Disorganization of Transcriptional Regulation and Alteration of Keratin Family Gene Expression in Hairy Ear Mice

TL;DR

This study explores how a genetic mutation in mice leads to abnormal ear hair growth by altering the expression of keratin-related genes.

Contribution

The study identifies specific keratin and keratin-associated protein genes that are dysregulated in hairy ear mice due to Hoxc gene disruption.

Findings

RNA-seq analysis revealed 2092 differentially expressed genes in hairy ear mice.

Hair keratin and keratin-associated protein genes were upregulated, while epithelial keratin genes were downregulated.

Genes like Car6, Gprc5d, and Dab2 showed altered expression, suggesting involvement in hair follicle regulation.

Abstract

Background: The hairy ear (Eh) mutation in heterozygous mice (Eh/+) results in elongated and additional ear hairs, along with altered pinna morphology compared to wild-type (+/+) mice. Previous studies suggest that disruption of the Hoxc gene cluster caused by the Eh inversion influences the hair growth cycle. Methods: To elucidate the molecular basis of this phenotype, we performed RNA-seq analysis on ear tissues from four-week-old Eh/+ and +/+ mice and compared their transcriptomic profiles. Results: Differential expression analysis identified 2092 genes, and subsequent Gene Ontology (GO) and overrepresentation analysis revealed significant alterations in hair growth-related processes, including the hair cycle and canonical keratinization in Eh/+ ears. Notably, numerous hair keratin and keratin-associated protein (Krtap) genes were markedly upregulated in Eh/+ mice. Validation by quantitative real-time PCR confirmed increased expression of randomly selected keratin genes (Krt34, Krt39, Krt71, Krt81, Krt84) and keratin-associated proteins (Krtap4-16 and Krtap22-2). In contrast, epithelial keratin genes such as Krt2 and Krt14 were downregulated in Eh/+ ears. In addition, genes associated with hair follicle growth, Car6 and Gprc5d, showed elevated expression, while Dab2, a telogen–anagen transition marker linked to hair follicle stem cell activation, was slightly increased at the telogen stage in Eh/+ compared with +/+ mice. Conclusions: These findings provide new insights into the role of Hoxc cluster genes in orchestrating the expression of hair keratin and Krtap genes and highlight potential regulatory mechanisms underlying the hairy ear phenotype.

Figures

Full text

1. Introduction

The hairy ear (Eh) mutation was first identified in the 1960s among the progeny of a neutron-irradiated male mouse [1]. Homozygous Eh/Eh mice exhibit perinatal lethality due to cleft palate formation, which results from impaired growth of the palatal shelves [2]. In contrast, heterozygous Eh/+ mice display no overt developmental abnormalities except for a smaller pinna and longer, denser hair compared to wild-type littermates [1]. The Eh mutation is caused by a paracentric inversion of the distal half of chromosome 15, with the proximal breakpoint located between Sntb1 (syntrophin basic 1) and Has2 (hyaluronan synthase 2), and the distal breakpoint between Hoxc4 (homeobox C4) and Smug1 (single-strand selective monofunctional uracil DNA glycosylase), without disrupting any coding sequences [1].

Another chromosomal inversion mutation, Koala (Koa), shares similarities with Eh, involving the distal half of chromosome 15, although the inverted region in Koa (51 Mb) is slightly larger than that in Eh (47 Mb) [3]. Koa heterozygotes (Koa/+) exhibit a hairy pinna phenotype similar to Eh/+ mice [4]. Increased expression of Hoxc cluster genes has been associated with hair follicle stem cell regeneration in Koa ear skin [5]. While changes in the expression of genes near inversion breakpoints have been reported [1], a comprehensive genome-wide transcriptomic analysis of Eh mutants has not yet been reported.

Hair follicles undergo cyclic phases collectively known as the hair cycle, which includes anagen (growth), catagen (regression), telogen (resting), and exogen (shedding), although exogen does not occur in every cycle [6]. Previous studies have reported altered expression of Hoxc cluster genes and an extended anagen phase in the ear hair follicles of Eh/+ mice compared to wild-type controls [1,2]. Multiple signaling pathways, including WNT, BMP, FGF, PDGF, TGF-β, and TNF-α, have been implicated in hair growth regulation [7,8,9,10,11], and defects in these pathways profoundly affect hair cycle progression [12,13,14,15].

Keratins (Krt) expressed in the skin can be broadly classified into epidermal keratins and hair (trichocyte) keratins, which differ in both their spatial expression patterns and biological functions. Epidermal keratins are predominantly expressed in basal and suprabasal layers of the interfollicular epidermis, where they provide mechanical stability and contribute to epidermal barrier function. In contrast, hair keratins are specifically expressed in differentiated trichocytes of the hair follicle and form the core intermediate filament network of the hair shaft [16,17].

In addition to keratins, keratin-associated proteins (KRTAP) represent a major structural component of hair. Krtap genes are expressed in hair follicle-derived cells and form an interfilamentous matrix that embeds and cross-links hair keratin intermediate filaments through extensive disulfide bonding. Rather than constituting intermediate filaments themselves, KRTAP contributes to the mechanical strength, rigidity, and resilience of the hair shaft by organizing keratin filaments into higher-order structures [18,19,20].

The coordinated and stage-specific regulation of hair keratin and Krtap gene families is essential for proper hair follicle differentiation and hair shaft formation, and alterations in their expression have been associated with diverse hair phenotypes across mammalian species. Given these well-established distinctions between epidermal and hair-specific keratin systems, differential regulation of these gene classes provides an important framework for interpreting transcriptomic changes associated with altered hair growth [21,22].

In this study, we conducted RNA-seq analysis to compare gene expression profiles between Eh/+ and wild-type ear tissues, aiming to elucidate the genome-wide impact of chromosomal inversion. Our findings reveal that the Eh/+ mutation alters the expression of numerous hair growth-related genes, including most hair Krt and Krtap. Furthermore, we demonstrate that hair Krt genes and epithelial Krt genes are subject to distinct regulatory mechanisms.

2. Materials and Methods

2.1. Animal and Sample Preparation

Eh/+ mice (B6By.Cg-Eh/J, RRID: IMSR_JAX:000523) were obtained from the Jackson Laboratory (Bar Harbor, ME, USA) and maintained under specific pathogen-free conditions at the Laboratory Animal Research Center of Konkuk University (KULARC). Animals were housed under a 12 h light/dark cycle with sterilized food and water provided ad libitum. Environmental conditions were controlled at 24 ± 1 °C and 55 ± 5% relative humidity. All experimental procedures were approved by the Konkuk University Animal Care and Use Committee (protocol number KU19224). Ear tissues were collected from four-week-old Eh/+ and wild-type (WT) mice and stored at −70 °C until further analysis.

2.2. Genotyping

Tail tips (5 mm) were excised using a sterile razor blade and incubated in 100 µL tail lysis buffer (Viagen Biotech Inc., Los Angeles, CA, USA) containing 25 mM proteinase K (Sigma-Aldrich, St. Louis, MO, USA) at 55 °C overnight. Lysates were heat-inactivated at 90 °C for 1 h. Genotyping PCR was performed using 50 ng genomic DNA, 0.5 µM of each primer, and the following primer sets: wild-type allele—WT-L (5′-ATGGGAGGCAAGAAGAACCT-3′) and WT-R (5′-ACTGCCCAGTGGTCCTTAAA-3′); hairy ear allele (Eh)—HE-L (5′-GGTGGGCCATATGGAGATTA-3′) and WT-R. The reaction mixture contained 100 µM dNTPs, 4× loading dye, 10× e-Taq Reaction Buffer (25 mM MgCl_2_), and 0.5 U Solg™ e-Taq DNA Polymerase (SolGent, Seoul, Republic of Korea). PCR was carried out on a Veriti 96-Well Thermal Cycler (Thermo Fisher Scientific, Waltham, MA, USA) under the following conditions: 95 °C for 5 min; 35 cycles of 95 °C for 30 s, 57 °C for 30 s, and 72 °C for 30 s; final extension at 72 °C for 5 min. PCR products were resolved on 1% agarose gels and visualized under UV illumination.

2.3. RNA Isolation and RNA-Seq

Ear tissues from Eh/+ and +/+ mice of both sexes were homogenized in liquid nitrogen, followed by the addition of 1 mL RNA extraction buffer (R&A-BLUE Total RNA Extraction Kit; iNtRON Biotechnology, Seongnam, Republic of Korea). Lysates were transferred to RNase-free tubes on ice, and total RNA was purified using the RNeasy Mini Kit (Qiagen, Valencia, CA, USA) according to the manufacturer’s protocol. RNA integrity was assessed using an Agilent Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA), and samples with RNA integrity number (RIN) > 8.0 were used for sequencing. RNA-seq libraries were prepared using the TruSeq RNA Sample Prep Kit (Illumina, San Diego, CA, USA) and sequenced on HiSeq 2000 or NovaSeq 6000 platforms (Illumina) to generate paired-end reads (151 bp). Library preparation and sequencing were performed by Macrogen (Seoul, Republic of Korea). The RNA-seq data was submitted to NCBI SRA under the accession number of PRJNA1390433.

2.4. Identification of Differentially Expressed Genes (DEGs) from RNA-Seq

Raw RNA-seq reads were processed using Trimmomatic (v0.39) to remove adapter sequences and low-quality bases (LEADING:3, TRAILING:3, SLIDINGWINDOW:4:15) [23]. Reads shorter than 101 bp were discarded. Trimmed reads were aligned to the mouse reference genome GRCm39 (RefSeq: GCF_000001635.27) using the STAR aligner (v2.7.11b) with ENCODE standard options and the following adjustments: maximum number of loci per read set to 20 (--outFilterMultimapNmax 20) and minimum overhang for novel splice junction detection set to 8 bp (--alignSJoverhangMin 8) [24]. GENCODE M37 (Ensembl 114) annotation was provided for junction guidance and downstream quantification.

Read counts mapped to gene exons were quantified using featureCounts (v2.0.6) with reverse-stranded mode (-s 2) [25]. Differential expression analysis was generated using edgeR (v4.40) [26,27]. Samples were grouped by genotype (Eh/+ and wild type (+/+)), pooling sexes as biological replicates. Genes with counts per million (CPM) > 1 in at least one genotype were retained. Library sizes were normalized using the trimmed mean of M-values (TMM) method, and statistical significance for two-group comparisons was assessed using the exact test. DEGs were defined as those with ≥2-fold change (FC) and a Benjamini–Hochberg false discovery rate (FDR) < 0.01.

2.5. Statistical Overrepresentation Analysis

Gene Ontology (GO) overrepresentation analysis of differentially expressed genes (DEGs) was performed using PANTHER (v19.0) (http://www.pantherdb.org) [28]. Statistical significance was assessed using Fisher’s exact test with false discovery rate (FDR) correction. Analyses were conducted against the GO Biological Process (BP) Complete dataset, and terms with fold enrichment greater than 2 and FDR < 0.01 were considered significant.

2.6. Real-Time Quantitative PCR (RT-qPCR)

Reverse transcription was carried out in 20 µL reactions using 2 µg of total RNA, oligo(dT)15 primer (Santa Cruz Biotechnology, Dallas, TX, USA), 10 µM dNTPs, and the SuperiorScript III cDNA Synthesis Kit (Enzynomics, Daejeon, Republic of Korea) at 50 °C for 50 min, followed by enzyme inactivation at 72 °C for 15 min. Genomic DNA contamination was evaluated by end-point PCR for Gapdh using exon–exon junction primers (5′-CTCACTCAAGATTGTCAGCA-3′ and 5′-GTCATCATACTTGGCAGGTT-3′) and intron-specific primers (5′-GCATCCTGACCTATGGCGTA-3′ and 5′-GCATCATCGAACCTCTCCCC-3′). PCR cycling conditions included initial denaturation at 95 °C for 5 min, followed by 40 cycles of 95 °C for 30 s, annealing at 57 °C (exon–exon primers) or 60 °C (intronic primers) for 40 s, and extension at 72 °C for 30 s, with a final extension at 72 °C for 5 min.

Real-time quantitative PCR (qPCR) was performed using SsoAdvanced Universal SYBR Green Supermix (Bio-Rad Laboratories, Hercules, CA, USA) with 100 ng cDNA and gene-specific primers (Table 1) whose primer efficiencies range between 82.6–106.6% (Table S1) on a Bio-Rad CFX Connect system, following the manufacturer’s protocol. The thermal profile consisted of 95 °C for 30 s (initial denaturation), followed by 45 cycles of 95 °C for 10 s and 60 °C for 30 s. Melt curve analysis was conducted by stepwise temperature increases from 65 °C to 90 °C in 0.5 °C increments, with a 5 s hold at each step. All reactions were performed in triplicate, and data were analyzed using Bio-Rad CFX Manager software (v3.1). Relative expression was calculated as −ΔC_q_, where ΔC_q_ = C_q_(target) − C_q_(Gapdh). For each gene, time point, and sex, qPCR was performed on three biological replicates (independent animals), with three technical replicates per sample. Group mean −ΔC_q_ values were compared between sexes using Welch’s two-sample t-test. Because no significant sex effect was detected (p > 0.01), data from males and females were pooled and treated as biological replicates for each genotype. All plots were generated using R (v4.4.2).

3. Results

3.1. Development of a Genotyping Protocol for the Hairy Ear Mutation (Eh)

We designed primers flanking the distal breakpoint region of both the Eh and wild-type alleles, based on their published sequences (GenBank accession numbers: wild-type—NC_000081.6; Eh—AY757367.1) [1]. Using a region-specific common primer, we developed a PCR-based genotyping assay capable of distinguishing between the Eh and wild-type alleles (Figure 1A). The resulting amplicon sizes were 440 base pairs (bp) for the wild-type allele and 230 bp for the Eh allele (Figure 1B). In heterozygous mice (Eh/+), both bands were consistently detected. Genotyping results aligned precisely with phenotypic observations (Figure 1C); all Eh/+ mice exhibited the characteristic hairy ear phenotype (Figure 1C). No Eh/Eh homozygous mice were recovered in surviving litters, consistent with the known postnatal lethality associated with homozygosity for the Eh allele.

3.2. Genome-Wide Transcriptomic Alterations in Eh/+ Ear Tissues Revealed by RNA-Seq

To compare gene expression profiles between Eh/+ and +/+ mouse ear tissues, we performed RNA sequencing (RNA-seq) on postnatal day 28 (P28) samples from four animals (one male and one female per genotype). Sequencing generated 157.96 million paired-end reads (151 bp each), totaling approximately 44.75 Gb of data. Reads were aligned to the mouse reference genome (GRCm39; RefSeq: GCF_000001635.27) with an average mapping rate of 89.5 ± 1.7% (Table S2).

Gene expression was normalized as counts per million (CPM) using the trimmed mean of M-values (TMM) method (Figure S1). Principal component analysis (PCA) revealed genotype as the major source of transcriptional variation, with PC1 explaining 70.0% of total variance and clearly separating Eh/+ from +/+ samples (Figure S2). Based on this clustering, male and female samples within each genotype were treated as biological replicates for differential expression analysis.

We identified 2092 differentially expressed genes (DEGs) between Eh/+ and +/+ ears (fold change (FC) > 2; FDR < 0.01), including 1175 upregulated and 917 downregulated genes among 16,082 expressed genes (CPM > 1; Figure S3A and Table S3). Of these, 108 DEGs mapped to chromosome 15 (Figure 2A–C and Table S4), with 83.3% (90/108) clustered within the Eh inversion region, suggesting direct transcriptional effects of the chromosomal rearrangement. The remaining 18 DEGs outside the inversion were all upregulated.

Gene Ontology (GO) Biological Process (BP) overrepresentation analysis of chromosome 15 DEGs highlighted processes such as keratinization (GO:0031424), intermediate filament organization (GO:0045109), intermediate filament cytoskeleton organization (GO:0045104), and keratinocyte differentiation (GO:0030216) (fold-enrichment > 2 & Fisher’s exact test, FDR < 0.01; Figure 2D and Table S5). These enrichments were largely driven by the strong upregulation of hair keratin genes. When the analysis was extended to all DEGs (n = 2092), additional hair-related processes emerged, including hair cycle (GO:0042633), hair follicle development (GO:0001942), and molting cycle (GO:0042303), while keratinization- and intermediate filament-associated terms remained significantly enriched (FDR < 0.01; Figure S3B and Table S6), consistent with the chromosome 15-specific profile. Furthermore, other BPs such as epidermis development (GO:0008544), skin development (GO:0043588), and transmembrane transport processes (organic anion transport, GO:0015711; inorganic anion transport, GO:0015698) were also significantly represented.

Given that the GO BP overrepresentation analysis identified keratinization as a significantly affected pathway, we next examined the expression profiles of all detected keratin (Krt) and keratin-associated protein (Krtap) genes in our dataset. A total of 52 keratin genes were expressed, including 29 epithelial keratins, 23 hair keratins, and one unclassified keratin (Krt222) (Figure 3A; Table S7). Remarkably, all 23 hair keratin genes were significantly upregulated in Eh/+ ear tissues, regardless of chromosomal location (Figure 3A).

In contrast, epithelial keratin genes displayed heterogeneous expression patterns. On chromosome 11, most epithelial keratins were modestly upregulated, with fold changes (FCs) lower than those observed for hair keratins. Notably, Krt14, Krt19, and Krt24 were significantly downregulated (FC < 0.5). On chromosome 15, epithelial keratins such as Krt18, Krt87, and Krt90 were strongly upregulated, whereas Krt2, Krt4, and Krt8 were significantly downregulated. Additionally, several epithelial keratins, including Krt76 and Krt88, were not expressed in both Eh/+ and +/+ samples. We also observed robust upregulation of 78 Krtap genes in Eh/+ samples, with no Krtap genes showing significant downregulation (Figure 3B and Table S7). Chromosome 11 contained the largest number of differentially expressed Krtap genes (35 of 78), followed by chromosome 16 (26 of 78).

3.3. Upregulation of Hoxc4, Hoxc5, and Hoxc13 in Eh/+ Ear Transcriptome of 4-Week-Old Mice

The Hoxc gene cluster within the inversion contains nine genes: Hoxc4, Hoxc5, Hoxc6, Hoxc8, Hoxc9, Hoxc10, Hoxc11, Hoxc12, and Hoxc13. Among these, we observed significant upregulation of Hoxc4 (FC = 42.88 & FDR = 2.34 × 10^−11^), Hoxc5 (FC = 7.38 & FDR = 1.67 × 10^−4^), and Hoxc13 (FC = 6.96 & FDR = 1.24 × 10^−16^) in the P28 Eh/+ ear transcriptome compared to +/+ controls (Figure 4 and Table S3), whereas the remaining cluster genes were not expressed at detectable levels (Table S4). These findings are consistent with the notion that increased expression of Hoxc genes contributes to the extended anagen phase in the hair cycle, although their expression appears to be developmentally regulated [1,29,30].

3.4. Hair Cycle Marker and Keratin Gene Expression Dynamics in Eh/+ and Wild-Type Mice

An extended anagen phase has been proposed as a key factor contributing to the increased hair length observed in Eh/+ ears compared to wild-type ears. To investigate this, we measured the expression of hair cycle stage marker genes [31,32]—Car6 (anagen), Gprc5d (mid- to late anagen), and Dab2 (telogen–anagen transition)—at three developmental time points (P10, P14, and P28) using real-time quantitative PCR (RT-qPCR) (Figure 5).

In both genotypes, all three genes showed increased expression from P10 (anagen) to P14 (transition to catagen), followed by a decline at P28 (telogen), with peak expression at P14. Notably, Car6 and Gprc5d levels were significantly higher in Eh/+ samples than in +/+ at P10 (p < 0.01). In contrast, Dab2 showed an opposing expression pattern during anagen and catagen, with slightly reduced expression in Eh/+ samples, followed by a modest increase at the telogen stage, which may contribute to the denser hair phenotype in mutant ears.

We further analyzed the expression of three hair keratin genes (Krt34, Krt71), two keratin-associated protein genes (Krtap4-16, Krtap22-2), and two epithelial keratin genes (Krt2, Krt14) across the same stages. All hair keratin and keratin-associated protein genes were upregulated in Eh/+ samples, peaking at P14 (Figure 5). Conversely, epithelial keratin genes showed lower expression in Eh/+ compared to wild-type mice.

4. Discussion

In this study, we performed differential expression analyses of Eh/+ and +/+ ear transcriptomes and identified a substantial upregulation of hair keratin (Krt) and keratin-associated protein (Krtap) genes in Eh/+ samples. Gene Ontology (GO) and pathway enrichment analyses indicated that these differentially expressed genes are primarily involved in biological processes such as intermediate filament organization, keratinization, and hair cycle. While Krt and Krtap genes are the major components of hair fibers and are important in hair follicle morphogenesis and the hair cycle [33], the causal relationship between their overexpression and the observed hairy ear phenotype remains unclear. Our findings suggest a potential link between elevated expression of these genes and hair growth, but whether this represents a driving mechanism or a secondary outcome of the phenotype requires further investigation. GO analysis revealed enrichment of DEGs associated with intermediate filament organization, keratinocyte differentiation, and hair cycle in Eh/+ ears (Table S6).

To conduct transcriptome analysis, a total of four animals with one biological replicate per genotype were used. Comparative analysis of ear transcriptomes from 4-week-old Eh/+ and wild-type mice identified 2092 DEGs. Principal component analysis (PCA) showed that PC1 accounted for approximately 70.0% of the variance, primarily reflecting genotype differences likely driven by the chromosomal inversion, whereas PC2 captured residual variation associated with sex differences within each group. This observation aligns with previous reports indicating that hair phenotypes are also influenced by sex [34]. To validate the RNA-seq findings, the expression patterns of selected DEGs were examined using real-time quantitative PCR. The qPCR results were largely consistent with RNA-seq data, supporting the robustness and reliability of our transcriptomic analysis despite the limited sample size (Figure S4).

The mouse genome contains more than 50 Krt genes, which can be classified into hair keratins and epithelial keratins [35]. Krtap genes are expressed in cell populations within hair follicles, nails, and cornified parts of filiform papillae on the tongue [22,36,37]. Several Krt and Krtap genes are not only structural components of hair but also play roles in regulating keratinocyte differentiation and the hair cycle [38,39,40].

Unlike hair keratins, which act as effector molecules for hair growth, epithelial keratins primarily function in keratinocyte migration and differentiation. Therefore, precise regulation of Krt and Krtap expression is essential for normal hair development [35]. Interestingly, the hairy ear mutation selectively upregulates hair keratin genes rather than epithelial keratins, suggesting distinct regulatory mechanisms governing these two keratin classes.

Krt2 was the most strongly downregulated keratin in the Eh/+ ear skin compared with wild-type controls (Figure 3 and Figure 5). The high basal expression of Krt2 and its distinct association with the ear phenotype are consistent with the tissue-specific expression pattern of Krt2 in mice, as it is predominantly expressed in ear epidermis but not in more densely haired skin regions such as the dorsal skin [41]. Furthermore, the restricted expression of Krt2 and the inverse correlation between Krt2 expression levels and hair growth observed in this study support a functional role for Krt2 in tissue-specific regulation of ear hair growth, alongside the upregulation of hair-type keratin genes.

Furthermore, the observation that hair keratin genes located on both the inversion-associated chromosome (MMU15) and a non-inverted chromosome (MMU11) were upregulated indicates the involvement of both cis- and trans-acting factors in this process, although the underlying mechanism remains unclear. In contrast, epithelial keratin genes exhibited inconsistent changes, with some upregulated and others downregulated. This variability suggests that epithelial keratin expression is also affected in Eh/+ ears, albeit without a clear directional pattern. Given this inconsistency, the altered expression of epithelial keratins is unlikely to be directly driven by Hoxc dysregulation.

Current knowledge on the regulatory mechanisms controlling keratin and Krtap gene expression remains limited. Previous studies suggest that Hoxc promotes hair follicle cycling, potentially through WNT signaling pathways [5,42,43]. In particular, Hoxc13 and WNT signaling have been identified as key regulators of hair keratin gene expression [36]. In contrast, epithelial keratin genes are primarily regulated by transcription factors such as AP1, AP2, SP1, POU domain proteins, C/EBP, glucocorticoid receptor, and NF-κB [44]. These distinct regulatory networks between hair keratin and epithelial keratin genes are consistent with our observation of robust upregulation of hair keratin genes in Eh/+ ears. Moreover, NF-κB has also been implicated in the transcriptional control of Krtap [45]. The concurrent upregulation of hair keratin and Krtap genes in the Eh/+ ear transcriptome suggests the involvement of a shared regulatory mechanism driving their coordinated expression.

Previous studies have shown that the expression of Hoxc cluster genes differs across developmental stages in Eh/+ and wild-type mice [1]. In our analysis, only Hoxc4, Hoxc5, and Hoxc13 were significantly upregulated at four weeks of age, corresponding to the telogen stage of the hair cycle, whereas other Hoxc genes were expressed at undetectable levels. We also observed age-dependent differences in hair keratin and Krtap expression between Eh/+ and wild-type mice (P10, P14, and P28; Figure 4). Consistent with previous findings in cashmere goats [36], Hoxc13 appears to regulate the expression of multiple hair keratin genes in mice. Because the entire Hoxc cluster gene is inverted in the Eh mutation, the observed upregulation of hair Krt and Krtap in Eh/+ ears may result from altered regulatory mechanisms affecting Hoxc expression.

A previous study reported that hair follicles in Eh/+ mice remained in the anagen (growth) phase for approximately three days longer (postnatal days 12–15) compared to +/+ littermates [1]. This observation is consistent with the higher expression of Car6 and Gprc5d, markers of anagen and mid- to late anagen, respectively, in Eh/+ ears compared to wild-type ears. Gprc5d has been implicated in hair cycle regulation and keratinization [46,47]. The increased expression of Dab2, a marker of the telogen–anagen transition, observed at the telogen stage in Eh/+ ears may facilitate subsequent hair follicle stem cell activation during anagen. This effect is potentially linked to dysregulation of Hoxc cluster genes, as Hoxc-dependent regulation of hair follicle stem cell regeneration has previously been demonstrated in the ear skin of Koa mutant mice [5].

It has also been shown that the hair cycle of mouse ear follicles remains in an extended telogen phase for at least three months, whereas dorsal hair follicles complete three cycles during the same period. This prolonged telogen phase has been suggested as a reason why the hairy phenotype is restricted to ear hair rather than extending to other body regions [48].

Chromosomal inversions are known to suppress genetic recombination and maintain haplotype linkages in natural populations, contributing to divergence in multiple traits [49,50,51]. Evaluating whether the Eh inversion suppresses recombination would provide valuable insight into its evolutionary implications.

Animal models have long been used in hair research to elucidate physiological and pathological processes underlying hair loss and regeneration [52]. In vivo studies have been conducted in various species, including mice, rats, hamsters, rabbits, sheep, and monkeys [53]. Eh/+ mice, together with Koa mutants, may serve as informative models for investigating the regulatory mechanisms of hair Krt and Krtap genes.

This study demonstrates that the Eh chromosomal inversion in mice profoundly alters ear transcriptomes, leading to significant upregulation of hair Krt and keratin-associated protein (Krtap) genes. Our findings suggest that Hoxc cluster dysregulation may contribute to the hairy ear phenotype through complex regulatory mechanisms. A limitation of this study is the small sample size used for RNA-seq analysis (n = 2 per genotype). However, the robustness of the transcriptomic findings was supported by validation using RT-qPCR with a larger number of biological replicates for selected genes. Although the precise causal relationship remains unclear, this work provides a foundation for future studies to elucidate cis- and trans-acting factors involved in keratin gene regulation and to explore the evolutionary implications of chromosomal inversions.