection properties of the glasswing butterfly. Nat. Commun. 6, 1–8 (2015).
Yoshida, A., Motoyama, M., Kosaku, A. & Miyamoto, K. Antireflective nanoprotuberance array in the transparent wing of a hawkmoth, Cephonodes hylas. Zoological Sci. 14, 737–741 (1997).
Pinna, C. S. et al. Mimicry can drive convergence in structural and light transmission features of transparent wings in Lepidoptera. Elife. 10, e69080 (2021).
Yoshida, A. & Aoki, K. Scale Arrangement Pattern in a Lepidopteran Wing. 1. Periodic Cellular Pattern in the Pupal Wing of Pieris rapae: (scale/cellular pattern/regular spacing/wing epidermis/Pieris rapae). Dev. Growth Differ. 31, 601–609 (1989).
Kimura, K., Kodama, A., Hayasaka, Y. & Ohta, T. Activation of the cAMP/PKA signaling pathway is required for post-ecdysial cell death in wing epidermal cells of Drosophila melanogaster. Development. 131, 1597–1606 (2004).
Kukalová-Peck, J. Origin of the insect wing and wing articulation from the arthropodan leg. Can. J. Zool. 61, 1618–1669 (1983).
BOPPRÉ, M. & Vane-Wright, R. Androconial systems in Danainae (Lepidoptera): functional morphology of Amauris, Danaus, Tirumala and Euploea. Zool. J. Linn. Soc. 97, 101–133 (1989).
Hall, J. P. & Harvey, D. J. A survey of androconial organs in the Riodinidae (Lepidoptera). Zool. J. Linn. Soc. 136, 171–197 (2002).
Rutowski, R. L. Male scent-producing structures in Colias butterflies. J. Chem. Ecol. 6, 13–26 (1980).
Omura, H. & Yotsuzuka, S. Male-specific epicuticular compounds of the sulfur butterfly Colias erate poliographus (Lepidoptera: Pieridae). Appl. Entomol. Zool. 50, 191–199 (2015).
Ômura, H., Itoh, T., Wright, D. M., Pavulaan, H. & Schröder, S. Morphological study of alar androconia in Celastrina butterflies. Entomol. Sci. 18, 353–359 (2015).
Tsai, C.-C. et al. Physical and behavioral adaptations to prevent overheating of the living wings of butterflies. Nat. Commun. 11, 1–14 (2020).
Futahashi, R. & Osanai-Futahashi, M. Pigments in insects. In Pigments, Pigment Cells and Pigment Patterns (eds Hashimoto, H., Goda, M., Futahashi, R., Kelsh, R. & Akiyama, T. Ch. 1, 3–43 (Springer, 2021).
Kayser, H. & Nimtz, M. Farnesyl biliverdins IXα are novel ligands of biliproteins from moths of the Noctuoidea superfamily: A chemosystematic view of the Lepidoptera. Insect Biochem. Mol. Biol. 78, 12–19 (2016).
Srygley, R. B. & Chai, P. Flight morphology of Neotropical butterflies: palatability and distribution of mass to the thorax and abdomen. Oecologia. 84, 491–499 (1990).
Woolley, J. T. Reflectance and transmittance of light by leaves. Plant physiol. 47, 656–662 (1971).
Constantino, L. M. & Salazar, J. A. A review of the Philaethria dido species complex (Lepidoptera: Nymphalidae: Heliconiinae) and description of three new sibling species from Colombia and Venezuela. Zootaxa. 2720, 1–27 (2010).
Eichenseer, H., Murphy, J. & Felton, G. Sequestration of host plant carotenoids in the larval tissues of Helicoverpa zea. J. Insect Physiol. 48, 311–318 (2002).
Sakudoh, T. et al. A CD36-related transmembrane protein is coordinated with an intracellular lipid-binding protein in selective carotenoid transport for cocoon coloration. J. Biol. Chem. 285, 7739–7751 (2010).
Pertuzatti, P. B., Sganzerla, M., Jacques, A. C., Barcia, M. T. & Zambiazi, R. C. Carotenoids, tocopherols and ascorbic acid content in yellow passion fruit (Passiflora edulis) grown under different cultivation systems. LWT-Food Sci. Technol. 64, 259–263 (2015).
Seely, G. & Meyer, T. The photosensitized oxidation of β-carotene. Photochem. Photobiol. 13, 27–32 (1971).
Wootton, R. J. Functional morphology of insect wings. Annu. Rev. Entomol. 37, 113–140 (1992).
Kayser, H. P. C. i. p. I. K., G.A. & Gilbert, L.I. (eds): Comparative Insect Physiology, Biochemistry, and Pharmacology, vol. 10. Pergamon Press, Oxford/New York, pp. 367–415.
Chattopadhyay, J. The structure and defensive efficacy of glandular secretion of the larval osmeterium in Graphium agamemnon agamemnon Linnaeus, 1758 (Lepidoptera: Papilionidae). Turk. J. Zool. 35, 245–254, doi:10.3906/zoo-0901-24 (2011).
Warren, A., Davis, K., Stangeland, E., Pelham, J. & Grishin, N. Illustrated lists of American Butterflies (North and South America) 21-XI-2017. Butterflies of America. (2017).
DeVries, P. J. The Butterflies of Costa Rica and Their Natural History. Papilionidae, Pieridae, Nymphalidae. (Princeton University Press, 1987).
Glassberg, J. A Swift Guide to Butterflies of Mexico and Central America. (Princeton University Press, 2018).

There is NO Competing Interest.

SupplementaryMovie01.mov
Another way to be green: pigmented liquid ink in butterfly wings – a novel form of colouration
SupplementaryMovie02.mov
Another way to be green: pigmented liquid ink in butterfly wings – a novel form of colouration
NishidaetalNaturecommunicationsSupplementaryinformation14iii2022.pdf
Another way to be green: pigmented liquid ink in butterfly wings – a novel form of colouration
rs.pdf
Reporting Summary

Download PDF

Version 1

posted

You are reading this latest preprint version

Another way to be green: pigmented liquid ink in butterfly wings – a novel form of colouration

Status:

Version 1

Abstract

Figures

Introduction

Results

Existence of green ink in wings

Wing structure in the area with and without green ink

Existence of living cells in the wing membrane

Identification of pigments responsible for the green colouration of the wings

Identification of green pigments in the hemolymph

Discussion

Why green ink?

How common is the ink system?

Materials And Methods

General information of the studied butterflies

Butterfly specimen supply

Observation of specimens

Measurement of the proportion of green area vs total area

SEM analysis

Observation of cuticular autofluorescence in wing sections

Nuclei staining

F-actin staining

Tracer experiment

Time-lapse fixed-point observation of wing colour change

Extraction, purification, and identification of green pigments

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1