The research began with a question of timing. Hair follicles do not simply appear; they arise through a tightly choreographed sequence of interactions between the embryonic epidermis and the underlying dermis, and the genes that drive this process are thought to be active only during narrow developmental windows. To find those windows, the team collected fetal skin from cashmere goats at four gestational stages—days 45, 55, 65, and 75—obtained by cesarean section from does maintained under identical environmental and dietary conditions. Using the same samples as a previously published study, and with approval from the Laboratory Animal Ethics Committee of Inner Mongolia Agricultural University, they profiled gene expression across these stages and searched for long non-coding RNAs whose behavior changed in a stage-specific, transient manner.

Long non-coding RNAs, or lncRNAs, are transcripts that do not code for proteins but can regulate gene expression in a variety of ways. Some act inside the nucleus, scaffolding chromatin complexes; others linger in the cytoplasm, where they can interact with microRNAs. The screen identified 32 candidate lncRNAs with transient, stage-specific expression changes at embryonic day 55, a finding that suggests this single day may represent an important hub of transcriptional regulation during follicle morphogenesis. Enrichment analysis of the accompanying gene expression data showed positive enrichment of Wnt-related gene sets in the day 55 versus day 45 comparison and in the day 65 versus day 55 comparison. The authors are careful to note that this enrichment does not constitute a direct measurement of Wnt pathway activity, but it places the day 55 window squarely in the territory of one of developmental biology’s most celebrated signaling systems.

Among the 32 candidates, one transcript stood out. Known by the unglamorous identifier MSTRG.18075.2, this lncRNA showed relatively high expression at embryonic day 55, exactly the stage flagged by the temporal screen. Subcellular localization experiments revealed that the molecule resides predominantly in the cytoplasm—a detail that matters enormously for what the researchers proposed next. Cytoplasmic lncRNAs are the raw material of one of molecular biology’s most debated regulatory models: the competing endogenous RNA, or ceRNA, hypothesis. In this model, a lncRNA acts as a molecular sponge, soaking up microRNAs that would otherwise bind to and suppress messenger RNAs. By titrating microRNAs away from their targets, the lncRNA indirectly protects those targets, effectively adding a layer of cross-talk between transcripts that never touch each other physically.

The microRNA at the center of the proposed circuit is chi-miR-145-5p, and the target is WNT16, a member of the Wnt family of secreted signaling proteins that play fundamental roles in embryonic patterning, including the initiation and development of hair follicles. The team first tested the predicted interactions in a heterologous reporter system using 293T cells, a standard laboratory workhorse derived from human embryonic kidney tissue. When chi-miR-145-5p was introduced, it reduced the activity of wild-type reporter constructs containing the predicted binding sites from either MSTRG.18075.2 or the 3′ untranslated region of WNT16. Crucially, mutant reporters in which those binding sites were disrupted did not show the same response. That pattern—suppression of the wild-type sequence but not the mutated one—is the classic signature of a sequence-specific microRNA interaction, and it lent credibility to the predicted pairing on both ends of the proposed axis.

Reporter assays, however, are performed in a foreign cellular context, and the authors acknowledge this limitation explicitly. The more biologically meaningful experiments took place in cashmere goat dermal fibroblasts, the dermal cells that provide the structural and signaling environment in which follicles form. When the researchers knocked down MSTRG.18075.2 in these cells, a coherent set of phenotypes emerged: apoptosis was reduced, cell-cycle progression slowed, proliferation and migration were impaired, and intracellular reactive oxygen species accumulated. Each of these cellular behaviors is relevant to morphogenesis, because follicle development requires precisely timed waves of cell proliferation, movement, and survival in the dermal compartment. A transcript whose perturbation disturbs all of these processes simultaneously is a plausible participant in the developmental program, even if the exact in vivo role remains unproven.

The genetic logic of the ceRNA model makes a specific prediction: if MSTRG.18075.2 protects WNT16 by soaking up chi-miR-145-5p, then removing the microRNA should partially rescue the effects of removing the lncRNA. That is essentially what the team observed. Combined knockdown of chi-miR-145-5p partially reversed some of the phenotypes induced by knocking down either MSTRG.18075.2 or WNT16. The rescue was partial, not complete, which is consistent with the messy reality of cellular networks—microRNA-145-5p almost certainly has other targets, and MSTRG.18075.2 may have other functions—but the direction of the effect supports the candidate regulatory model the authors propose: MSTRG.18075.2 sponges chi-miR-145-5p, thereby relieving repression of WNT16.

To their credit, the researchers draw the boundaries of their claims with unusual precision. They note that endogenous RNA complexes were not examined, that downstream Wnt signaling activity was not directly measured, and that no experiments tested whether the axis influences hair follicle morphogenesis in living embryos. The findings, they write, support a relationship between MSTRG.18075.2 and the candidate chi-miR-145-5p/WNT16 axis at the levels of temporal expression, cytoplasmic localization, sequence-specific reporter responses, and in vitro cellular phenotypes—but they do not directly establish an endogenous ceRNA mechanism or a causal relationship in vivo. In a field where the ceRNA hypothesis has been criticized for overreach, this restraint is notable and strengthens rather than weakens the study’s contribution.

The broader significance of the work lies in what it offers for the future of cashmere production. Cashmere goats are the economic backbone of many herding communities in Inner Mongolia, and the fineness and density of the undercoat fibers they produce are determined largely by the number and quality of secondary hair follicles established during embryonic development. If regulatory circuits like the one described here can be confirmed in vivo, they could eventually inform selective breeding strategies or even molecular interventions aimed at improving fiber yield and quality. The study was supported by the Inner Mongolia Autonomous Region Science and Technology Program and the Innovative Research Team Program for Higher Education Institutions of Inner Mongolia Autonomous Region, reflecting regional investment in the genetics of this prized livestock resource.

For now, MSTRG.18075.2 remains a candidate rather than a confirmed conductor of follicle development, and the day 55 window it marks is a lead rather than a conclusion. But the study exemplifies a productive middle path in modern genomics: a temporal screen to find the moment, localization studies to find the compartment, reporter assays to test the chemistry, and cell-based phenotyping to test the biology. Each step alone would be suggestive; together they form a coherent, testable hypothesis about how a non-coding transcript in the cytoplasm of embryonic skin cells might help decide how many follicles a future goat will grow. The next chapter—demonstrating the mechanism inside the developing follicle itself—will be the one that determines whether this molecular whisper becomes a roar.

Subject of Research: A candidate lncRNA/microRNA/WNT16 regulatory axis in embryonic hair follicle development of cashmere goats

Article Title: Cytoplasmic lncRNA MSTRG.18075.2 associated with embryonic hair follicle development in cashmere goats: dermal fibroblast phenotypes and a candidate chi-miR-145-5p/WNT16 regulatory axis

Article References: Bao, X., Ma, R., Pan, J., Wang, Y., Qiao, J., Ma, Q., Wang, R., Shang, F., & Zhang, Y. (2026). Cytoplasmic lncRNA MSTRG.18075.2 associated with embryonic hair follicle development in cashmere goats: dermal fibroblast phenotypes and a candidate chi-miR-145-5p/WNT16 regulatory axis. BMC Genomics. https://doi.org/10.1186/s12864-026-13305-6

Image Credits: AI Generated

DOI: 10.1186/s12864-026-13305-6

Keywords: cashmere goat, lncRNA, MSTRG.18075.2, chi-miR-145-5p, WNT16, ceRNA, hair follicle morphogenesis, dermal fibroblasts, Wnt signaling, embryonic development, BMC Genomics, Inner Mongolia

Hoxc13 belongs to the Hox family of homeobox transcription factors, master regulators of embryonic patterning that specify body plans along the head-to-tail axis. Within this family, Hoxc13 has long been recognized as a critical controller of integumentary appendages, the skin-derived structures that include hair, nails, and claws. Previous work by the same team had shown that during mammalian evolution, the Hoxc13 protein acquired a characteristic polyglycine repeat insertion that distinguishes it from its non-mammalian homologs, the related proteins found in birds, reptiles, and other vertebrates. The new study set out to test whether this small structural addition actually changes what the protein can do at the genome-wide level, and whether that change could plausibly have contributed to the emergence of hair.

To answer this question, the researchers employed an elegant experimental design built around two versions of the Hoxc13 protein from cashmere goats. The first, designated Hoxc13-W, is the wild-type protein carrying the intact polyglycine repeat. The second, Hoxc13-S, is a shortened variant in which the polyglycine fragment has been deleted, mimicking the ancestral, non-mammalian form of the protein. Both proteins were synthesized using an in vitro expression system, allowing the team to compare their DNA-binding behavior directly without the confounding effects of living cells. This reductionist approach isolates the intrinsic DNA-binding capacity of each protein variant, a critical consideration when the goal is to attribute functional differences to a specific structural feature.

The centerpiece of the study was DNA affinity purification sequencing, or DAP-seq, a technique that uses purified proteins to fish out genomic DNA fragments they can bind, which are then identified by high-throughput sequencing. The results were striking. Hoxc13-W, the version with the polyglycine repeat, identified 12,679 binding peaks across the goat genome. Hoxc13-S, lacking the insertion, identified only 3,634 peaks. Even more telling was the overlap: just 243 peaks were shared between the two proteins. In other words, the vast majority of genomic sites bound by the wild-type protein were not recognized by the polyglycine-deleted variant, and vice versa. A single short amino acid insertion had fundamentally rewired the DNA-binding landscape of this transcription factor, transforming it from a relatively restricted regulator into one with a vastly expanded reach.

Functional enrichment analysis of the genes associated with these binding peaks revealed why this expansion matters for hair. The targets of Hoxc13-W were specifically enriched in pathways governing hair follicle morphogenesis and cycling, the developmental programs that build follicles and drive them through their growth, regression, and resting phases. By contrast, the targets of Hoxc13-S clustered primarily in the axon guidance pathway, a set of genes involved in steering neuronal growth cones rather than in skin biology. This shift in target preference suggests that the polyglycine insertion did not merely add more binding sites but redirected the protein toward an entirely different set of biological functions, aligning its regulatory activity with the demands of building hair-producing skin.

One of the most consequential findings concerned the Wnt signaling pathway, a cascade of intercellular communication that plays a central and well-documented role in hair follicle development, stem cell activation, and the hair growth cycle. Within this pathway, Hoxc13-W bound 30 genes, while Hoxc13-S bound only 5. The sixfold difference in Wnt pathway coverage implies that the polyglycine-bearing form of Hoxc13 can exert far broader control over one of the most important signaling networks in hair biology. Because Wnt signaling is repeatedly recruited at multiple stages of follicle morphogenesis and regeneration, an expanded capacity to regulate Wnt pathway genes could have provided mammalian ancestors with finer and more powerful control over hair production.

The study also probed how the polyglycine repeat affects the recognition of cis-regulatory motifs, the short DNA sequences to which transcription factors attach. Motif analysis demonstrated that the polyglycine fragment substantially influenced the motifs associated with Hoxc13 binding beyond the core Hoxc13-binding motif, which has the consensus sequence 5′-ATAAA-3′. This suggests that the insertion does not simply strengthen or weaken binding to the same sequence; rather, it changes the sequence context in which the protein operates, likely through cooperation with other DNA-bound factors or through altered preferences for flanking sequence features. Such changes in motif usage can ripple through regulatory networks, opening or closing access to large suites of target genes across the genome.

Molecular docking analysis added a structural dimension to the story. The computational modeling indicated that the polyglycine fragment alters the N-terminal structure of the Hoxc13 protein, the region involved in protein-protein interactions. Glycine is the smallest amino acid and confers exceptional conformational flexibility, so a repeat of glycine residues can act as a flexible hinge or spacer within a protein. By reshaping the N-terminal architecture, the insertion appears to affect how Hoxc13 interacts with partner proteins, which in turn influences how it engages DNA in a cellular context, since many transcription factors bind DNA effectively only as part of larger protein complexes. The polyglycine repeat, in this view, functions as a gain-of-function modification that enhances both the DNA-binding breadth and the interaction capacity of the protein.

Taken together, the findings support a model in which the polyglycine repeat insertion in Hoxc13 acted as an evolutionary gain-of-function mutation that amplified the Hoxc13-mediated gene regulatory network during the transition from non-mammalian to mammalian vertebrates. Rather than requiring wholesale changes in gene sequence or the emergence of entirely new genes, this scenario illustrates how a modest insertion of flexible, low-complexity sequence within a single transcription factor can reconfigure its binding profile, redirect its pathway targets, and reshape its protein interactions. Such low-complexity insertions are increasingly appreciated as a rich source of evolutionary novelty, capable of tuning protein behavior with relatively small mutational steps.

The research, conducted by Guanghao Yang, Chaoyong Huang, and colleagues at Inner Mongolia Minzu University, also carries practical implications beyond evolutionary biology. Cashmere goats, the species from which the two protein variants were derived, are prized for their fine undercoat fibers, and a deeper understanding of the Hoxc13-centered regulatory network could inform breeding strategies aimed at improving fiber quality and yield. More broadly, the study demonstrates the power of combining DAP-seq with pathway enrichment, motif analysis, and molecular docking to dissect the functional consequences of protein sequence evolution. While the polyglycine repeat is certainly not the sole driver of mammalian hair, the evidence presented here makes a compelling case that this tiny stretch of glycine helped tip the balance toward the furry integument that now defines the class Mammalia, from the wool of sheep to the whiskers of cats to the hair on our own heads.

Subject of Research: The role of a Hoxc13 polyglycine repeat insertion in mammalian hair and integument evolution

Article Title: Hoxc13 polyglycine repeat gain-of-function contributes to mammalian integument evolution by altering targeted genes and interactions

Article References: Yang, G., Huang, C., Yu, M., Li, W., Wu, D., Li, C., Hu, S., & Wu, J. (2026). Hoxc13 polyglycine repeat gain-of-function contributes to mammalian integument evolution by altering targeted genes and interactions. BMC Genomics. https://doi.org/10.1186/s12864-026-13314-5

Image Credits: AI Generated

DOI: 10.1186/s12864-026-13314-5

Keywords: Hoxc13, polyglycine repeat, hair follicle, integument evolution, transcription factor, DAP-seq, Wnt signaling, gene regulation, mammalian evolution, cashmere goat, molecular docking, BMC Genomics