Summary

MicroRNAs (miRNAs) are endogenous small non-coding RNAs that play crucial roles in numerous biological processes. However, the role of miRNAs in skin color determination in fish has not been completely determined. Here, we identified that 13 miRNAs are differentially expressed between red and white skin. The analysis of miRNA spatial and temporal expression patterns suggests that miR-429 is a potential regulator of skin pigmentation. miR-429 silencing results in an obvious change in skin pigmentation. Bioinformatics analysis and a luciferase reporter assay show that miR-429 directly regulates expression of Foxd3 by targeting its 3′-untranslated (3′-UTR) region. miR-429 silencing leads to a substantial increase in the expression of Foxd3 in vivo, thereby repressing the transcription of MITF and its downstream genes, such as TYR, TYRP1 or TYRP2. These findings would provide a novel insight into the determination of skin color in fish.

Introduction

Some fish usually display fascinating color patterns in their skin, which have important roles in numerous biological processes, such as mate choice, camouflage and the perception of threatening behavior (Kelsh, 2004; Protas and Patel, 2008). The determination of skin color is a very complicated process in fish, which is associated with a series of cellular, genetic, environmental and physiological factors (Aspengren et al., 2009). How skin color patterns form, however, is a longstanding question among biologists. The genetics of skin pigmentation in vertebrates has been extensively studied in a range of laboratory animal models. To date, a series of genes have been reported to be involved in the determination of skin color, such as pro-opiomelanocortin (POMC), melanocyte-stimulating hormone (MSH), microphthalmia-associated transcription factor (MITF), kit oncogene (KIT), tyrosinase (TYR), tyrosine related protein-1 (TYRP1) and tyrosine related protein-2 (TYRP2) (Hubbard et al., 2010; Kelsh, 2004). Yet surely the majority of gene resources involved in the determination of fish skin color remain to be discovered.

Skin color pattern is governed by complex and well-balanced programs of gene activation and silencing. MicroRNAs (miRNAs) are a set of single-stranded, non-coding RNA molecules with an average size of approximately 22 nucleotides. They recognize and bind 3′-untranslated regions (UTRs) of mRNA, blocking translation of the gene or inducing cleavage of the mRNA. The crosstalk between miRNAs and mRNAs is important for the steadiness of signal transduction and the transcriptional activities as well as the maintenance of homeostasis in many organs, including the skin (Mo, 2012; Sand et al., 2009; Yi and Fuchs, 2010). Accordingly, skin-expressed miRNAs might have a crucial role in skin development, body color formation and skin diseases. In a previous study, miR-203 was found to define a molecular boundary between proliferative basal progenitors and terminally differentiating suprabasal cells, ensuring proper identity of neighboring layers (Yi et al., 2008). The expression level of miR-137 can affect the body color pattern in mice (Dong et al., 2012). In Drosophila, the loss of miR-8 shows a substantially decreased pigmentation of the dorsal abdomen (Kennell et al., 2012). This evidence suggests that miRNAs could be involved in the formation of body color. However, no miRNA has been reported in the process of the determination of fish skin color.

The common carp (Cyprinus carpio) is a widespread freshwater fish of eutrophic waters in lakes and large rivers in Europe and Asia. They have numerous skin colors in the natural environment, such as red, white, orange or black (Wang et al., 2009). Over the past ten years, we have built two pure strains of common carp, which have white or red skin color. Moreover, the trait of skin color is inherited in accordance with the law of Mendelian inheritance, as shown in supplementary material Fig. S1. Thus, they offer an appealing model system to investigate the genetic basis of skin pigment patterns. However, the unclear genetic background hinders its application for basic biological study. Tilapia is an important model for the research of fish physiology, endocrinology and the evolutionary mechanisms in vertebrates. It also has a diversity of color patterns, including red and white skin color. Scientists have successfully completed a tilapia whole-genome sequencing project, providing relatively clear genome information (Guyon et al., 2012). Common carp and tilapia would be adopted as research subjects to investigate the role of miRNAs in the determination of skin color.

In this study, we used the RNA-Seq method to identify skin-related miRNAs, and quantitive PCR to screen these differentially expressed miRNAs between white and red skin. We found that the miR-429 is a positive regulator of skin color in fish, and that loss of miR-429 affects the pigment content in the skin. These effects are mainly mediated by the direct interaction between miR-429 and Foxd3, suggesting a function of miRNA in the determination of skin color in fish.

Results

miRNA expression signature in the skin of common carp

To obtain the miRNA expression signature in the skin of common carp, we constructed a small RNA library from RNA sample pools isolated from different colored skin. Total RNA was size-fractionated through PAGE, and small RNAs corresponding in molecular weight to the mature miRNA population (18–24 nt fraction) were extracted from the gel and processed for reverse transcription and PCR amplification to create cDNA libraries. Small RNA sequencing using the Illumina Genome Analyzer II generated 15,505,485 raw reads. After removing low quality sequences and adapter sequences, 15,326,145 clean reads were left for further analysis. Of these, 60.06% of the small RNA sequences obtained were 20–24 nt in size, which is the typical size range for Dicer-derived products (Fig. 1). Currently, the complete genome sequence of common carp is still unavailable. We selected the zebrafish genome as the reference genome for subsequent analysis. The clean reads mentioned above were mapped to the zebrafish genome by using short oligonucleotide alignment program (SOAP) software. The results show that approximately 76,595 unique sRNAs (7,976,459 total sRNAs) were found to match the zebrafish genome perfectly (supplementary material Table S1). Next, we performed a database search to find rRNA, tRNA, snRNA and snoRNA deposited at Rfam and NCBI GenBank database (Fig. 1).

Fig. 1.

Overview of small RNA library sequencing. (A) The raw reads were primarily analyzed to obtain clean reads. (B) Length distribution of small RNAs in common carp. (C) The clean reads were blasted against the GenBank and Rfam databases to annotate rRNA, tRNA, snRNA or snoRNA.

Fig. 1.

Overview of small RNA library sequencing. (A) The raw reads were primarily analyzed to obtain clean reads. (B) Length distribution of small RNAs in common carp. (C) The clean reads were blasted against the GenBank and Rfam databases to annotate rRNA, tRNA, snRNA or snoRNA.

To identify the conserved miRNA homologs in the skin of common carp, BLASTn search with an E-value cutoff of 10 was used to search for the remnant small RNAs with predicted hairpin structures against the miRBase (http://microrna.sanger.ac.uk/sequences/) (Kozomara and Griffiths-Jones, 2011). With this similarity search, 73 miRNAs were found to be the same as at least one of the published miRNAs in the miRBase. All of these conserved miRNA precursors were identified and had a hairpin structure (data not shown). On the basis of sequence similarity, these known miRNAs were divided into 37 families (supplementary material Table S2).

miRNAs show differential expression in white and red skin

To gain insight into the function of miRNAs in physiological processes, it is essential to have precise information on their temporal and spatial expression patterns. First, we detected the expression levels of skin-related miRNAs in different skin color in common carp. The results indicate that 14 distinct miRNAs are differentially expressed between white and red skin. miR-25, miR-15a-3p, miR-146b, miR-184, miR-429 and miR-141 are abundantly expressed in red skin, whereas miR-18a, miR-137, miR-17a, miR-203a, miR-9-3p, miR-9-5p, miR-129-5p and miR-204 are abundantly expressed in white skin (Table 1). In tilapia, we also detected the expression levels of these differentially expressed miRNAs identified in common carp. The results show that excluding miR-141, miR-129-5p or miR-17a, other differentially expressed miRNAs have a similar expression pattern between common carp and tilapia (supplementary material Table S3).

Table 1.
Differential miRNA expression in white and red skin of common carp
graphic
graphic

The expression of miRNAs is tightly regulated in a time- and space-dependent manner. Tissue-specific or high-abundance miRNAs expressed in certain tissues imply that these miRNAs have crucial functions in the maintenance of tissue function. The observation that miR-25, miR-137, miR-203a and miR-429 are strongly expressed in the skin, implies an important role for these miRNAs in the skin constitutive process (Table 2). In addition, we examined the miRNA expression profile during five different developmental stages in common carp, including zygote, blastula, gastrula, segmentation and larvae. Interestingly, we found that miR-429 shows a dynamic expression pattern. Expression of miR-429 is first detected at the beginning of the gastrulae stage, and its expression is substantially upregulated in the segmentation stage; this expression is then sustained through to the larvae stage (Table 2; supplementary material Fig. S2). Given that pigment cells are initially derived from the neural crest during the gastrula stage (Betancur et al., 2010), we speculated that miR-429 is a potential regulator of the pigmentation process.

Table 2.
Overview of the expression pattern of these differential expressed miRNAs in different color skin
graphic
graphic

0: no expression;

1: weak ubiquitous or background expression;

2: strong ubiquitous expression.

miR-429 silencing affects the pigmentation process

In previous studies, we used an miRNA antagomir method to conduct miRNA loss-of-function experiments in vivo (Yan et al., 2012; Yan et al., 2013). In this study, we also observed that miR-429 expression could be efficiently blocked by its corresponding antagomir (miR-429 antagomir) but not the mismatched miRNA (miR-203 antagomir) in vivo. Furthermore, the silencing effect of the miRNA antagomir was detected at different time points (supplementary material Fig. S3). To evaluate the effect of miR-429 silencing on skin pigmentation, we injected red tilapia weighing ∼5 g with miR-429 antagomir for 30 days. Expression of miR-429 is efficiently silenced by the miR-429 antagomir but not by the mismatched antagomir (Fig. 2A). There is no significant difference observed between the wild-type fish and miR-429 silencing fish in survival rate (Data not shown). We compared the melanin production between wild-type and miR-429 silencing fish. We found that the red fish injected with miR-429 antagomir has a lower melanin content relative to its matched red group (Fig. 2B). In red common carp, we also observed that the treatment with the miR-429 antagomir significantly decreases the melanin content in skin (supplementary material Fig. S4). These data suggest that miR-429 expression levels could affect the process of skin pigmentation in fish.

Fig. 2.

miR-429 silencing affects the pigmentation process. (A) Red tilapia weighing ∼5 g was received miR-429 antagomir (or mismatched antagomir) at a dose of 60 mg/kg body weight or left untreated for 30 days. miR-429 expression was detected by using real-time PCR. 18S rRNA expression was detected as the internal control. The data were expressed as the relative change compared with the wild-type group. (B) Tilapia was treated as shown in Fig. 2A. Total melanin content in the skin was detected as described in Materials and Methods. Data represent the mean ±s.e.m. from three independent experiments. Asterisk (*) indicates a significant difference compared with the wild-type group (P<0.05).

Fig. 2.

miR-429 silencing affects the pigmentation process. (A) Red tilapia weighing ∼5 g was received miR-429 antagomir (or mismatched antagomir) at a dose of 60 mg/kg body weight or left untreated for 30 days. miR-429 expression was detected by using real-time PCR. 18S rRNA expression was detected as the internal control. The data were expressed as the relative change compared with the wild-type group. (B) Tilapia was treated as shown in Fig. 2A. Total melanin content in the skin was detected as described in Materials and Methods. Data represent the mean ±s.e.m. from three independent experiments. Asterisk (*) indicates a significant difference compared with the wild-type group (P<0.05).

miR-429 acts directly at the 3′-UTR of Foxd3

To determine the role of miR-429 in the pigmentation process, we performed an in silico functional annotation analysis of its predicted target genes using the TargetScan and miRanda program on the basis of the tilapia genome sequence (Chen et al., 2005). We screened the KEGG pathway database (http://www.kegg.jp/kegg/) (Kanehisa et al., 2008) for candidate genes of the melanogenesis signaling pathway, and found that Foxd3 is a potential target gene of miR-429. The sequence alignment between miR-429 and the 3′-UTR segment of Foxd3 is shown in Fig. 3A. Foxd3 has been shown to have a function in the specification of various downstream neural crest derivatives. To verify the interaction between Foxd3 and miR-429, we engineered two luciferase reporters, which are the wild-type 3′-UTR of Foxd3 gene, or the mutant UTR of Foxd3 gene. One luciferase reporter with or without miR-429 mimic or a scrambled miRNA mimic were transfected into an HEK 293T cell. The scrambled miRNA mimic with no homology to the tilapia genome was used to control the nonspecific effects. As shown in Fig. 3B, the transfection of scrambled miRNA mimic does not affect the luciferase activity of the Foxd3 3′-UTR wild-type reporter. However, miR-429 mimic transfection results in a significant decrease in the luciferase activity of Foxd3 3′-UTR wildytpe reporter. By contrast, the luciferase activity of Foxd3 3′-UTR mutant reporters is not repressed by the miR-429 mimic (Fig. 3B). In addition, we found that in vivo administration of miR-429 antagomir but not NaCl results in a profound decrease in the endogenous expression of miR-429 (Fig. 3C). Meanwhile, we detected a significant increase in Foxd3 expression at mRNA level and protein level (Fig. 3D,E). The inverse expression correlation between miRNAs and putative target genes also suggests that miR-429 directly regulates Foxd3 expression in vivo. Taken together, these data indicate that miR-429 regulates Foxd3 expression through targeting of the 3′-UTR of Foxd3 gene.

Fig. 3.

miR-429 acts directly at the 3′-UTR of Foxd3. (A) The alignment between miR-429 and the 3′-UTR segment of Foxd3. (B) Luciferase reporters were linked with Foxd3 3′-UTRs containing either putative miR-429-binding sites (3′-UTR wt) or mutant miR-429 binding sites (3′-UTR mutant). miR-429 mimic or scrambled mimic (Scr mimic) were cotransfected with a luciferase-UTR construct into HEK 293T cells, and then the luciferase activity was determined. The cells transfected with Scr mimic plus 3′-UTR wt were used as the control group. Data represent the mean ±s.e.m. from three independent experiments. P<0.05. (C and D) Tilapia weighing ∼5 g received a tail-vein injection of NaCl, or miR-429 antagomir at a dose of 60 mg/kg body weight for the indicated times. The untreated group was taken as the control group. The relative expression of miR-429 (C) and Foxd3 (D) was detected using real-time PCR. 18S rRNA expression was used as the internal control. The data were expressed as the relative change compared with the untreated group. Asterisk (*) indicates a significant difference compared with the control group (P<0.05). (E) Tilapia was treated as shown in Fig. 3D, and the expression of Foxd3 was detected by using western blots. GAPDH was detected as the internal control. Shown in representative image.

Fig. 3.

miR-429 acts directly at the 3′-UTR of Foxd3. (A) The alignment between miR-429 and the 3′-UTR segment of Foxd3. (B) Luciferase reporters were linked with Foxd3 3′-UTRs containing either putative miR-429-binding sites (3′-UTR wt) or mutant miR-429 binding sites (3′-UTR mutant). miR-429 mimic or scrambled mimic (Scr mimic) were cotransfected with a luciferase-UTR construct into HEK 293T cells, and then the luciferase activity was determined. The cells transfected with Scr mimic plus 3′-UTR wt were used as the control group. Data represent the mean ±s.e.m. from three independent experiments. P<0.05. (C and D) Tilapia weighing ∼5 g received a tail-vein injection of NaCl, or miR-429 antagomir at a dose of 60 mg/kg body weight for the indicated times. The untreated group was taken as the control group. The relative expression of miR-429 (C) and Foxd3 (D) was detected using real-time PCR. 18S rRNA expression was used as the internal control. The data were expressed as the relative change compared with the untreated group. Asterisk (*) indicates a significant difference compared with the control group (P<0.05). (E) Tilapia was treated as shown in Fig. 3D, and the expression of Foxd3 was detected by using western blots. GAPDH was detected as the internal control. Shown in representative image.

Altered expression of pigmentation genes in miR-429 silencing tilapia

Skin color is primarily determined by the amount, type and distribution of melanin (Parra, 2007). MITF has been reported as a key gene in the regulation of melanin production. It is involved in melanocyte development and melanogenesis. Foxd3 can bind the MITF promoter and repress MITF transcription, thereby affecting the process of melanogenesis. Because miR-429 silencing results in the upregulation of Foxd3 expression, MITF would be upregulated during this process. Indeed, quantitive PCR analysis reveals that miR-429 silencing results in a significant increase in Foxd3 expression, thereby upregulating MITF expression (Fig. 4A,B). MITF is required for the expression of TYR, TRP-1 and TRP-2 genes that encode enzymes implicated in the production of melanin (Levy et al., 2006). Quantitive PCR and western blot analysis indicate that the expression of MITF downstream genes, TYR, TYRP1 and TYRP2, are markedly downregulated in miR-429 silencing fish (Fig. 4C,D). Taken together, these results show that miR-429 silencing can increase the level of Foxd3, which represses the transcription of MITF and its downstream genes.

Fig. 4.

miR-429 silencing changes the expression of MITF and its downstream genes. Tilapia weighing ∼5 g received miR-429 antagomir at a dose of 60 mg/kg body weight or were left untreated for the indicated times. The untreated group was taken as the control group. The relative mRNA expression of Foxd3 (A), MITF (B), and MITF downstream genes, including TYR, TRP-1, and TRP-2 (C) was detected using real-time PCR. 18S rRNA expression was detected as the internal control. The data were expressed as the relative change compared with the untreated group. Asterisk (*) indicates a significant difference compared with the control group (P<0.05). (D) The expression of TYR, TYRP1 and TYRP2 proteins was detected by western blot analysis. GAPDH was used as the internal control. A representative result of three repeated experiments is shown.

Fig. 4.

miR-429 silencing changes the expression of MITF and its downstream genes. Tilapia weighing ∼5 g received miR-429 antagomir at a dose of 60 mg/kg body weight or were left untreated for the indicated times. The untreated group was taken as the control group. The relative mRNA expression of Foxd3 (A), MITF (B), and MITF downstream genes, including TYR, TRP-1, and TRP-2 (C) was detected using real-time PCR. 18S rRNA expression was detected as the internal control. The data were expressed as the relative change compared with the untreated group. Asterisk (*) indicates a significant difference compared with the control group (P<0.05). (D) The expression of TYR, TYRP1 and TYRP2 proteins was detected by western blot analysis. GAPDH was used as the internal control. A representative result of three repeated experiments is shown.

miR-429-mediated pigmentation is distinct from α-MSH-mediated pigmentation upon UV radiation

Ultraviolet (UV) radiation is a major environmental hazard that can lead to skin inflammation, photoaging and skin cancer. The skin pigment can provide essential protection against UV radiation (Matsumura and Ananthaswamy, 2004). Here, we found that UV radiation significantly upregulates the level of miR-429 expression in the melanocyte. The change in miR-429 expression could be detected as early as 1 hour after UV treatment (Fig. 5A). UV treatment also could significantly increase MC1R and α-MSH expression at the mRNA and protein level, which stimulates the keratinocytes to secrete to enhance the synthesis of skin melanin (Fig. 5A). By comparison, we found that the change in UV-mediated miR-429 expression occurs substantially earlier than that of UV-mediated gene and protein expression, suggesting that miRNA regulation occurs earlier than most gene transcription responses (Fig. 5A). In addition, we found that miR-429 expression was significantly increased in fish upon UV radiation. Meanwhile, we detected a significant reduction in Foxd3 expression and a marked increase in MITF expression at mRNA level and protein level (Fig. 5B–D). To determine whether miR-429 is involved in MSH-mediated pigmentation, we compared the expression profiling of α-MSH and MC1R between the miR-429 silencing group and the wild-type group in response to UV radiation. The results show that the miR-429 level does not change the expression of α-MSH and the level of MC1R under the same conditions, suggesting that miR-429-mediated pigmentation is distinct from α-MSH-mediated pigmentation (Fig. 5E,F,G).

Fig. 5.

miR-429-mediated pigmentation is distinct from α-MSH-mediated pigmentation upon exposure to UV radiation. (A) The melanocyte was exposed to UV radiation or left untreated for the indicated times. The expressions of miR-429, MC1R and α-MSH levels were detected by real-time PCR or western blot analysis. The results were expressed as the change in treated groups relative to the untreated group. (BD) Tilapia was exposed to UV radiation for the indicated time. The untreated group was taken as the control group. The relative expressions of MC1R and α-MSH levels were detected by real-time PCR or western blot analysis. 18S rRNA expression (for real-time PCR) or GAPDH (for western blot) was detected as the internal control. The data were expressed as the relative change compared with the untreated group. (E, F and G) The melanocytes were transfected with miR-429 inhibitor or left untreated, and then they were exposed to UV radiation for the indicated times. The relative expressions of MC1R and α-MSH levels were detected by real-time PCR and western blot analysis. Each sample was analyzed in triplicate. Asterisk (*) indicates a significant difference compared with the control group (P<0.05).

Fig. 5.

miR-429-mediated pigmentation is distinct from α-MSH-mediated pigmentation upon exposure to UV radiation. (A) The melanocyte was exposed to UV radiation or left untreated for the indicated times. The expressions of miR-429, MC1R and α-MSH levels were detected by real-time PCR or western blot analysis. The results were expressed as the change in treated groups relative to the untreated group. (BD) Tilapia was exposed to UV radiation for the indicated time. The untreated group was taken as the control group. The relative expressions of MC1R and α-MSH levels were detected by real-time PCR or western blot analysis. 18S rRNA expression (for real-time PCR) or GAPDH (for western blot) was detected as the internal control. The data were expressed as the relative change compared with the untreated group. (E, F and G) The melanocytes were transfected with miR-429 inhibitor or left untreated, and then they were exposed to UV radiation for the indicated times. The relative expressions of MC1R and α-MSH levels were detected by real-time PCR and western blot analysis. Each sample was analyzed in triplicate. Asterisk (*) indicates a significant difference compared with the control group (P<0.05).

Discussion

miRNAs post-transcriptionally regulate gene expression by promoting mRNA degradation or inhibiting mRNA translation. They have pivotal roles in a variety of developmental processes, and their dysregulations are linked to numerous skin diseases (Mo, 2012). When miRNAs are globally ablated in skin epithelium by conditionally targeting Dicer 1, an miRNA-processing enzyme, this operation could distort epidermal morphology, implying a key role of miRNAs in skin development (Yi et al., 2006). To gain insight into the possible significance of skin miRNAs in fish, we first identified the expression profile of miRNAs in the skin tissue of fish, and then compared the miRNA expression pattern between white and red skin. We found that 11 miRNAs are differentially expressed in different colored skin, implying that these differentially expressed miRNAs are involved in skin pigmentation. The knowledge of tissue-specific and cell-specific expression patterns of miRNAs can directly inform functional studies (Aboobaker et al., 2005). Previous studies have revealed that miR-375 is specifically expressed in pancreatic islet cells, where it regulates the expression of insulin secretion (Poy et al., 2004). miR-1 is expressed exclusively in muscle, where it regulates cardiomyocyte proliferation in vertebrates and muscle physiology in flies (Sokol and Ambros, 2005; Zhao et al., 2005). Here, we found that miR-429, miR-25 and miR-137 are highly expressed in fish skin. In particular, miR-429 displays dynamic expression patterns during embryonic development. miR-429 is first detected at the gastrulae stage, and then its expression is gradually upregulated until the larvae stage. Because pigment cells are initially derived from the neural crest during gastrula stage, we speculated that miR-429 has a key function in the regulation of skin pigmentation.

Skin pigmentation in fish is a complex process that involves a series of cellular, genetic and physiological factors (Colihueque, 2010). The role of miRNAs in pigmentation has been reported in some species, including mouse, alpaca and Drosophila (Dong et al., 2012; Kennell et al., 2012; Zhu et al., 2010). In this study, we revealed that miR-429 is a potential regulator of fish pigmentation. Our previous study found that miR-429 directly regulates the expression of OSTF1, an osmotic stress transcriptional factor, revealing a role of miR-429 in fish osmoregulation (Yan et al., 2012). Here, we found that miR-429 is highly expressed in red skin, and inhibition of miR-429 function causes a substantial decrease in skin pigmentation. This study further extends the biological role of miR-429 in fish. miR-429 is a member of the miR-8 family, which has been predicted or experimentally confirmed in a wide range of species. Previous studies have identified miR-8 as a regulator of osmoregulation, growth, apoptosis and neuronal survival by targeting multiple mRNAs (Hyun et al., 2009; Loya et al., 2009; Vallejo et al., 2011). In Drosophila, miR-8 is required for proper spatial patterning of pigment on adult female abdomens. Loss of miR-8 in the developing cuticle results in cell-autonomous loss of pigmentation (Kennell et al., 2012). Sequence alignment suggests that the miR-8 family is highly conserved between invertebrates and vertebrates, which might indicate that its function has been conserved. Our study provides an example that suggests the role of miR-8 in pigmentation is highly conserved between invertebrates and vertebrates.

miRNAs control biological processes by regulating the expression of their target genes. Here, we found a binding site of miR-429 in the 3′-UTR region of Foxd3, and characterized their effects on Foxd3 using a reporter assay. Foxd3 is one of the earliest molecular markers of the neural crest lineage; it is expressed in many organisms in the premigratory and migrating neural crest, and its expression is downregulated as the cells differentiate into most derivatives (Abel and Aplin, 2010; Thomas and Erickson, 2009). Foxd3 can also control the lineage choice between neural or glial and pigment cells by repressing MITF during the early phase of neural crest migration. In this study, we found that miR-429 expression begins in the gastrula stage, which leads to a gradual decrease in expression of Foxd3. The ablation of MITF repression could contribute to the generation of pigment cells.

MITF is a member of the Myc-related family of basic helix-loop-helix leucine zipper (bHLH-Zip) transcription factors and is highly conserved across different vertebrate species. MITF-positive cells are also observed in the optic neuroepithelial layer, the presumptive retinal pigment epithelium (RPE), as well as in cells behind the optic cup that were probably derived from the neural crest and could develop into choroidal and iris pigment cells (Nakayama et al., 1998). Mutations of the MITF gene cause a variety of phenotypes, most notably in pigmented cells (Levy et al., 2010). This evidence suggests that MITF expression is tightly associated with the development of pigment cells. Expression of MITF begins in neural crest-derived cells in mouse embryo (Hou et al., 2000). Here, we found a similar result, which shows that miR-429 is activated at the gastrula stage, which directly represses Foxd3 expression, thereby releasing MITF inhibition by Foxd3. MITF directly regulates the expression of multiple genes that are necessary for melanophore development, including tyrosinase, tyrosinase-related protein-1, tyrosinase-related protein-2 and so on (Levy et al., 2006). We also observe the expression change in the MITF downstream gene when the level of miR-429 expression is altered.

Ultraviolet radiation is a major environmental hazard and brings harmful effect on skin tissue. Protection against UV-mediated damage is afforded by melanin production in melanocyte. Melanin absorbs UV radiation and dissipates the energy as harmless heat, blocking the UV from damaging skin tissue. A key component of this process is the UV-induced α-MSH, pro-opiomelanocortin (POMC) and MC1R gene expression (Miller and Tsao, 2010). α-MSH is the main physiological regulator of skin pigmentation, which is produced by the proteolytic cleavage of the large precursor protein, POMC. α-MSH-bound MC1R can activate adenylyl cyclase, lead to phosphorylation of cAMP responsive-element-binding protein (CREB) transcription factor family members and in turn transcriptionally activate various genes, such as MITF. Our results show that UV treatment in fish leads to the activation of α-MSH and the MC1R gene. The result is similar to that previously reported in keratinocytes (Dong et al., 2012). By contrast, miR-429 knockdown has no effect on the level of α-MSH and MC1R expression with or without UV radiation. On the basis of these results, we speculate that miR-429-mediated melanin production is distinct from α-MSH-mediated melanin production. We proposed a model for the avoidance of UV injury in fish (Fig. 6). Once a fish is exposed to UV radiation, miR-429 expression is rapidly upregulated, and earlier than other pigmentation gene activation. miR-429 upregulation could release Foxd3-mediated MITF inhibition, thus activating its downstream pathway and pigment synthesis (Segura et al., 2009). If fish are exposed to sustained UV injury, the pituitary and keratinocytes secrete α-MSH and induce the production of MC1R in melanocytes, which activates the α-MSH pathway. miRNA-mediated gene regulation operates earlier than most transcriptional responses. The fast regulation of miRNAs after UV treatment indicates that miRNA-mediated gene silencing acts earlier than most gene transcriptional responses after UV damage. Fish have evolved numerous strategies for effectively escaping UV injury through distinct signaling pathways. miRNAs are implicated in buffering developmental processes against the effects of environmental fluctuations.

Fig. 6.

The biological effects of miR-429 in the determination of skin color.

Fig. 6.

The biological effects of miR-429 in the determination of skin color.

In summary, we revealed a new regulatory mechanism for skin pigmentation in fish from an miRNA viewpoint. We found that miR-429 is differentially expressed between white and red skin, and its silencing using antagomir leads to a substantial change in skin melanin content. The post-transcriptional regulation of Foxd3 by miR-429 could affect the expression of MITF and its downstream genes, including TYR, TYRP1 and TYRP2, which in turn affects the pigmentation process in fish skin. miRNA-mediated melanogenesis might provide an additional option for fish to avoid UV injury.

Materials and Methods

Experimental fish

Tilapia and common carp were obtained from the fishery farm of Shanghai Ocean University (Shanghai, China). They were kept in a water circulation system in 200-liter tanks, and the water temperature was kept at 26±2°C under a 12-hour light and 12-hour dark photoperiod. All experiments were conducted under the Guidance of the Care and Use of Laboratory Animals in China. This research was approved by the Committee on the Ethics of Animal Experiments of Shanghai Ocean University.

Small RNA library construction and sequencing

RNA samples were harvested from different color skin of common carp, and immediately frozen in liquid nitrogen. Small RNA libraries were constructed using a Small RNA Cloning Kit (Takara). ∼20 µg of small RNA was submitted for sequencing. Briefly, the Solexa sequencing was performed as follows: RNA was purified by PAGE to enrich for the molecules in the range of 17–27 nt, then was ligated with 5′- and 3′- adaptors. The resulting samples were used as templates for cDNA synthesis followed by PCR amplification. The obtained sequencing libraries were subjected to the Solexa sequencing-by-synthesis method. After the run, image analysis, sequencing quality evaluation and data production summarization were performed by using the Illumina/Solexa pipeline.

Identification of skin-related miRNAs

The sequencing data was pretreated to discard low quality reads, no 3′-adaptor reads, 5′-adaptor contaminants and sequences shorter than 18 nucleotides. After trimming the 3′-adaptor sequence, sequence tags were mapped onto the zebrafish genome using the SOAP software (http://soap.genomics.org.cn) with a tolerance of one mismatch. The matched sequences were then queried against non-coding RNAs from the Rfam database (http://www.sanger.ac.uk/Software/Rfam) and NCBI GenBank database (http://www.ncbi.nlm.nih.gov/blast/Blast.cgi) by performing a BLASTn search. Any small RNAs with exact matches to these sequences were excluded from further analysis. The remnant reads were compared with the miRBase (19.0) to annotate conserved miRNAs. The secondary structures of the predicted miRNAs were confirmed using RNAfold (http://rna.tbi.univie.ac.at/cgi-bin/RNAfold.cgi).

Real-time RT-PCR analyses of mRNAs and miRNAs

Total RNA was extracted using Trizol reagent (Invitrogen), and miRNAs were extracted using the miRNeasy kit (Qiagen) according to the manufacturer's instruction. RNA integrity was assessed by electrophoresis on a 1.0% agarose gel. For mRNA quantification, reverse transcription was performed using a High Fidelity primeScriptTM RT-PCR Kit as instructed (Takara, Dalian, China). Real-time RT-PCR was performed by using the SYBR Green Real-time PCR Master Mix (Toyobo, Shanghai, China) and the StepOne Real-time PCR system (Applied Biosystems Inc., Foster City, CA) according to the manufacturer's protocol. miRNA abundance was detected using the stem-loop PCR method, and 18S rRNA expression was detected as the internal control. The relative gene or miRNA expression was detected using the comparative threshold cycle (CT) method, also referred to as the 2- ΔΔCT method. All reactions were performed in triplicate on the MyiQ5 Real-time PCR Detection System (Bio-Rad, Hercules, CA, USA).

Luciferase reporter assay

The 3′-UTR of Foxd3 was amplified from tilapia genomic DNA and individually cloned into the pGL3 vector (Promega) by directional cloning. The mutant Foxd3 3′-UTR reporters were created by mutating the seed region of the predicted miR-429 site. HEK293T cells were co-transfected with 0.4 µg of firefly luciferase reporter vector and 0.02 µg of the control vector containing Renilla luciferase (PRL-CMV, Promega) using lipofectamine 2000 (Invitrogen) in 24-well plates (Costar). Each transfection was performed in four wells. Luciferase assays were carried out 24 hours after transfection using the Dual Luciferase Reporter Assay System (Promega). Firefly luciferase activity was normalized to Renilla luciferase activity.

Spectrophotometric assay of melanin

Skin sample suspensions were solubilized in 8 M urea/1 M sodium hydroxide and cleared by centrifugation at 10,700 g for 10 minutes. Chloroform was added to the supernatants to remove fatty impurities. Skin containing pheomelanins were cleared by centrifugation at 10,700 g for 10 minutes and analyzed for absorbance at 400 nm. A400/mg = spectrophotometric pheomelanin (Sp.PM) (Dong et al., 2012).

Western blot analysis

The samples were lysed in the lysis buffer (20 mM Tris, pH 7.4, 150 mM NaCl, 5 mM EDTA, 1% Triton X-100) with 1 mM phenylmethylsulfonyl fluoride, 10 mM NaF, 1 mM Na3VO4 and protease inhibitor cocktail. The protein lysates were cleared by centrifugation at 4°C for 15 minutes at 12,000 g, quantified using the Bio-Rad protein assay, resolved on 8–12% SDS-PAGE gels, and transferred onto PVDF membranes (Millipore). After blocking, the PVDF membranes were incubated with the primary antibody, and then the membrane was washed and incubated with goat anti-mouse IgG (H+L)-HRP conjugate antibody (Santa Cruz). Specific complexes were visualized by echo-chemiluminescence (Amersham).

Data analysis

Data was expressed as mean ±s.e.m. unless otherwise stated. Statistical significance was assessed by one-way ANOVA followed by Bonferroni's multiple comparison tests. Statistical significance was defined as P<0.05.

Acknowledgements

We are grateful to Xian-qiang Du and Ling Wang in Shanghai Oebiotech CO., LTD for advice on data analysis. The authors declare that there is no conflict of interest that would affect the impartiality of this scientific work.

Author contributions

B.Y. and C.H.-W. designed the experiment and wrote the manuscript. B.Y., B.L. and C.D.-Z. carried out the experiments, data organization and statistical analyses. K.L.-L. and L.J.-Y. provided the experimental samples. J.L.-Z. and X.L.-G. participated in the study design and discussed the manuscript. All authors read and approved the final manuscript.

Funding

This work was supported by Special Fund for Agro-scientific Research in the Public Interest [grant number 200903045 to C.H.W.]; the National Natural Science Foundation of China [grant number 30972250 to C.H.W.]; the Specialized Research Fund for the Doctoral Program of Higher Education of China [grant number 20123104120005 to B.Y.]; the Shanghai Educational Development Foundation [grant number 12CG56 to B.Y.]; and the Shanghai University Knowledge Service Platform-Shanghai Ocean University aquatic animal breeding center [grant number ZF1206 to C.H.W.].

Reference

Abel
E. V.
,
Aplin
A. E.
(
2010
).
FOXD3 is a mutant B-RAF-regulated inhibitor of G(1)-S progression in melanoma cells.
Cancer Res.
70
,
2891
2900
.
Aboobaker
A. A.
,
Tomancak
P.
,
Patel
N.
,
Rubin
G. M.
,
Lai
E. C.
(
2005
).
Drosophila microRNAs exhibit diverse spatial expression patterns during embryonic development.
Proc. Natl. Acad. Sci. USA
102
,
18017
18022
.
Aspengren
S.
,
Sköld
H. N.
,
Wallin
M.
(
2009
).
Different strategies for color change.
Cell. Mol. Life Sci.
66
,
187
191
.
Betancur
P.
,
Bronner-Fraser
M.
,
Sauka-Spengler
T.
(
2010
).
Assembling neural crest regulatory circuits into a gene regulatory network.
Annu. Rev. Cell Dev. Biol.
26
,
581
603
.
Chen
P. Y.
,
Manninga
H.
,
Slanchev
K.
,
Chien
M.
,
Russo
J. J.
,
Ju
J.
,
Sheridan
R.
,
John
B.
,
Marks
D. S.
,
Gaidatzis
D.
 et al. (
2005
).
The developmental miRNA profiles of zebrafish as determined by small RNA cloning.
Genes Dev.
19
,
1288
1293
.
Colihueque
N.
(
2010
).
Genetics of salmonid skin pigmentation: clues and prospects for improving the external appearance of farmed salmonids.
Rev. Fish Biol. Fish.
20
,
71
86
.
Dong
C.
,
Wang
H.
,
Xue
L.
,
Dong
Y.
,
Yang
L.
,
Fan
R.
,
Yu
X.
,
Tian
X.
,
Ma
S.
,
Smith
G. W.
(
2012
).
Coat color determination by miR-137 mediated down-regulation of microphthalmia-associated transcription factor in a mouse model.
RNA
18
,
1679
1686
.
Guyon
R.
,
Rakotomanga
M.
,
Azzouzi
N.
,
Coutanceau
J. P.
,
Bonillo
C.
,
D'Cotta
H.
,
Pepey
E.
,
Soler
L.
,
Rodier-Goud
M.
,
D'Hont
A.
 et al. (
2012
).
A high-resolution map of the Nile tilapia genome: a resource for studying cichlids and other percomorphs.
BMC Genomics
13
,
222
.
Hou
L.
,
Panthier
J. J.
,
Arnheiter
H.
(
2000
).
Signaling and transcriptional regulation in the neural crest-derived melanocyte lineage: interactions between KIT and MITF.
Development
127
,
5379
5389
.
Hubbard
J. K.
,
Uy
J. A. C.
,
Hauber
M. E.
,
Hoekstra
H. E.
,
Safran
R. J.
(
2010
).
Vertebrate pigmentation: from underlying genes to adaptive function.
Trends Genet.
26
,
231
239
.
Hyun
S.
,
Lee
J. H.
,
Jin
H.
,
Nam
J. W.
,
Namkoong
B.
,
Lee
G.
,
Chung
J.
,
Kim
V. N.
(
2009
).
Conserved MicroRNA miR-8/miR-200 and its target USH/FOG2 control growth by regulating PI3K.
Cell
139
,
1096
1108
.
Kanehisa
M.
,
Araki
M.
,
Goto
S.
,
Hattori
M.
,
Hirakawa
M.
,
Itoh
M.
,
Katayama
T.
,
Kawashima
S.
,
Okuda
S.
,
Tokimatsu
T.
 et al. (
2008
).
KEGG for linking genomes to life and the environment.
Nucleic Acids Res.
36 Database issue
,
D480
D484
.
Kelsh
R. N.
(
2004
).
Genetics and evolution of pigment patterns in fish.
Pigment Cell Res.
17
,
326
336
.
Kennell
J. A.
,
Cadigan
K. M.
,
Shakhmantsir
I.
,
Waldron
E. J.
(
2012
).
The microRNA miR-8 is a positive regulator of pigmentation and eclosion in Drosophila.
Dev. Dyn.
241
,
161
168
.
Kozomara
A.
,
Griffiths-Jones
S.
(
2011
).
miRBase: integrating microRNA annotation and deep-sequencing data.
Nucleic Acids Res.
39
,
D152
D157
.
Levy
C.
,
Khaled
M.
,
Fisher
D. E.
(
2006
).
MITF: master regulator of melanocyte development and melanoma oncogene.
Trends Mol. Med.
12
,
406
414
.
Levy
C.
,
Khaled
M.
,
Robinson
K. C.
,
Veguilla
R. A.
,
Chen
P. H.
,
Yokoyama
S.
,
Makino
E.
,
Lu
J.
,
Larue
L.
,
Beermann
F.
 et al. (
2010
).
Lineage-specific transcriptional regulation of DICER by MITF in melanocytes.
Cell
141
,
994
1005
.
Loya
C. M.
,
Lu
C. S.
,
Van Vactor
D.
,
Fulga
T. A.
(
2009
).
Transgenic microRNA inhibition with spatiotemporal specificity in intact organisms.
Nat. Methods
6
,
897
903
.
Matsumura
Y.
,
Ananthaswamy
H. N.
(
2004
).
Toxic effects of ultraviolet radiation on the skin.
Toxicol. Appl. Pharmacol.
195
,
298
308
.
Miller
A. J.
,
Tsao
H.
(
2010
).
New insights into pigmentary pathways and skin cancer.
Br. J. Dermatol.
162
,
22
28
.
Mo
Y. Y.
(
2012
).
MicroRNA regulatory networks and human disease.
Cell. Mol. Life Sci.
69
,
3529
3531
.
Nakayama
A.
,
Nguyen
M. T. T.
,
Chen
C. C.
,
Opdecamp
K.
,
Hodgkinson
C. A.
,
Arnheiter
H.
(
1998
).
Mutations in microphthalmia, the mouse homolog of the human deafness gene MITF, affect neuroepithelial and neural crest-derived melanocytes differently.
Mech. Dev.
70
,
155
166
.
Parra
E. J.
(
2007
).
Human pigmentation variation: evolution, genetic basis, and implications for public health.
Am. J. Phys. Anthropol.
134
,
Suppl 45
85
105
.
Poy
M. N.
,
Eliasson
L.
,
Krutzfeldt
J.
,
Kuwajima
S.
,
Ma
X.
,
Macdonald
P. E.
,
Pfeffer
S.
,
Tuschl
T.
,
Rajewsky
N.
,
Rorsman
P.
 et al. (
2004
).
A pancreatic islet-specific microRNA regulates insulin secretion.
Nature
432
,
226
230
.
Protas
M. E.
,
Patel
N. H.
(
2008
).
Evolution of coloration patterns.
Annu. Rev. Cell Dev. Biol.
24
,
425
446
.
Sand
M.
,
Gambichler
T.
,
Sand
D.
,
Skrygan
M.
,
Altmeyer
P.
,
Bechara
F. G.
(
2009
).
MicroRNAs and the skin: tiny players in the body's largest organ.
J. Dermatol. Sci.
53
,
169
175
.
Segura
M. F.
,
Hanniford
D.
,
Menendez
S.
,
Reavie
L.
,
Zou
X.
,
Alvarez-Diaz
S.
,
Zakrzewski
J.
,
Blochin
E.
,
Rose
A.
,
Bogunovic
D.
 et al. (
2009
).
Aberrant miR-182 expression promotes melanoma metastasis by repressing FOXO3 and microphthalmia-associated transcription factor.
Proc. Natl. Acad. Sci. USA
106
,
1814
1819
.
Sokol
N. S.
,
Ambros
V.
(
2005
).
Mesodermally expressed Drosophila microRNA-1 is regulated by Twist and is required in muscles during larval growth.
Genes Dev.
19
,
2343
2354
.
Thomas
A. J.
,
Erickson
C. A.
(
2009
).
FOXD3 regulates the lineage switch between neural crest-derived glial cells and pigment cells by repressing MITF through a non-canonical mechanism.
Development
136
,
1849
1858
.
Vallejo
D. M.
,
Caparros
E.
,
Dominguez
M.
(
2011
).
Targeting Notch signalling by the conserved miR-8/200 microRNA family in development and cancer cells.
EMBO J.
30
,
756
769
.
Wang
C. H.
,
Wang
J.
,
Xiang
S. P.
,
Lv
Y. P.
,
He
X. Z.
(
2009
).
Parental genetic effects evaluation of growth-related traits of red common carp in China.
Fish. Sci.
75
,
1301
1305
.
Yan
B.
,
Zhao
L. H.
,
Guo
J. T.
,
Zhao
J. L.
(
2012
).
miR-429 regulation of osmotic stress transcription factor 1 (OSTF1) in tilapia during osmotic stress.
Biochem. Biophys. Res. Commun.
426
,
294
298
.
Yan
B.
,
Zhu
C. D.
,
Guo
J. T.
,
Zhao
L. H.
,
Zhao
J. L.
(
2013
).
miR-206 regulates the growth of the teleost tilapia (Oreochromis niloticus) through the modulation of IGF-1 gene expression.
J. Exp. Biol.
216
,
1265
1269
.
Yi
R.
,
Fuchs
E.
(
2010
).
MicroRNA-mediated control in the skin.
Cell Death Differ.
17
,
229
235
.
Yi
R.
,
O'Carroll
D.
,
Pasolli
H. A.
,
Zhang
Z.
,
Dietrich
F. S.
,
Tarakhovsky
A.
,
Fuchs
E.
(
2006
).
Morphogenesis in skin is governed by discrete sets of differentially expressed microRNAs.
Nat. Genet.
38
,
356
362
.
Yi
R.
,
Poy
M. N.
,
Stoffel
M.
,
Fuchs
E.
(
2008
).
A skin microRNA promotes differentiation by repressing ‘stemness’.
Nature
452
,
225
229
.
Zhao
Y.
,
Samal
E.
,
Srivastava
D.
(
2005
).
Serum response factor regulates a muscle-specific microRNA that targets Hand2 during cardiogenesis.
Nature
436
,
214
220
.
Zhu
Z.
,
He
J.
,
Jia
X.
,
Jiang
J.
,
Bai
R.
,
Yu
X.
,
Lv
L.
,
Fan
R.
,
He
X.
,
Geng
J.
 et al. (
2010
).
MicroRNA-25 functions in regulation of pigmentation by targeting the transcription factor MITF in Alpaca (Lama pacos) skin melanocytes.
Domest. Anim. Endocrinol.
38
,
200
209
.

Supplementary information