Here I describe a sooty wing phenotype in offspring produced by a wild-inseminated female of Eurema mandarina (Lepidoptera: Pieridae) collected in Tanegashima Island, Japan in May, 2014. Seven (3 females and 4 males) out of 36 offspring had a sooty phenotype on their wings, which is likely due to a simple genetic mutation (most likely a single recessive allele in an autosome). The expression of this phenotype, restricted to the ventral side of the wings, was not uniform—more conspicuous in hindwings than in forewings, and the affected areas were variable among individuals. Microscopic observation revealed that the sooty phenotype was attributed to the increased proportion of pigmented scales.
In May, 2014, I collected female adults of the yellow butterfly Eurema mandarina (Lepidoptera: Pieridae) in Tanegashima Island (Kagoshima Prefecture, Japan). They were brought into the laboratory and allowed to lay eggs on fresh leaves of Lespedeza juncea var. subsessilis (Fabales: Fabaceae). Among 18 broods fed on an artificial diet containing Albizia julibrissin (Fabales: Fabaceae) leaves (Narita et al. 2007a) at 25°C with 16h : 8h (light : dark) photoperiod, one brood developed to adults consisting of individuals with a sooty phenotype in their wings (here referred to as sooty-wing or sw) as well as individuals with a wild phenotype (Table 1). The sw phenotype was expressed only on the ventral side of the wing (both forewing and hindwing) (Fig. 1; Fig. 2a), implying that this mutation may be governed by a gene responsible for dorsal/ventral separation of wing cells (such as apterous in Drosophila; Blair 1994, reviewed by Held 2002). The expression of this phenotype was more conspicuous in hindwings than in forewings, and the affected areas were variable among individuals—in the female hindwings for instance, the marginal area (code 013) and the interior area (codes 011 and 032) were conspicuously affected (Fig. 1a).
In E. mandarina, there are peculiar spots in the specific positions of the ventral side of the wings (both forewings and hindwings in males and females; Yata 1995). Each of these spots was also present in sw individuals (Fig. 3), indicating that the sw phenotype is not due to the mislocation of melanic scales which are to constitute these spots.
Because all the insects were reared at a constant temperature (25°C), the sw phenotype is not likely to be due to the thermal effects as has been observed in monarch butterflies, Danaus plexippus (Davis et al. 2005). The ratio of sw individuals (n = 7) to wild-type individuals (n = 29) was not significantly deviated from 1:3 (Table 1), a ratio predicted from the Mendelian segregation of an autosomal recessive allele sw, which was carried by both of their parents in the heterozygous condition (sw/+) (P > 0.05 by Fisher's exact probability test). On the other hand, the ratio of sw to wild-type individuals was significantly deviated from 1:1, a ratio predicted from the assumption that one of the parents was homozygous (sw/sw) (P = 0.0125 by Fisher's exact probability test). Although other possibilities such as polygenic effects remain, a single autosomal recessive allele is most likely to be responsible for the phenotype.
Close inspection under dissecting microscope revealed that the sw phenotype is due to the marked increase in number of blackish or brownish scales (Fig. 2c,d,e,f), which is probably due to the pigmentation by melanins. Thus, a disruption in the regulation of melanin biosynthesis may be responsible for the phenotype (True 2003). Darkening of color was first recognized at the late pupal stage, implying that the appearance of this phenotype is associated with the extension of invaginated wing discs in E. mandarina (Švácha 1992, Heming 2003).
Table 1.
Phenotype of 36 offspring produced by a wild-caught E. mandarina female
Fig. 1.
Wings with sw phenotype. (a) Females. Left forewing of a sw individual (code 011) was crumpled and was not suitable for photograph (missing panels). (b) Males. Red rectangles indicate wings showing sw phenotype (restricted to the ventral side). Bars represent 10 mm.
Fig2.
Individuals with sw phenotype (a,c,e) and wild-type individuals (b,d,f). (a,b) Butterflies drinking honey solutions. (c,d) Magnified image of peripheral areas of hindwings. Bars represent 1 mm. (e,f) Scales removed from the wing surface. Bars represent 200 m m.
Fig3.
Magnified images of some panels of Fig. 1. (a) Female left wings. (b) Male left wings. Arrows represent markings present in both wild-type and sw individuals. Bars represent 10 mm.
When an E. mandarina butterfly is at rest, the ventral side of the wings is more conspicuous than the dorsal side (Fig. 2a). Hence, sw phenotype would be prone to be subjected to natural selection, be it positive or negative. In some butterflies, adaptive significance of wing melanism as a thermoregulatory mechanism is established (Watt 1968, Roland 1982, Guppy 1986).
E. mandarina is a model species for the study of Wolbachia, a host manipulating intra-cellular bacteria (Werren et al. 2008, Kageyama et al. 2012). Two strains of Wolbachia were described in E. mandarina—a cytoplasmic-incompatibility-inducing wCI, being almost fixed except for the northernmost part of Japan (Hiroki et al. 2005; personal observation by DK); a feminizing wFem (always coexisting with wCI), being found only from Tanegashima Island and Okinawa-jima Island (Hiroki et al. 2004, Narita et al. 2007a, b, Kern et al. 2011). Here, I diagnosed all the 36 individuals of the brood (both sw and wild-type individuals) by PCR that was designed to specifically detect each of the strains (Hiroki et al. 2004), and found that they were all singly infected with wCI.
Acknowledgements
I am grateful to Masatsugu Hatakeyama for helpful discussion on wing development, and Markus Riegler, an anonymous reviewer, and the editor Keith S. Summerville for constructive comments to the early version of the manuscript.
