Ectopic Wnt6 expression induces twin—spot markings in the epidermis of the silkworm Bombyx mori – Potential molecular target to discover evolutionary mechanisms of adaptive body coloration in insects

Abstract

Animal body color patterning, as a naturally selected adaptive trait, plays critical ecological roles (e.g., warning coloration and mimicry). These patterns develop through spatiotemporally precise effector gene expression (e.g., pigment synthetases), regulated by pattern genes. However, the key pattern genes and the underlying mechanisms that govern the formation of marking patterns remain poorly defined. This study investigated the silkworm mutant, multilunar (L), which exhibited twin‐spot markings on sequential segments. We found that Wnt6 mRNA expression was significantly upregulated in the twin‐spot epidermal region. The cuticular protein RR‐1 motif 4 (CPR4) and cuticular protein RR‐1 motif 38 (CPR38) promoters were identified as core cis‐acting elements for epidermal specificity throughout larval development. CPR4‐driven Wnt6 overexpression induced twin‐spot markings on the epidermis of silkworm larvae, resembling the mutant L pattern, while exhibiting growth advantages compared with the CPR38‐driven system. Ultrastructural analysis indicated that there was a positive correlation between epidermal bulge structures and the formation of twin spots. Mechanistically, quantitative real‐time PCR (qRT‐PCR) analysis confirmed that epidermis‐specific overexpression of Wnt6 did not induce co‐activation of coclustered Wnt members. Furthermore, the analysis results of RNA‐seq and qRT‐PCR suggest that Wnt6 may promote the formation of the twin‐spot patterns by participating in hormone regulation, pigment production, and cell proliferation signaling pathways. This study, for the first time, establishes a heritable Wnt6 ectopic‐expression system that induces a twin‐spot phenotype resembling that of the mutant L, thereby providing a potential molecular target for elucidating the evolutionary mechanisms of adaptive body coloration in insects.

The CPR4 promoter drives ectopic Wnt6 expression in the silkworm (B. mori) epidermis, inducing twin‐spot markings at the dorsal end of body segments that resemble the L mutant. Bulge structures are closely associated with Wnt6‐mediated patterning, and Wnt6 may regulate hormonal responses, pigment synthesis, and cell proliferation pathways to promote epidermal pigmentation and local morphological changes, providing a potential molecular target for studying adaptive body color evolution.

Abbreviations

20E

20‐hydroxyecdysone

Ccnd2

cyclin D2

Chek2

checkpoint kinase 2

CPR38

cuticular protein RR‐1 motif 38

CPR4

cuticular protein RR‐1 motif 4

Ddc

dopa decarboxylase

DEG

differential gene expression

Dpyd

dihydropyrimidine dehydrogenase

DsRed2

Discosoma Red fluorescent protein 2

EcR

ecdysone receptor

EGFP

enhanced green fluorescent protein

epM

epidermis at the molting stage

epV3

epidermis at the feeding stage

EST

expressed sequence tag

Gadd45

growth arrest and DNA damage‐inducible 45

Grhpr

glyoxylate and hydroxypyruvate reductase

IE1

immediate‐early protein 1

JH

juvenile hormone

JHBP

juvenile hormone binding protein

KEGG

Kyoto Encyclopedia of Genes and Genomes

L1D1

the first day of first larval instar

L2D3

the third day of second larval instar

L3D2

the second day of third larval instar

L3D3

the third day of third larval instar

L4D1

the first day of fourth larval instar

L4D3

the third day of fourth larval instar

L5D3

the third day of fifth larval instar

L‐S5

the 5th segments of mutant L

L‐S6

the 6th segments of mutant L

Mthfd

methylenetetrahydrofolate dehydrogenase

NBAD

N‐β‐alanyldopamineN‐beta‐alanyldopamine

Npc1

npc intracellular cholesterol transporter 1

ORF

open reading frame

PAH

phenylalanine hydroxylase

Paics

phosphoribosylaminoimidazole succinocarboxamide synthetase

pCPR38>Wnt6

CPR38‐driven Wnt6 overexpressing silkworms

pCPR4>Wnt6

CPR4‐driven Wnt6 overexpressing silkworms

qRT‐PCR

quantitative real‐time PCR

Rchy1

ring finger and CHY zinc finger domain containing 1

Rp49

ribosomal protein 49

WT

wild‐type

WT‐S6

the 6th segment of WT Bombyx mori

Introduction

Genetic polymorphism is the basis for both convergent and divergent evolution of various phenotypes, and the diverse color patterns in caterpillars are closely related to natural selection, making them of significant value in evolutionary research [1]. Among lepidopteran insects, the domesticated silkworm, Bombyx mori (B. mori), is an ideal species for studying color pattern genes due to its rich variety of pigmentation pattern mutants in the larval stage [2], as well as the availability of genomic analysis and gene‐editing technologies [3]. Molecular genetic studies have identified several genes related to color pattern mutants in silkworms. For example, the dominant Zebra (Ze) mutant of B. mori exhibits black stripes on the anterior edge of each body segment, and research has shown that the Bmspz3 gene plays a major role in the deposition of melanin in the striped region of the Ze mutant [4]. Additionally, variations in the silkworm p allele are caused by differential expression of the apt‐like gene, leading to an increase in the expression of genes associated with melanin synthesis [5]. The black dilution (bd) mutant larvae of B. mori exhibit a dark gray phenotype, caused by an allelic mutation regulated by the transcription factor mamo [6]. Despite these findings, many genes involved in the regulation of insect body color remain undiscovered. Identifying novel regulatory genes and studying their molecular mechanisms will enhance our understanding of the molecular basis of animal coloration and the processes underlying phenotypic diversification.

A mutant strain of B. mori named ‘multi lunar (L)’ exhibits a dominant trait, characterized by multiple consecutive segments on the dorsal side displaying twin‐spots, which serve aposematic coloration [7]. Twin‐spot markings can also be observed in the 5th and 8th segments of the wild‐type (WT) B. mori and its ancestral species, Bombyx mandarina. The mutant L is believed to have spontaneously arisen in ancient China and was introduced to Japan over 100 years ago [7]. These mutants L are similar to the WT B. mori in terms of viability, mating, morphology, and growth. Previous studies have shown that the Wnt1 gene is a key regulatory factor in the formation of markings in the mutant L [8, 9]. Wnt family genes exhibit distinct expression patterns during both embryonic development and in various larval tissues, indicating functional diversification among family members [10, 11, 12, 13]. With the exception of Wnt1, the roles of other members of the Wnt gene family in the development of pigmentation patterns have yet to be fully elucidated.

To address this, we performed an in‐depth analysis of RNA‐seq data from nonspotted regions of WT and spotted regions of mutant L in larval epidermis. We found that within the Wnt gene cluster on chromosome 4 (Wnt1, Wnt6, Wnt9, and Wnt10), only Wnt1 and Wnt6 were upregulated in the mutant L compared with WT. Notably, Wnt6 expression was approximately two‐ to sevenfold higher than that of Wnt1 in the mutant L group, whereas Wnt9 and Wnt10 were undetectable in both groups. Furthermore, quantitative real‐time PCR (qRT‐PCR) confirmed that Wnt6 was specifically and highly expressed in the twin‐spot regions of the epidermis in mutants L when compared to both the non‐spotted epidermal regions of WT and the nonspotted regions of the same mutant. Importantly, epidermis‐specific overexpression of Wnt6 induced twin‐spot markings in a WT background that closely resembled those of the mutant L, indicating that differences in Wnt6 expression levels are tightly linked to the natural phenotypic variation in larval epidermal markings in the L strain. Recently, Ding et al. showed that electroporation‐mediated siRNA knockdown of epidermal Wnt6 in the same L strain markedly suppresses twin‐spot formation [14]. Together, these loss‐of‐function data and our gain‐of‐function results establish that Wnt6 is both necessary and sufficient for development of the twin‐spot markings.

This study aimed to elucidate the role of Wnt6 in twin‐spot pattern formation in the mutant L and its underlying regulatory mechanisms.

Results

Wnt6 is highly expressed in the epidermis of the mutant L

Twin‐spot markings were observed on the 5th and 8th segments of wild‐type (WT) B. mori. In contrast to WT B. mori, mutant L exhibited a dominant trait characterized by the formation of twin‐spot markings on the dorsal side of multiple consecutive segments (Fig. 1A). To investigate the molecular mechanisms underlying the formation of the larval color patterns in mutant L, we extracted epidermal RNA from the 6th segment of WT B. mori larvae (WT‐S6) at the third day of fifth larval instar (L5D3) stage and the 5th and 6th segments of mutant L larvae (L‐S5 and L‐S6) at the same stage, followed by RNA‐seq analysis. The result showed that within the Wnt gene cluster on chromosome 4 (including Wnt1, Wnt6, Wnt9, and Wnt10), only Wnt1 and Wnt6 were upregulated in the marked epidermal regions of both the L‐S5 and L‐S6 compared to the nonmarked epidermal region of WT‐S6 (Fig. 1B). This pattern is consistent with previous studies reporting Wnt1 upregulation in mutant L [8]. Within the marked epidermal region of mutant L, Wnt6 expression levels were approximately two‐ to sevenfold higher than those of Wnt1 (Fig. 1B). In contrast, Wnt9 and Wnt10 exhibited minimal or undetectable expression levels in both WT and mutant L samples (Fig. 1B). Further qRT‐PCR analysis confirmed that Wnt6 expression was significantly elevated in the marked epidermal regions of both the L‐S5 and L‐S6 compared with the nonmarked epidermal region of WT‐S6 (Fig. 1C). To further elucidate the spatial expression pattern of Wnt6 in mutant L, we conducted precise microdissection of the epidermis from both the spot‐marked region and the adjacent nonmarked region within the same body segment, followed by qRT‐PCR analysis. The result demonstrated that Wnt6 was expressed in both regions, with significantly higher expression levels observed in the spot‐marked region compared with the nonmarked region (Fig. 1D). Additionally, we examined the expression of Wnt6 in the marked epidermal region of mutant L larvae across different developmental stages. The result showed that Wnt6 expression exhibited periodic fluctuations. Specifically, its expression level was significantly elevated during the molting stage of each instar compared with the feeding and dormant stages (Fig. 1E). To further elucidate the genetic mechanisms contributing to the elevated expression of Wnt6 in mutant L, we carried out whole‐genome sequencing using embryo samples obtained from mutant L. Our findings identified multiple single‐nucleotide mutations within a putative cis‐regulatory element located approximately 2 kilobases upstream of the Wnt6 transcription start site (Fig. 1F). These mutations may alter transcription factor binding affinity or chromatin accessibility, thereby driving the cis‐regulatory upregulation of Wnt6 expression in the epidermis of the mutant L. Collectively, these findings suggest that the upregulation of Wnt6 may play a pivotal role in the development of the twin‐spot marking pattern in mutant L.

Fig. 1.

Wnt6 is highly expressed in the epidermis of B. mori mutant L larvae. (A) Representative images of wild‐type (WT) and mutant L Bombyx mori larvae on the third day of fifth larval instar (L5D3). The red dashed boxes represented the non‐marked epidermis of the sixth segment in WT larvae (WT‐S6) and the marked epidermis of the fifth and sixth body segments in mutant L larvae (L‐S5 and L‐S6). Scale bar: 0.5 cm. (B) Heatmap showing relative expression patterns of Wnt family genes in dorsal epidermal samples from WT‐S6, L‐S5, and L‐S6 larvae. Expression levels were quantified as Fragments Per Kilobase of transcript per Million mapped reads (FPKM) values from StringTie and normalized with Z‐score‐transformed log₂ fold changes, scaled to a 0–3 range for cross‐sample comparison. Color intensity indicates relative expression levels (0 = low, 3 = high). WT‐S6 serves as the baseline control for comparison with mutant samples. n = 3. (C) Quantification of Wnt6 expression by quantitative real‐time PCR (qRT‐PCR) in WT‐S6, L‐S5, and L‐S6 at the L5D3 developmental stage. Data were presented as mean ± SEM (n = 3), *P < 0.05, ***P < 0.001. Significance was determined using One‐way ANOVA with Tukey's multiple comparison test. (D) Quantification of Wnt6 expression in the epidermis of marked and non‐marked regions of mutant L (L‐S5) at the L5D3 stage by qRT‐PCR. Data were presented as mean ± SEM (n = 3), ***P < 0.001. Significance was determined using Student's t‐test. The blue arrow indicated the non‐marked region, and the red arrow indicated the marked region. (E) Quantitative analysis of Wnt6 expression in the marked epidermal regions of mutant L larvae at different stages from the 2nd (L2) to 5th (L5) instars by qRT‐PCR. Data were presented as mean ± SEM (n = 3), *P < 0.05, **P < 0.01. Significance was determined using One‐way ANOVA with Tukey's multiple comparison test. FS, feeding stage; DS, dormant stage; MS, molting stage. (F) Detection of mutation sites in the putative Wnt6 promoter region. Arrows indicated the types of base mutations. The relative expression levels of Wnt6 in panels C–E were normalized using ribosomal protein 49 (Rp49) as the internal reference gene.

The CPR4 and CPR38 promoters exhibit epidermis‐specific regulatory activity in B. mori

To investigate the role of Wnt6 in twin‐spot markings of mutant L larvae, we first screened and verified promoters with epidermis‐specific regulatory activity. According to the expressed sequence tag (EST) database, cuticular protein genes are predominantly expressed in the silkworm epidermis, with some expressed during both molting and feeding periods [15]. The relative mRNA expression levels of these cuticular protein genes in different larval tissues were analyzed using qRT‐PCR. Tissues from the head, Malpighian tube, midgut, middle silk gland, epidermis, fat body, testis, ovary, and wing disk of L5D3 larvae were used (Fig. 2A,B, Fig. S1). The results demonstrated that the mRNAs of cuticular protein RR‐1 motif 4 (CPR4) and cuticular protein RR‐1 motif 38 (CPR38) were exclusively and significantly expressed in the epidermal tissue (Fig. 2A,B). However, it is unclear whether the CPR4 and CPR38 gene promoters have the activity to drive gene expression in vivo. Therefore, the 2‐kb fragment lengths of the upstream regions of CPR4 and CPR38 genes were used to generate piggyBac‐based transgenic plasmids pBac[CPR4‐EGFP‐IE1‐DsRed2] and pBac[CPR38‐EGFP‐IE1‐DsRed2] (Fig. 2C). Both putative promoter sequences contained a TATA box (TATATAA). The enhanced green fluorescent protein (EGFP) gene was inserted immediately downstream of the CPR4 and CPR38 promoters, respectively, followed by a 240‐bp SV40 3′ untranslated region (UTR) terminator sequence (Fig. 2C). A Discosoma Red fluorescent protein 2 (DsRed2), derived from Discosoma sp., and driven by the Immediate‐Early Protein 1 (IE1) promoter, was used for selection (Fig. 2C). Additionally, a control plasmid was constructed, lacking the CPR4/CPR38 promoter sequences, with all other elements identical to those in the experimental plasmids (Fig. 2C).

Fig. 2.

The CPR4 and CPR38 promoters drive the ectopic expression of EGFP in the epidermis of B. mori. (A) Relative expression levels of the cuticular protein RR‐1 motif 4 (CPR4) in various tissues isolated from the third day of fifth larval instar (L5D3) larvae, including head, Malpighian tubules (MT), midgut (MG), middle silk gland (MSG), epidermis (Epi), fat body (Fat), testis (TE), ovary (OV), and wing disk. Relative expression of CPR4 was normalized to Rp49. (B) Relative expression levels of the cuticular protein RR‐1 motif 38 (CPR38) in various tissues isolated from L5D3 larvae. Relative expression of CPR38 was normalized to Rp49. (C) Schematic representation of the piggyBac vector. In the pBac[CPR4‐EGFP‐IE1‐DsRed2] and pBac[CPR38‐EGFP‐IE1‐DsRed2] plasmids, the enhanced green fluorescent protein (EGFP) expression was driven by the CPR4 and CPR38 promoters, respectively. The control plasmid did not drive EGFP expression via any promoter. All plasmids contained the Discosoma Red fluorescent protein 2 (DsRed2) expression cassette driven by the Immediate‐Early Protein 1 (IE1) promoter. (D) Representative images of control and transgenic silkworms at the embryonic, first instar, and fifth instar larval stages under optical and fluorescence microscopy. The scale bar was 1 mm for the embryonic and first‐instar stages, and 0.5 cm for the fifth‐instar stage. (E) Representative images and quantitative analysis of EGFP expression in various tissues of CPR4 transgenic fifth instar larvae. Scale bar: 0.5 cm. (F) Representative images and quantitative analysis of EGFP expression in various tissues of CPR38 transgenic fifth instar larvae. Scale bar: 0.5 cm. The relative expression levels of EGFP in panels E–F were normalized to Rp49. All data were presented as mean ± SEM (n = 3).

To evaluate the CPR4 and CPR38 promoter activity, we conducted in vivo genetic transformation. Among 400 preblastoderm embryos injected with pBac[CPR4‐EGFP‐IE1‐DsRed2], 195 hatched (48.8%), and 98 developed into adults. For pBac[CPR38‐EGFP‐IE1‐DsRed2], 178 hatched (44.5%), and 85 reached adulthood. In the control group, 183 hatched (45.8%), and 93 became adults. G0 individuals were sib‐mated or crossed with WT, yielding 50 G1 broods per condition. Fluorescence microscopy identified four DsRed2‐positive G1 broods in each experimental set (Table S1).

Transgenic silkworms carrying CPR4/CPR38 promoters showed no significant differences in hatching, larval growth, survival, or reproduction compared to controls, indicating no adverse effects on development. To explore their regulatory roles, we analyzed spatiotemporal EGFP expression via fluorescence stereomicroscopy, characterizing their in vivo expression dynamics. The CPR4 transgenic individuals did not show any green fluorescence during the embryonic stage but exhibited widespread green fluorescence throughout their bodies during the first‐ and fifth‐larval instars (Fig. 2D). In contrast, the CPR38 transgenic individuals exhibited green fluorescence during the embryonic stage as well as in the first‐ and fifth‐larval instars (Fig. 2D). The control group did not show any green fluorescence at any stage (Fig. 2D). These observations indicate that both CPR4 and CPR38 promoters possess regulatory activity in silkworms, with the CPR38 promoter demonstrating regulatory effects earlier than the CPR4 promoter. Furthermore, by observing the green fluorescence in various tissues (epidermis, testis, ovary, fat body, middle silk gland, and midgut) of fifth‐instar larvae from CPR4 and CPR38 transgenic individuals, we found that these two transgenes exhibited intense green fluorescence exclusively in the epidermis (Fig. 2E,F). qRT‐PCR analysis of EGFP gene expression in various tissues of the transgenic individuals revealed that, compared to other tissues, EGFP expression was significantly higher in the epidermis, with CPR38‐driven EGFP expression surpassing that driven by the CPR4 promoter (Fig. 2E,F). In conclusion, these findings emphasize that both the CPR4 and CPR38 promoters demonstrate regulatory activity in silkworms, and the regulatory capacity of the CPR38 promoter exceeds that of the CPR4 promoter.

Epidermal expression of Wnt6 driven by CPR4 or CPR38 promoters induces the formation of twin‐spot markings in B. mori larvae

To further validate the role of Wnt6 in the formation of twin‐spot markings on the epidermis of mutant L larvae, we utilized the CPR4 promoter to drive the ectopic expression of Wnt6 in the epidermis of Nistari. The transgenic vector pBac[IE1‐EGFP‐CPR4‐Wnt6], based on the PiggyBac transposon, was constructed first. The vector contains a Wnt6 open reading frame (ORF) driven by the CPR4 promoter, along with an EGFP fluorescent marker expression cassette driven by the IE1 promoter (Fig. 3A). The control plasmid lacks both the IE1 and CPR4 promoters (Fig. 3A). Among 400 preblastoderm embryos injected with pBac[IE1‐EGFP‐CPR4‐Wnt6], 185 embryos (46%) hatched successfully, and 90 larvae developed to the adult stage (Table S1). These G0 moths were mated with WT or sib‐mated moths, resulting in 45 G1 broods. Using fluorescence microscopy (Nikon AZ100), 3 G1 broods were identified to contain a total of 32 EGFP‐positive individuals (Table S1), which were designated as pCPR4>Wnt6.

Fig. 3.

The ectopic expression of Wnt6 in the epidermis driven by the CPR4 promoter promotes the formation of twin‐spot markings. (A) Schematic of the piggyBac vector. In the pBac[IE1‐EGFP‐CPR4‐Wnt6] plasmid, Wnt6 expression was driven by the CPR4 promoter, while EGFP expression was driven by the IE1 promoter. The control plasmid lacked the IE1 promoter and CPR4 promoter entirely. (B–E) Representative optical and fluorescence microscopy images of transgenic (pCPR4>Wnt6) and control silkworms at the first day of first larval instar (L1D1), the third day of second larval instar (L2D3), the third day of third larval instar (L3D3), and the first day of fourth larval instar (L4D1). All scale bars represented 1 mm, except for the magnified view in C, where the scale bar corresponded to 0.5 mm. (F) Representative images showing green fluorescence of EGFP expression and epidermal color pattern in the third day of fifth larval instar (L5D3) stage of control and pCPR4>Wnt6 silkworms. Scale bar: 0.5 cm. Red dashed lines indicated the magnified epidermal views in body segments 5–8 (S5–S8). Panels (B–F) showed representative images from six independent biological replicates.

Using a stereomicroscope, we systematically characterized the marking patterns on the dorsal side of multiple sequential segments in pCPR4>Wnt6 transgenic larvae throughout their developmental stages. The results revealed no significant difference in the epidermal color patterns between the control group and pCPR4>Wnt6 individuals at the first day of first larval instar (L1D1) and the third day of second larval instar (L2D3) (Fig. 3B,C). However, at the third day of third larval instar (L3D3) stage, the epidermal color of pCPR4>Wnt6 individuals was slightly darker compared to the controls (Fig. 3D). Notably, starting from the first day of fourth larval instar (L4D1) and encompassing the L5D3, pCPR4>Wnt6 larvae exhibited a distinct twin‐spot marking pattern, resembling that of mutant L, in contrast to the control group (Fig. 3E,F). All pCPR4>Wnt6 transgenic individuals exhibited the same phenotype (Fig. 4A). These findings demonstrate that epidermis‐specific ectopic expression of Wnt6, driven by the CPR4 promoter, facilitates the development of twin‐spot pigmentation patterns in B. mori larvae.

Fig. 4.

Effects of CPR4‐driven ectopic Wnt6 expression on the development of silkworm larvae, pupae and adults. (A) Representative images displaying phenotypic traits of control and CPR4‐driven Wnt6 overexpressing silkworms (pCPR4>Wnt6) at the third day of fifth larval instar (L5D3) stage within a single field of view. Scale bar: 1 cm. The blue arrow indicated control silkworms, and the red arrow indicated pCPR4>Wnt6 silkworms. (B) Comparison of developmental timing between control and pCPR4>Wnt6 silkworms across the first to fifth instar larvae (L1–L5) and the pupal stage (P). The x‐axis represented developmental time (hours; h), and the y‐axis represented the percentage of individuals at each developmental stage. Control: con. n = 25. (C, D) Morphological observations of pupae (C) and wing disks at the L5D3 stage (D) in the control and pCPR4>Wnt6 groups. Scale bar: 0.5 cm. (E, F) Morphological features of wings (E) and adults (F) from both groups. Scale bars: 1 cm. ♀ denoted female; ♂ denoted male. Panels (C–F) showed representative images from six independent biological replicates.

To further elucidate the effects of CPR4‐driven ectopic expression of Wnt6 on the growth and development of silkworms, we conducted systematic observations and statistical analyses of developmental timing and morphological characteristics at each metamorphic stage. The results showed that the developmental duration of the first‐instar larvae was comparable between control and pCPR4>Wnt6 individuals (Fig. 4B). However, by the end of the second‐instar larval stage, the developmental progression of pCPR4>Wnt6 individuals lagged behind the control group by 24 h, and this delay further increased to 48 h by the end of the pupal stage (Fig. 4B). Additionally, no significant differences were observed in the developmental size and morphology of the pupae (Fig. 4C), wing disks of the L5D3 larvae (Fig. 4D), adult wings (Fig. 4E), or adults (Fig. 4F) between control and pCPR4>Wnt6 individuals.

On this basis, we constructed another transgenic vector, pBac[IE1‐EGFP‐CPR38‐Wnt6], to compare the effects of different promoters driving Wnt6 expression on larval marking patterns. In this construct, Wnt6 expression was regulated by the CPR38 promoter (Fig. 5A). Except for the promoter, the vector is structurally identical to the CPR4 construct, ensuring direct comparability between the two (Fig. 5A). Following the microinjection of 400 preblastoderm embryos, 198 individuals hatched (49.5%), with 102 of them developing into adults (Table S1). The G0 generation moths were crossed with either WT or sibling individuals to generate 48 G1 broods. Fluorescence screening identified 44 EGFP‐positive individuals across three G1 broods, which were designated as pCPR38>Wnt6 (Table S1).

Fig. 5.

The CPR38 promoter‐driven ectopic expression of the Wnt6 gene in the epidermis triggers the formation of twin‐spots in larvae. (A) Schematic of the piggyBac vector construction. In the pBac[IE1‐EGFP‐CPR38‐Wnt6] plasmid, the expression of Wnt6 was controlled by the CPR38 promoter, while EGFP expression was driven by the IE1 promoter. The control plasmid lacked both the IE1 and CPR38 promoters. (B) Morphological characteristics at the second day of the third larval instar (L3D2) in control and CPR38‐driven Wnt6 overexpressing silkworms (pCPR38>Wnt6) groups. Red dotted lines indicated magnified regions of the epidermis on the dorsal side of the 7th and 8th segments. Scale bar: 0.5 mm. (C) Representative images of epidermal markings in the third day of fifth larval instar (L5D3) stage of the mutant L and pCPR38>Wnt6 transgenic silkworms, together with green fluorescence of EGFP expression in the transgenic larvae. Red dashed lines indicated magnified regions of the epidermis on the dorsal side of the 6th and 7th segments. Scale bar: 0.5 cm. Panels (B, C) showed representative images from six independent biological replicates. (D) Representative images illustrating the phenotypic traits of control and pCPR38>Wnt6 larvae at the L5D3 developmental stage, captured within a single field of view. Scale bar: 1 cm. The blue arrow indicated control silkworms, red arrow indicated pCPR38>Wnt6 silkworms. (E) Survival curves of control and pCPR38>Wnt6 silkworms across the first to fifth instar larvae (L1–L5) and the pupal stage (P). n = 30.

Bright‐field stereomicroscopic examination of larval epidermis demonstrated that pCPR38>Wnt6 larvae exhibited slightly darker epidermal pigmentation compared to the control group at the second day of third larval instar (L3D2), closely resembling the phenotype observed in pCPR4>Wnt6 larvae at the L3D3 stage (Fig. 5B). More importantly, twin‐spot markings appeared on the epidermis of pCPR38>Wnt6 larvae at the L5D3 stage, displaying a pattern analogous to mutant L (Fig. 5C). All pCPR38>Wnt6 transgenic individuals demonstrated identical phenotypic characteristics (Fig. 5D). However, unlike pCPR4 > Wnt6, pCPR38 > Wnt6 positive individuals began to die at the fourth instar, with survival rates stabilizing at approximately 50%–60% during the pupal stage, significantly lower than the control group (Fig. 5E). Despite these individuals originating from independent transformation events, they exhibited high consistency in phenotypic characteristics, timing of death, and survival rates. Therefore, the increased mortality is unlikely to be caused by random insertion site effects.

In conclusion, both the CPR4 and CPR38 promoters are capable of effectively driving Wnt6 expression to induce the formation of twin‐spot patterns. However, the stronger or broader activity of the CPR38 promoter may cause additional physiological stress or interfere with unknown regulatory pathways, thereby leading to a reduction in the overall survival rate of larvae.

The epidermal bulge structure is associated with the formation of twin‐spot markings in larvae

To explore the positioning mechanism of the twin‐spot markings, we performed a combined analysis using optical microscopy and scanning electron microscopy on the epidermis of larvae at the L3D3 stage. The results revealed that in the dorsal epidermis of the 5th to 8th segments in both the control and pCPR4>Wnt6 groups, a distinct convex morphological structure with positional conservation (referred to as the ‘bulge’) was present. This structure exhibited a beetle‐like shape and was accompanied by three to four characteristic bristles. Notably, the bulge displayed a strict spatial colocalization pattern with the crescent‐shaped and star‐shaped markings on the 5th and 8th segments (Fig. 6A,B). These findings suggest that the epidermal bulge structures during the early larval development stage may contribute to the spatial distribution of twin‐spot markings at a later stage.

Fig. 6.

Bristles and bulge structures on the epidermis of control and pCPR4>Wnt6 transgenic silkworm larvae at the L3D3 stage. (A) Epidermal images of the S5‐S8 in control and CPR4‐driven Wnt6 overexpressing silkworms (pCPR4>Wnt6) transgenic silkworm larvae at the third day of third larval instar (L3D3) developmental stage. Red dashed boxes demarcate crescent markings in S5 and star spot markings in S8. Scale bar: 1 mm. (B) Scanning electron microscopy (SEM) images of bristles and bulge structures on the epidermis of the S5‐S8 in control and pCPR4>Wnt6 transgenic silkworm larvae at the L3D3 stage. Yellow arrows indicate bristles; red arrows indicate bulges. Scale bar: 50 μm; magnification: 300×. Panels (A, B) show representative images from six independent biological replicates.

Starting from L4D1, pCPR4>Wnt6 larvae exhibited significant twin‐spot markings on sequential segments, compared with the control group. We then examined the epidermal characteristics of larvae from both groups during this period. The results showed that bulge structures on the epidermis of larvae in both groups were detectable only at the spot marking region, with no such structures found in the non‐spot region (Fig. 7A,B). However, the characteristic bristles on the epidermis of the 5th to 8th body segments (S5–S8) were present in both groups (Fig. 7A,B). To elucidate the mechanisms underlying the initiation of twin‐spot pattern formation at the fourth instar following ectopic Wnt6 expression, we collected epidermal samples from the marked regions of pCPR4>Wnt6 transgenic silkworms across the second to fifth instars, encompassing the dormancy, molting, and feeding periods, to evaluate Wnt6 transcription levels. The results showed that Wnt6 expression levels were significantly higher during the molting stage of the fourth instar compared to other stages (Fig. 7C), suggesting that the abnormally high expression of Wnt6 at this stage may contribute to the initiation of twin‐spot formation. Additionally, epidermal samples were collected from both the marked and non‐marked regions within the same body segment of pCPR4>Wnt6 transgenic silkworms, followed by qRT‐PCR analysis. No significant differences in Wnt6 transcript levels were observed between these regions (Fig. 7D), indicating that Wnt6 is broadly expressed in the epidermis, consistent with the widespread regulatory activity of the CPR4 promoter. In summary, the twin‐spot patterns induced by the ectopic expression of Wnt6 are closely related to the epidermal bulge structures in terms of spatial distribution. Its formation not only depends on the expression level of Wnt6 in the epidermis but may also be regulated by the specific microenvironment provided by the local bulge structure.

Fig. 7.

At the L4D1 stage, the bulge structures on the epidermis of both control and pCPR4>Wnt6 transgenic silkworm larvae are closely associated with the formation of twin‐spot patterns. (A) Epidermal patterns on body segments 5–8 (S5–S8) of control and CPR4‐driven Wnt6 overexpressing silkworms (pCPR4>Wnt6) at the first day fourth larval instar (L4D1) stage. Scale bar: 1 mm. (B) Scanning electron microscopy (SEM) images showing bristles and bulge structures on the epidermis of the S5–S8 in control and pCPR4>Wnt6 transgenic larvae at the L4D1 stage. Yellow arrows indicated bristles; red arrows indicated bulges. Scale bar: 50 μm; magnification: 300×. Panels (A, B) showed representative images from six independent biological replicates. (C) Quantification of Wnt6 expression levels in pCPR4>Wnt6 transgenic silkworms across developmental stages from the second to fifth instar larvae (L2–L5) by qRT‐PCR. Data were presented as mean ± SEM (n = 3), ***P < 0.001. Significance was determined using Student's t‐test. DS, dormant stage; FS, feeding stage; MS, molting stage. (D) Quantification of Wnt6 expression levels in marked and non‐marked regions of the epidermis in pCPR4>Wnt6 silkworms at the L4D1 stage by qRT‐PCR. The relative expression levels of Wnt6 in panels C, D were normalized to Rp49. Data were presented as mean ± SEM (n = 3). ns for no significant difference. Significance was determined using Student's t‐test. Rp49 was employed as the reference gene in all qRT‐PCR assays.

Wnt6 upregulates epidermal pigment synthesis‐related gene expression

To investigate the mechanism by which Wnt6 induces the formation of larval spot pigmentation, we analyzed the expression of key genes involved in the pigment synthesis pathway in the epidermis of the non‐spot areas in the control group and the spot areas in the pCPR4>Wnt6 group at various developmental stages. These stages included the first‐instar (L1) feeding stage, fourth‐instar (L4) dormant stage, L4 molting stage, the third day of fourth larval instar (L4D3) feeding stage, fifth‐instar (L5) dormant stage, L5 molting stage, and L5D3 feeding stage. All epidermal samples were collected from the dorsal side of the sixth segment (S6).

In the melanogenesis pathway, the TH gene encodes tyrosine hydroxylase, a rate‐limiting enzyme located upstream in the pathway, which plays a crucial role in melanogenesis and body pigmentation [16]. The Ebony gene, through its product N‐β‐alanyl dopamine (NBAD) synthase, converts β‐alanine and dopamine into NBAD, thereby promoting the formation of light‐colored pigments [17]. The Black gene, which encodes aspartate decarboxylase, catalyzes the conversion of aspartate to β‐alanine, thereby contributing to epidermal pigment synthesis [18]. The Yellow family genes are key regulators of epidermal pattern formation. Among these, Drosophila Yellow‐f is identified as dopachrome‐conversion enzymes responsible for catalyzing the transformation of dopachrome into 5,6‐dihydroxyindole within the melanization pathway [19]. Additionally, Laccase2, a phenol oxidase, promotes melanogenesis [20]. Based on this, we used qRT‐PCR to measure the expression levels of Wnt6, TH, Ebony, Black, Yellow‐f, and Laccase2. This aimed to uncover the molecular mechanisms by which ectopic Wnt6 expression promotes the formation of epidermal twin‐spot markings in larvae.

Our findings demonstrated that, compared to the control group, the expression of the Wnt6 gene in the epidermis of pCPR4>Wnt6 larvae was upregulated across all developmental stages (Fig. 8A). Notably, during the molting stages, its expression level exhibited a more pronounced upregulation trend relative to the feeding and dormant stages (Fig. 8A). This expression pattern showed a significant correlation with stage‐specific activation of enzyme genes associated with pigment synthesis pathways. In comparison with controls, the expression levels of the TH and Ebony genes in the pCPR4>Wnt6 larval epidermis were significantly increased during both the L4 and L5 molting stages (Fig. 8B,C). The Black gene displayed significant upregulation during the L5 molting stage (Fig. 8D), while the Yellow‐f gene showed a marked elevation specifically during the L4 molting stage (Fig. 8E). Furthermore, Laccase2 expression in the pCPR4>Wnt6 larval epidermis was significantly enhanced during the L5 molting stage compared to controls, although it was also upregulated during the L4 dormant stage (Fig. 8F). Furthermore, we analyzed EGFP expression driven by the CPR4 promoter and observed a similar periodic pattern, with peak expression during the molting stage (Fig. 8G). This suggests that the cyclical expression of Wnt6 is at least partly governed by the temporal regulation inherent to the CPR4 promoter. Overall, these findings suggest that Wnt6 expression driven by the CPR4 promoter may enhance pigment synthesis gene expression during the molting stage, thereby coordinating the formation of twin‐spot pigmentation on the epidermis and synchronizing cuticle remodeling with pigment metabolism.

Fig. 8.

Expression levels of Wnt6 and pigment synthesis‐related genes in the epidermis of control and pCPR4>Wnt6 transgenic silkworm larvae at various developmental stages. (A–F) Quantification of mRNA expression levels of Wnt6 and pigment synthesis‐related genes (TH, Ebony, Black, Yellow‐f, and Laccase2) in the epidermis of control and CPR4‐driven Wnt6 overexpressing silkworms (pCPR4>Wnt6) larvae at various developmental stages, including the feeding stage of the first instar (L1 FS), dormant stage of the fourth instar (L4 DS), molting stage of L4 (L4 MS), feeding stage on the third day of the fourth instar (L4D3 FS), dormant stage of the fifth instar (L5 DS), molting stage of L5 (L5 MS), and feeding stage of L5D3 (L5D3 FS), by qRT‐PCR. Expression levels were normalized to Rp49. Data were presented as the mean ± SEM (n = 3), *P < 0.05, **P < 0.01, ***P < 0.001. Significance was determined using Two‐way ANOVA with Tukey's multiple comparison test. (G) Quantification of EGFP mRNA expression levels in the epidermis of pCPR4>EGFP (CPR4‐driven EGFP expression) transgenic silkworm larvae at the above developmental stages by qRT‐PCR. EGFP expression levels were normalized to Rp49. Data were presented as the mean ± SEM (n = 3). ***P < 0.001. Significance was determined using One‐way ANOVA with Tukey's multiple comparison test. Rp49 was used as the reference gene in all qRT‐PCR assays.

Ectopic expression of Wnt6 does not upregulate other Wnt genes within the same gene cluster

The Wnt gene family typically exhibits a compact, clustered arrangement within the genomes of most species, and these gene clusters are highly conserved across evolution. Genome analysis of the silkworm reveals a conserved cluster comprising Wnt9, Wnt1, Wnt6, and Wnt10 on chromosome 4 [10]. To investigate the effect of high Wnt6 expression driven by the CPR4 promoter on other Wnt genes in this cluster, we used qRT‐PCR to measure changes in the expression of Wnt6, Wnt1, Wnt9, and Wnt10 across various tissues, including the Malpighian tubules, midgut, middle silk gland, dorsal epidermis of the sixth segment, fat body, testes, ovaries, and wing disks, in the control and pCPR4>Wnt6 larvae at the L5D3 stage. The results showed that, compared to the control group, Wnt6 was specifically and highly expressed in the epidermis of pCPR4>Wnt6 larvae (Fig. 9A). In contrast, Wnt1 expression was significantly reduced in the Malpighian tubules, midgut, and epidermis of pCPR4>Wnt6 larvae (Fig. 9B). Wnt9 expression was significantly increased only in the Malpighian tubules of pCPR4>Wnt6 larvae (Fig. 9C), and Wnt10 expression was elevated only in the testes of pCPR4>Wnt6 larvae (Fig. 9D). These results indicate that the role of ectopic Wnt6 expression in the formation of epidermal twin‐spot patterns is not mediated by upregulating other Wnt genes within the same gene cluster.

Fig. 9.

Ectopic expression of Wnt6 in the larval epidermis does not promote the expression of other Wnt genes within the same cluster. (A–D) qRT‐PCR analysis of Wnt6 (A), Wnt1 (B), Wnt9 (C), and Wnt10 (D) expression levels in various tissues, including Malpighian tubules (MT), midgut (MG), middle silk gland (MSG), epidermis (Epi), fat body (Fat), testis (TE), ovary (OV), and wing disk, in control and CPR4‐driven Wnt6 overexpressing silkworms (pCPR4>Wnt6) larvae at the third day of fifth larval instar (L5D3) stage. Rp49 was used as the reference gene. All data were presented as the mean ± SEM (n = 3), *P < 0.05, ***P < 0.001. Significance was determined using Student's t‐test.

RNA‐seq and qRT‐PCR analysis of epidermis from pCPR4 >Wnt6 larvae

To elucidate the molecular mechanisms underlying Wnt6‐induced epidermal spot formation in B. mori, we conducted RNA‐seq analysis of nonspot epidermal regions in control larvae and spot regions in pCPR4>Wnt6 transgenic larvae of the sixth segment at both the fifth‐instar molting stage and the L5D3 developmental stage. At the molting stage, compared with the control group, the pCPR4>Wnt6 group showed 2031 significantly upregulated and 370 significantly downregulated genes (Fig. 10A), whereas at L5D3, 191 genes were significantly upregulated and 105 were downregulated (Fig. 10B). These results indicate that Wnt6 overexpression causes more extensive transcriptional reprogramming during the molting stage.

Fig. 10.

Effects of ectopic Wnt6 expression in the epidermis of silkworm larvae on downstream genes and associated enriched pathways during the fifth‐instar molting and L5D3 stages. (A, B) Volcano plots of differentially expressed genes (DEGs) in the dorsal epidermis of the body S6 in larvae, comparing control and CPR4‐driven Wnt6 overexpressing silkworms (pCPR4>Wnt6) at the fifth‐instar molting stage (L5 MS) (A) and the third day of fifth larval instar (L5D3) stage (B). n = 3. Red and blue dots represented significantly upregulated and downregulated genes, respectively (|log₂FC| ≥ 2, Padjust <0.05), while gray dots denoted non‐significant genes (Padjust ≥ 0.05 or |log₂FC| < 2). (C, D) Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis of DEGs in the dorsal epidermis of the S6 in control and pCPR4>Wnt6 larvae during the L5 MS (C) and at the L5D3 stage (D). n = 3. The red solid‐line boxes represented pathways exclusively enriched during the L5 MS, while the blue solid‐line boxes indicated pathways co‐enriched in both the L5 MS and the L5D3 stage. (E) Venn diagram of KEGG pathway enrichment for DEGs in the control and pCPR4>Wnt6 groups during the L5 MS and the L5D3 stage. (F) Quantification of representative DEGs from enriched pathways by qRT‐PCR. Expression levels were normalized to Rp49 and presented relative to the control group at the L5 MS stage. Data were presented as mean ± SEM (n = 3); *P < 0.05, **P < 0.01, ***P < 0.001. Significance was determined using Two‐way ANOVA with Tukey's multiple comparison test. (G) Quantification of representative p53 pathway genes by qRT‐PCR in non‐spot regions of WT larval epidermis and spot regions of mutant L larval epidermis during the L5 MS. Expression levels were normalized to Rp49 and presented relative to the WT control group. Data were presented as mean ± SEM (n = 3), *P < 0.05, ***P < 0.001. Significance was determined using Student's t‐test. (H) Measurement of 20‐hydroxyecdysone (20E) and juvenile hormone (JH) levels in the hemolymph of control and pCPR4>Wnt6 transgenic larvae during the L5 MS. Data were presented as mean ± SEM (n = 3); *P < 0.05. Significance was determined using Student's t‐test.

Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis demonstrated that the differentially expressed genes (DEGs) commonly present across both developmental stages were predominantly enriched in three functional categories: Pigment precursor metabolism (Tyrosine metabolism, Phenylalanine metabolism, Glycine–Serine–Threonine metabolism), which provides substrates for melanin and pteridine pigment synthesis; Cell proliferation support (Purine metabolism, One‐carbon pool by folate), which supplies nucleotides and methyl donors required for sustained local proliferation and methylation‐dependent regulatory processes; and Hormonal lipid precursor supply (Pantothenate and CoA biosynthesis), which furnishes lipid precursors essential for the biosynthesis of ecdysteroids and juvenile hormone (JH) (Fig. 10C–E). Based on these findings, we selected representative differentially expressed genes associated with the pathways for qRT‐PCR validation, aiming to elucidate their functional roles in Wnt6‐mediated spot formation. DOPA decarboxylase (Ddc) catalyzes the conversion of DOPA to dopamine, which represents a key step in melanin synthesis [21]; phenylalanine hydroxylase (PAH) catalyzes the conversion of phenylalanine to tyrosine, thereby influencing the rate of melanin production [22]; glyoxylate and hydroxypyruvate reductase (Grhpr) contributes glycine for the synthesis of pteridine pigments [23]; phosphoribosylaminoimidazole succinocarboxamide synthetase (Paics) and methylenetetrahydrofolate dehydrogenase (Mthfd) are involved in the production of nucleotides and methyl donors through purine synthesis and folate‐mediated one‐carbon metabolism [24, 25, 26]; dihydropyrimidine dehydrogenase (Dpyd) participates in the β‐alanine to pantothenate to CoA pathway, supplying lipid precursors essential for the biosynthesis of ecdysteroids and JH [27]. qRT‐PCR results demonstrated that these genes exhibited significantly higher expression levels during the 5th instar molting period and at the L5D3 stage in the pCPR4>Wnt6 group compared to the control group. In both the control group and the pCPR4>Wnt6 group, the expression levels of these genes were significantly higher during the 5th‐instar molting period compared to the L5D3 stage. (Fig. 10F). These findings suggest that Wnt6 may regulate pigmentation, local cell proliferation, and hormonal changes during the 5th‐instar molting period and L5D3 stage, with the molting period likely being critical for pattern formation.

Notably, pathways specifically enriched during the fifth‐instar molting period, but not at L5D3, include the p53 signaling pathway, which suppresses cell proliferation by inducing cell cycle arrest and apoptosis [28, 29], and the cholesterol metabolism pathway, which supplies precursors for steroid hormone synthesis [30] (Fig. 10C–E). To further verify these findings, we examined the expression of genes related to the p53 pathway, such as cyclin D2 (Ccnd2; a G1/S transition cyclin [31]) and ring finger and CHY zinc finger domain containing 1 (Rchy1; a p53 E3 ubiquitin ligase [32, 33, 34]), as well as genes involved in hormone regulation, including juvenile hormone binding protein (JHBP) [35], ecdysone receptor (EcR) [36, 37, 38], and npc intracellular cholesterol transporter 1 (Npc1; a sterol transporter) [39, 40, 41]. qRT‐PCR analysis revealed significantly elevated expression levels of these genes in the pCPR4>Wnt6 group compared to the control group during the fifth‐instar molting period, whereas no significant differences were observed at L5D3 (Fig. 10F). Moreover, the expression levels of these genes were significantly higher during the fifth instar molting period than at the L5D3 stage in both the control group and the pCPR4>Wnt6 group (Fig. 10F). These results suggest that during the fifth instar molting period of pCPR4>Wnt6 silkworms, Wnt6 may promote localized epidermal cell proliferation through negative regulation of the p53 signaling pathway and may also modulate hormone signaling to induce epidermal remodeling, thereby facilitating pattern formation. In addition, qRT‐PCR analysis of the p53 signaling pathway genes growth arrest and DNA damage‐inducible 45 (Gadd45; a downstream target of p53 responsive to DNA damage) [42, 43], checkpoint kinase 2 (Chek2; an upstream activator of p53) [44, 45], and p53 itself was conducted on the non‐marked epidermis of WT silkworms and the marked epidermis of the mutant L during the 5th‐instar molting stage. The results demonstrated that, compared with the WT group, the expression levels of these genes were markedly downregulated in the mutant L group (Fig. 10G). This finding suggests that the high endogenous expression of Wnt6 in the mutant L likely promotes the formation of epidermal bulges in spot‐patterned regions through suppression of p53 signaling. Hemolymph analysis conducted during the fifth‐instar molting stage demonstrated that JH and 20‐hydroxyecdysone (20E) levels in the pCPR4>Wnt6 group were significantly elevated relative to those in the control group (Fig. 10H), indicating that Wnt6‐mediated regulation of the JH/20E axis contributes to pigmentation patterning. Collectively, these findings indicate that the expression of ectopic Wnt6 likely drives the formation of twin‐spot patterns in B. mori through a “hormone‐pigment‐proliferation” regulatory network.

Discussion

Our study established three key advances in developmental patterning. First, we characterized the CPR4 promoter as a pan‐developmental regulatory element in the silkworm larval epidermis. Second, we observed that epidermis‐specific overexpression of Wnt6 promoted the formation of twin‐spot marking on larval dorsal body segments, phenocopying the mutant L pattern. Third, our results demonstrated a close association between epidermal bulge structures and Wnt6‐mediated generation and spatial patterning of twin‐spot markings across the larval epidermis. At the mechanistic level, Wnt6 orchestrated epidermal gene expression without upregulating other Wnt genes in the same cluster. It was involved in regulating hormone‐responsive factors, melanogenesis components, and cell proliferation signaling pathways. This regulatory network is likely involved in pigment biosynthesis and bulge morphogenesis, potentially leading to the twin‐spot phenotype.

The cuticular patterns of lepidopteran larvae exhibit dynamic remodeling characteristics during the molting process, in which epidermal proteins play a crucial role in cuticle formation [46, 47]. In comparison with the epV3 (epidermis at the feeding stage) library, the epM (epidermis at the molting stage) library exhibits significantly higher diversity and abundance of epidermal protein gene expression, whereas RR‐1‐type epidermal proteins predominate in the epV3 library [15]. Notably, the glycine‐rich protein CPG25, which contains the CPG domain, is specifically and highly expressed in the epM library [15]. Recent studies have verified that the CPG25 promoter exhibits specific regulatory activity in the epidermal tissue during the molting period of the silkworm [48]. However, epidermal‐specific promoters that retain regulatory activity during both the molting and feeding periods have not yet been identified. This technical gap directly hinders the in‐depth exploration of the functions of regulatory genes involved in epidermal morphogenesis. In this study, based on the gene expression data published by Fujiwara et al. from the epM and epV3 libraries [15], we screened via qRT‐PCR and found that the CPR4 and CPR38 genes were specifically and highly expressed in the epidermis. Additionally, our research identified, for the first time, that the promoters of these two genes possessed regulatory activity in the epidermis of silkworm larvae at all developmental stages. Notably, the activity of the CPR38 promoter was significantly higher than that of the CPR4 promoter. In functional assays, both CPR4 and CPR38 promoters effectively drove ectopic Wnt6 expression, inducing twin‐spot patterns on the larval epidermis resembling those of the mutant L. This is the first time the formation mechanism of the spot pattern in the mutant L has been visually revealed in vivo through genetic regulation. While the CPR38 promoter displayed stronger regulatory activity in vivo, excessive promoter activity significantly decreased silkworm survival during Wnt6 expression, probably due to the activation of unknown negative regulatory factors. This suggests that the intensity of promoter‐driven gene expression should not be maximized indiscriminately. Instead, tools for tissue‐specific gene function analysis must strike a balance between driving efficiency and biological adaptability. Previous studies have shown that certain CPR family genes exhibit periodic expression in the larval epidermis at different developmental stages, closely associated with fluctuations in ecdysteroid levels. Specifically, elevated ecdysteroid levels suppress the transcription of some CPR genes, whereas their expression resumes and reaches a peak as hormone levels decline [15]. Moreover, several CPR genes can be induced by ecdysteroid pulses through the action of transcription factors such as FtzF1 [49]. These findings suggest that the promoter regions of some CPR genes may contain ecdysone response elements and related regulatory sites, thereby enabling temporal regulation by hormonal signals. Our qRT‐PCR data further revealed that Wnt6, driven by the CPR4 promoter, exhibited periodic high expression during molting stage, which may reflect the sensitivity of the CPR4 promoter to ecdysteroids. Collectively, these results suggest that CPR promoters not only drive epidermis‐specific gene expression but also may mediate hormone‐dependent temporal regulation, highlighting their potential multifunctional regulatory roles.

The Lepidoptera order, known for its extensive species diversity and larval pigmentation variability, serves as an excellent model for studying color pattern evolution [50, 51]. This study investigates the B. mori mutant L, characterized by distinctive twin‐spot markings on consecutive body segments. Previous research by Fujiwara et al. demonstrated that Wnt1 is upregulated in the epidermis of the twin‐spot regions in mutant L, with ectopic Wnt1 expression inducing additional larval pigmentation via electroporation‐mediated somatic transgenesis [8]. Wnt family genes exhibit varied expression patterns during embryonic and larval development, indicating functional specialization [10, 11, 12, 13]. However, the role of other Wnt genes, beyond Wnt1, in B. mori larval marking formation remains underexplored. Our study revealed significantly elevated Wnt6 expression in the epidermal spotted regions of the mutant L strain, surpassing levels observed in both WT and adjacent non‐spotted regions within the same mutant. Moreover, Wnt6 expression in these spotted regions consistently exceeded that of Wnt1. These results indicate a strong correlation between the spatially restricted expression of Wnt6 and the formation of spot patterns in mutant L. Consistent with this, ectopic expression of Wnt6 driven by an epidermis‐specific promoter (CPR4) was sufficient to induce spot patterns resembling those of mutant L. Interestingly, qRT‐PCR analysis at the L5D3 stage showed that Wnt1 expression was lower in the epidermal spotted regions of the pCPR4>Wnt6 group relative to non‐spotted control regions. This finding contrasts with Fujiwara et al. [8], who observed elevated Wnt1 expression in mutant L, attributable to a Wnt1 promoter mutation. This discrepancy is likely explained by the intact wild‐type Wnt1 promoter in our overexpression system. Additionally, Wnt1 expression is naturally low at the L5D3 feeding stage in both control and transgenic groups, unlike its peak during the molting stage reported by Fujiwara et al. [8]. Thus, the apparent reduction in Wnt1 expression in our system most likely reflects its low baseline rather than direct suppression by Wnt6. Mechanistically, Wnt6 may influence Wnt1 expression indirectly, potentially through competition for shared receptors or by perturbing feedback loops within the Wnt signaling network. Additionally, studies in butterflies show that Wnt1 influences pigmentation in both red and white regions of butterfly hindwings, whereas Wnt6 specifically affects red region. Knocking down Wnt6 via siRNA selectively reduces red region pigmentation [52]. These findings support the view that Wnt6 functions in a region‐ and cell‐specific manner without broadly affecting Wnt1 expression. In summary, these results demonstrate that ectopic expression of Wnt6 is sufficient to induce epidermal spot formation without concomitant upregulation of other Wnt genes within the same genomic cluster, thereby providing important insights into the molecular mechanisms underlying the evolutionary diversification of body‐color patterns. Recently, Ding et al. showed that electroporation‐mediated siRNA knockdown partially suppresses epidermal Wnt6 expression in the mutant L, markedly attenuating twin‐spot formation [14]. This loss‐of‐function evidence, together with our gain‐of‐function results, strongly supports a critical regulatory role for Wnt6 in twin‐spot development. Nevertheless, several limitations remain: epidermis‐specific Wnt6 knockout data generated by genome editing are currently unavailable, the epistatic relationship with Wnt1 (e.g., whether Wnt6 overexpression can rescue the phenotype in a Wnt1‐knockdown background) remains untested, and the precise spatiotemporal expression pattern of Wnt6 in the mutant L epidermis awaits higher‐resolution characterization. To address these gaps, future studies will integrate CRISPR/Cas9‐mediated genome editing, optimized protocols for terminating embryonic diapause, and improved in situ hybridization techniques to systematically determine, in a stable genetic background, the absolute requirement for Wnt6, its detailed spatial and temporal expression dynamics, and its functional and positional relationship with Wnt1, thereby further refining the regulatory network that governs pigmentation pattern formation.

In this study, we found that overexpressing Wnt6 in the B. mori epidermis resulted in widespread expression, but pigment deposition was restricted to specific regions. qRT‐PCR analysis showed that Wnt6 transcription levels did not differ significantly between spotted and nonspotted regions within the same segment of pCPR4>Wnt6 transgenic larvae, indicating broad Wnt6 distribution consistent with the pan‐epidermal activity of the CPR4 promoter. Electron microscopy revealed that on the first day of the fourth instar larvae, when spots initially appeared, local bulged structures form at the base of epidermal bristles, coinciding with the region of pigment deposition. These structures likely create a microenvironment that promotes pigment deposition. Consequently, the formation of twin‐spot patterns appears to depend not on the spatial distribution of Wnt6 expression but on the responsiveness of specific epidermal cells to Wnt6 signaling, influenced by factors such as the local microenvironment, co‐regulatory elements, and chromatin accessibility. Notably, previous studies show that ectopic Wnt1 expression induces pigmentation at any epidermal site, suggesting mechanistic differences between Wnt1 and Wnt6 in promoting pigment synthesis. Additionally, pCPR4>Wnt6 transgenic larvae did not exhibit twin‐spot patterns before the fourth instar, consistent with the temporal pattern of the mutant L. Prior research indicates that mutant L larvae lack distinct twin spots before the fourth instar, with Wnt1 expression significantly higher during the fourth‐instar molting stage than in earlier instars [8], suggesting that twin‐spot formation is closely tied to Wnt gene expression levels. Further analysis of Wnt6 expression in pCPR4>Wnt6 transgenic larvae across developmental stages showed significantly higher expression during the fourth‐instar molting stage compared to other stages, which indicated that elevated Wnt6 levels might be critical for triggering spot formation. From the fourth instar onward, epidermal bulges correlated positively with pigment deposition, which suggested that epidermal morphological changes might promote spot formation. In addition, it is reported that larvae of the citrus swallowtail (Papilio xuthus) display distinct body patterns and coloration at different developmental stages, which are coordinately regulated by ecdysteroids and JH [53]. In this study, we found that during the fifth‐instar molting stage, genes associated with the ecdysone and JH signaling pathways (such as EcR and JHBP) were significantly upregulated in the epidermal marking regions of pCPR4>Wnt6 larvae compared with non‐marked regions of the control larvae. Hemolymph analysis further confirmed that levels of JH and 20E were elevated in pCPR4>Wnt6 individuals relative to controls, suggesting that Wnt6 might influence pigment metabolism and marking formation by modulating hormone levels. Based on these observations, we speculate that the emergence of markings from the fourth instar onward may stem from elevated Wnt6 expression exerting a stronger influence on hormonal regulation and other downstream pathways (including epidermal remodeling) than in earlier instars, while the absence of an L mutant‐like phenotype prior to this stage likely reflects insufficient expression to reach the activation threshold.

The generation of black and brown melanin in insects is derived from DOPA, which is produced from tyrosine by the enzyme tyrosine hydroxylase (TH), or from dopamine, which is generated from DOPA through the action of Ddc [16, 54]. This metabolic pathway also involves several effector genes, including those that inhibit melanin synthesis (e.g., ebony and black), which convert dopamine into NBAD sclerotin (yellow/brown pigment), and those that promote melanin formation (e.g., yellow and Laccase2) [17, 18, 19, 20]. In multiple model organisms such as Drosophila, Tribolium castaneum, and B. mori, functional loss of these genes leads to significant changes in whole‐body melanin distribution [16, 17, 18, 19, 20]. These findings suggest that the region‐specific expression of effector genes plays a critical role in the evolution of insect body color patterns current studies widely support the view that regulatory genes (pattern genes) determine melanin distribution and localization by providing precise spatiotemporal expression cues to effector genes [3]. For instance, in Drosophila, the Dll (Distal‐less) and wg (wingless) genes upregulate the yellow gene by regulating the CRE element, thereby inducing wing pigment deposition [55, 56]. Abdominal pigmentation is controlled by the optomotor‐blind gene, while male‐specific pigment patterns are regulated by Abd‐B and bric‐à‐brac genes [57]. In lepidopteran insects, such as B. mori, Dll, wg, and WntA genes jointly regulate the distribution of wing melanin [58, 59, 60]. Additionally, in B. mori larvae, the formation of specific black stripes and spots is controlled by multiple factors, including the ligand Spz‐3 and its receptor Toll 8 [4], the imaginal disk growth factor Idgf [61], and the tsh/tio homologous gene [62]. These regulatory genes function in a species‐specific manner, activating effector genes under specific temporal and spatial conditions. Our studies found that specific overexpression of Wnt6 in the epidermis not only induced the twin‐spot marking phenotype but also significantly activated the expression of pigment synthesis‐related pathways and genes in this region. This suggests that Wnt6 may function as a pattern gene, influencing the distribution and localization of epidermal pigments in silkworm larvae, thus impacting the marking pattern. These findings provide a new perspective on the molecular mechanisms and evolutionary processes underlying pigment deposition patterns.

The Wnt/β‐catenin signaling pathway drives mammary tumorigenesis by upregulating cyclin D1 and c‐Myc expression [63]. Upon activation, β‐catenin collaborates with TCF/LEF transcription factors to promote proliferation and metastasis of ovarian cancer stem cells [64]. Additionally, elevated β‐catenin levels accelerate hepatocyte proliferation via the cyclin D1/c‐Myc pathway, thereby enhancing liver regeneration in mice [65]. Collectively, these studies highlight the critical role of Wnt signaling in regulating cell cycle progression and proliferation. Consistent with Fujiwara et al.'s report of bulge structures in mutant L larvae [8]. We observed localized bulge‐like structures with distinct positional characteristics in the epidermal marking regions of fourth‐instar silkworm larvae in both the control group and the pCPR4>Wnt6 transgenic line. Mechanistically, during the fifth‐instar molting stage, the p53 signaling pathway was suppressed in both mutant L and pCPR4>Wnt6 individuals. This suggests that Wnt6 may promote local cell proliferation by suppressing the p53 signaling pathway, thereby facilitating the formation of epidermal bulge structures. These findings further indicate that bulge structures are not artifacts of promoter activation but rather downstream developmental effects driven by enhanced Wnt6 activity.

In summary, by inducing ectopic Wnt6 expression in the silkworm epidermis, this study generated twin‐spot markings that closely resemble the mutant L phenotype, demonstrating that Wnt6 is sufficient to drive localized epidermal pigmentation. The Wnt6 loss‐of‐function evidence reported by Ding et al. [14], together with our gain‐of‐function findings, strongly supports an essential regulatory role for Wnt6 in twin‐spot formation in the mutant L. Building on these findings, the present work not only elucidates the mechanisms underlying Wnt6 function but also identifies candidate downstream pathways mediating pigment deposition, thereby broadening the molecular and genomic framework for understanding body‐color pattern development. Furthermore, the epidermis‐specific transgenic marker system established here provides a powerful tool for engineering visible genetic markers in silkworm and holds considerable promise for translational applications, including agricultural pest management and the breeding of functional silkworm varieties.

Materials and methods

Bombyx mori strains

The domesticated silkworm, B. mori, used in this experiment was from the multivoltine, non‐diapausing Nistari strain. The larvae were fed fresh mulberry leaves at 25 °C [62].

Plasmid construction

To construct the plasmid for CPR4 promoter‐driven EGFP gene expression, a 2‐kb upstream regulatory region of the CPR4 gene, including the putative promoter, was amplified from the B. mori genome using primers CPR4‐F and CPR4‐R. The fragment was then inserted into the pJET‐1.2 vector for sequencing validation. Subsequently, the sequence was amplified from the pJET‐1.2 vector using primers CPR4‐F1 and CPR4‐R1 and inserted into the pBac [EGFP‐IE1‐DsRed2] plasmid via homologous recombination [66]. This process finally generated the piggyBac transposon system‐based transgenic plasmid pBac [CPR4‐EGFP‐IE1‐DsRed2]. The control plasmid pBac [EGFP‐IE1‐DsRed2] lacked the CPR4 promoter.

Similarly, for the CPR38 promoter‐driven EGFP expression plasmid, a 2‐kb upstream sequence of the CPR38 gene was amplified using primers CPR38‐F and CPR38‐R. The fragment was then inserted into the pBac [EGFP‐IE1‐DsRed2] plasmid via homologous recombination using primers CPR38‐F2 and CPR38‐R2, generating the final construct pBac [CPR38‐EGFP‐IE1‐DsRed2].

To construct the Wnt6 expression plasmids driven by the CPR4/CPR38 promoters, total RNA was first extracted from the epidermis of B. mori and reverse‐transcribed into cDNA. ORF of the Wnt6 gene was then amplified using primers Wnt6‐F/R. Using homologous recombination technology, the amplified fragments were inserted into the specific recombination sites of the pBac [IE1‐EGFP‐CPR4] and pBac [IE1‐EGFP‐CPR38] vectors with the universal primer pair Wnt6‐F1/R1. The resulting transgenic plasmids, pBac [IE1‐EGFP‐CPR4‐Wnt6] and pBac [IE1‐EGFP‐CPR38‐Wnt6], expressed Wnt6 under the regulation of the CPR4 or CPR38 promoters, respectively. The plasmid pBac [EGFP‐Wnt6] was used as the control in the experiment. The primer sequences are provided in the Table S2. The sequences of the CPR4 and CPR38 promoters, along with the complete coding sequence of the Wnt6 gene ORF, are provided in the Sequences S1–S3.

Bombyx mori germline transformation

Haruhiko Fujiwara's team previously isolated epidermal cells from final instar silkworm (Bombyx mori) larvae during the molting period and constructed a cDNA library (epM), identifying 1380 predicted genes, of which 48% encode epidermal proteins, including RR‐1, RR‐2, glycine‐rich, and other subtypes [15]. These findings suggest that epidermal protein genes are promising candidates for identifying epidermis‐specific promoters, offering valuable tools to elucidate the molecular regulatory networks governing insect epidermal morphogenesis. In this study, we used data obtained by Haruhiko Fujiwara's team from the epM and epV3 cDNA libraries (the latter derived from the feeding stage) [15] and identified CPR4 and CPR38 as genes that are highly and specifically expressed in the epidermis through qRT‐PCR screening. Therefore, we selected CPR4 and CPR38 as candidate promoters for driving epidermis‐specific gene expression. To generate transgenic lines with EGFP expression driven by the CPR4 or CPR38 promoter, a mixture of pBac [CPR4‐EGFP‐IE1‐DsRed2] or pBac [CPR38‐EGFP‐IE1‐DsRed2] plasmids, piggyBac transposase mRNA and helper plasmids (each at a final concentration of 400 ng·μL⁻¹) was injected into G0 embryos at the preblastoderm stage.

The embryos were then incubated in a humidified incubator at 25 °C until hatching. The G0 larvae grew to the moth stage and underwent either sib mating or backcrossing with wild‐type moths to obtain the G1 generation. Positive individuals were screened based on the expression of marker genes under a fluorescence microscope (Nikon AZ100, Tokyo, Japan). The control group was processed under the same conditions.

To establish transgenic lines with specific Wnt6 expression in the epidermis, a similar procedure was followed, using pBac [IE1‐EGFP‐CPR4‐Wnt6] or pBac [IE1‐EGFP‐CPR38‐Wnt6] plasmids, piggyBac transposase mRNA, and helper plasmids, all at a final concentration of 400 ng·μL⁻¹. The G0 larvae were allowed to reach the moth stage, followed by sib mating or backcrossing with wild‐type moths. G1 transgenic individuals (designated pCPR4>Wnt6 and pCPR38>Wnt6 based on their promoter‐Wnt6 constructs) with green fluorescent markers were screened using a fluorescence microscope (Nikon AZ100). The control construct was used exclusively to rule out potential effects of the microinjection procedure on development.

RNA extraction, reverse transcription and quantitative real‐time PCR (qRT‐PCR) analysis

Total RNA was extracted from the various anatomical tissues of B. mori larvae, including the head, Malpighian tubules, midgut, middle silk gland, epidermis, fat body, testis, ovary, and wing disk, using TRIzol reagent (Invitrogen, Carlsbad, CA, USA). The RNA was then treated with DNase I (Invitrogen) to eliminate genomic DNA contamination, followed by purification with phenol/chloroform and dissolution in UltraPure distilled water (Invitrogen). RNA integrity was assessed by agarose gel electrophoresis. For cDNA synthesis, 1 μg of total RNA was reverse‐transcribed in a 20 μL reaction mixture using the Omniscript Reverse transcriptase kit (Qiagen, CA, USA) according to the manufacturer's protocol. The synthesized cDNA served as the template for qRT‐PCR, which was performed using a SYBR Green Real‐time PCR Master Mix Kit (Toyobo, Osaka, Japan). The qRT‐PCR conditions were as follows: initial denaturation at 95 °C for 3 min, followed by 40 cycles of 95 °C for 15 s, 60 °C for 20 s, and 72 °C for 45 s. The primers used for qRT‐PCR analysis of the target gene are listed in Table S2. B. mori ribosomal protein 49 (Rp49) was used as the internal reference [67].

RNA‐seq

The dorsal epidermis from the sixth segment in WT B. mori larvae and the fifth and sixth segments in mutant L larvae at the L5D3 stage, as well as the sixth‐segment dorsal epidermis from control and pCPR4>Wnt6 larvae at the L5D3 stage and during the L5 molting stage, were collected for RNA‐seq analysis. Total RNA was extracted from each epidermal sample using TRIzol reagent (Invitrogen) according to the manufacturer's instructions. RNA concentration and purity were measured using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). Only samples with an RNA Integrity Number (RIN) ≥ 8.0 were used for sequencing. Qualified RNA samples were subsequently submitted to Shanghai Majorbio Bio‐pharm Technology Co., Ltd. for library construction and sequencing. mRNA was enriched using Sera‐Mag oligo (dT) magnetic beads (Illumina, San Diego, CA, USA) and fragmented to approximately 300 base pairs. Library construction was conducted with the Illumina^® Stranded mRNA Prep, Ligation Kit, involving double‐stranded cDNA synthesis, end repair, adapter ligation, and library amplification, following the manufacturer's protocol. Sequencing adapters were ligated to the cDNA through PCR amplification, and the libraries were constructed using the NovaSeq Reagent Kit (Illumina). Sequencing was performed in a 2 × 150 bp paired‐end conffguration on an Illumina Novaseq 6000 platform, with all raw sequence read data stored in FastQ format. Raw paired‐end reads were trimmed and quality‐controlled using fastp [68] with default parameters. Clean reads were aligned in orientation mode to the B.mori reference genome (SilkBase, version 2.1; available at http://silkbase.ab.a.u‐tokyo.ac.jp/cgi‐bin/index.cgi) using HISAT2 software (http://ccb.jhu.edu/software/hisat2/index.shtml) [69]. The mapped reads of each sample were assembled by StringTie [70] in a reference‐based approach. Gene expression levels were quantified as FPKM (Fragments Per Kilobase of transcript per Million mapped reads) using StringTie. Differential expression analysis was performed using DEGseq [71], with DEGs defined as those with an adjusted P‐value <0.05 and |Log₂FC| ≥ 2. Functional enrichment analysis of KEGG pathways was performed to identify DEGs significantly enriched in KEGG pathways with a Benjamini‐Hochberg‐corrected P‐value <0.05, relative to the whole‐transcriptome background. KEGG pathway analysis was performed using Python scipy software. All bioinformatics analyses and visualizations were conducted on the Majorbio Cloud Platform (https://cloud.majorbio.com/) [72].

Whole genome sequencing

Genomic DNA was extracted from mutant L embryo samples by digesting with proteinase K, lysing in a standard sodium dodecyl sulfate (SDS)‐phenol buffer, and subsequently treating with RNase and purifying to obtain high‐quality DNA. The purified DNA was then submitted to Shanghai Majorbio Bio‐pharm Technology Co., Ltd. for genome sequencing. After passing library quality control, paired‐end sequencing (PE150) was performed on the Illumina NovaSeq 6000 platform, yielding reads with a total length of 300 bp. Raw sequencing data were subjected to quality control on the Majorbio Cloud Platform to remove low‐quality reads, resulting in clean data. SNPs and small insertions/deletions (InDels) were identified following the GATK Best Practices workflow. Structural variations (SVs) and copy number variations (CNVs) were analyzed using Delly and CNVnator, respectively. The putative promoter region of Wnt6 was defined as genomic coordinates 4 294 878–4 296 878 based on the silkworm reference genome (SilkBase database: http://silkbase.ab.a.u‐tokyo.ac.jp). Base mutation sites within this region were extracted from high‐throughput sequencing data, and the ggplot2 package in R was used to generate a mutation site distribution map annotated with genomic coordinates and the corresponding base substitutions.

Photography and microscopic imaging

Photographs of B. mori larvae and adults were acquired using a digital camera (Nikon DS‐Ri1, Tokyo, Japan). Bright‐field and fluorescence images of embryos, larvae, and dissected tissues were acquired using a stereo fluorescence microscope (Nikon, Tokyo, Japan).

Scanning electron microscopy (SEM) analysis

Epidermal samples from the control and pCPR4>Wnt6 transgenic of B. mori at L3D3 and L4D1 were fixed overnight in FAA fixative (37–40% formaldehyde, acetic anhydride, and 50% ethanol at a ratio of 1:1:18). The samples were then dehydrated through an ethanol gradient (50%, 70%, 80%, 90%, 95%, 100%), with each step lasting 10 min. The samples were dried in a critical point dryer with carbon dioxide for 6 h, subsequently coated with gold using a JFC‐1600 sputter coater (JEOL, Tokyo, Japan), and then examined by SEM with a JSM‐6360LV microscope (JEOL) to observe the surface structure of the epidermis.

Measurement of 20E titer

Hemolymph from three fifth‐instar molting stage silkworm larvae was pooled to obtain 300 μL of mixture per biological replicate (n = 3). 20E was extracted following previously described methods [73]. Briefly, the mixture was combined with methanol at a 1:1 (v/v) ratio, thoroughly vortexed, and centrifuged at 12 000  g for 10 min. The pellet was re‐extracted with ethanol, and the extracts were dried using a SpeedVac concentrator. Prior to enzyme immunoassay (EIA), the dried extracts were resuspended in EIA buffer and incubated overnight at 4 °C. 20E titers were measured using a 20E EIA Kit (Cayman Chemical, Ann Arbor, MI, USA) according to the manufacturer's instructions. Standard curves were generated using nonlinear regression analysis in the Origin software.

Measurement of JH titer

Six fifth‐instar molting stage silkworm larvae were selected, and a total of 600 μL of hemolymph from these larvae was pooled to constitute one biological replicate (n = 3). JH was extracted following previously described methods [74]. Each 600 μL hemolymph sample was mixed with 3000 μL of n‐hexane and 1500 μL of 70% methanol, and centrifuged at 12 000  g for 10 min. The upper n‐hexane phase was collected and dried under a stream of nitrogen gas. The dried extract was dissolved in 50 μL of 50% methanol, sonicated for 10 min, and centrifuged twice at 14 000  g for 10 min each. The resulting supernatant was collected for liquid chromatography (LC) analysis. LC‐mass spectrometry (LC–MS) analysis was performed by Tgene Biotech (Shanghai) Co., Ltd. Chromatography was carried out on a Vanquish UHPLC System (Thermo Fisher Scientific) using an ACQUITY UPLC^® HSS T3 column (2.1 × 100 mm, 1.8 μm; Waters, Milford, MA, USA) maintained at 40 °C. The flow rate and injection volume were set at 0.3 mL·min⁻¹ and 2 μL, respectively. Mass spectrometry was subsequently performed on a Thermo Q Exactive mass spectrometer (Thermo Fisher Scientific) equipped with an electrospray ionization (ESI) source, with data acquired separately in positive and negative ion modes.

Statistical analysis

All data were expressed as the mean ± the standard error of the mean. One‐way ANOVA with Tukey's multiple comparisons test, two‐way ANOVA with Tukey's multiple comparisons test, or Student's t‐test were used to evaluate statistical significance. P < 0.05 was considered statistically significant. All statistical analyses were conducted using graphpad prism 8.3.0 software (GraphPad Software).

Conflict of interest

All authors declare that they have no conflict of interest.

Author contributions

XX and ZY analyzed experimental data and drafted the manuscript. ZJ acquired and analyzed qPT‐PCR data and drafted the manuscript. YW and XG prepared transgenic silkworms and provided intellectual guidance on study design and data analysis. JZ acquired additional experimental data and revised the manuscript according to reviewer comments. FJ analyzed additional experimental data and revised the manuscript according to reviewer comments. RZ acquired experimental data, designed the study, and revised the manuscript.

Supporting information

Fig. S1. Spatial expression profiles of cuticular protein genes in silkworm larvae.

Table S1. Transgenic silkworm injection and strain construction statistics.

Table S2. Primers used in this study.

Sequences S1. CPR4 promoter.

Sequences S2. CPR38 promoter.

Sequences S3. Wnt6 CDS.

Data availability statement

The data supporting the findings of this study, along with the experimental materials and analytical methods, are available from the corresponding author upon reasonable request.