ENKD1 Modulates Skin Elasticity Through Microtubule Stability Regulation

ABSTRACT

Skin elasticity is critical for maintaining skin function, yet the molecular mechanisms governing this process remain incompletely understood. Herein, we identify enkurin domain‐containing protein 1 (ENKD1) as a key regulator of skin elasticity by modulating microtubule stability in basal keratinocytes. In Enkd1 knockout mice, impaired migration of basal keratinocytes results in reduced epidermal elasticity compared to wild‐type controls. Mechanistically, ENKD1 localizes to the centrosome and microtubules, where its expression enhances microtubule stability. Conversely, the absence of ENKD1 destabilizes microtubules, which likely impedes keratinocyte migration and compromises epidermal elasticity. Further investigations suggest that ENKD1 exerts its effects on microtubule stability via EB1. Collectively, these findings establish ENKD1 as a pivotal regulatory factor of mammalian epidermal elasticity, providing new insights into the molecular underpinnings of skin function.

1. Introduction

As the largest organ in the human body, the skin functions as a versatile protective barrier critical to sustaining physiological homeostasis. It consists of three distinct layers: the epidermis, dermis, and subcutaneous tissue (Everett and Sommers 2013). Epidermal stem cells (EPSCs) reside in the basal layer of the epidermis, anchored to the basement membrane that demarcates the epidermal and dermal compartments (Blanpain and Fuchs 2009; Ge and Fuchs 2018). During differentiation, these cells migrate upward through the stratified epidermal layers to reach the skin surface. The epidermal basement membrane zone (BMZ), located between the basal epidermis and papillary dermis, comprises basal keratinocytes and dermal fibroblasts. This specialized structure anchors actin microfilaments through focal adhesion, ensuring robust epidermal‐dermal cohesion. Disruption of key anchoring components, such as hemidesmosomes, increases dermal‐epidermal fragility, potentially leading to conditions like epidermolysis bullosa (EB) (Harvey et al. 2022; Natsuga et al. 2021).

The dermal elastic fiber network is a vital structural component that endows skin with its characteristic elasticity (Zi et al. 2025). Elastic protein is initially biosynthesized as a soluble precursor called tropoelastin, which subsequently matures into an insoluble, cross‐linked network capable of stretching under tension and recoiling elastically (Baumann et al. 2021; Chocarro‐Wrona et al. 2025). Fibroblasts are the primary cell type tasked with the production of elastin, displaying robust expression of elastin genes, while keratinocytes exhibit comparatively lower gene expression (Chowdhury et al. 2021; Plikus et al. 2021). The degradation or structural damage to elastic fibers may lead to disorders characterized by cutaneous laxity (Edgar et al. 2018).

Re‐epithelialization is a pivotal event in skin wound healing, relying on the directed collective migration of keratinocytes to re‐establish epidermal barrier continuity (Friedl and Wolf 2010; Grinnell 2008; Peña and Martin 2024). Unlike isolated single‐cell migration, collective keratinocyte migration is governed by biomechanical cues from the extracellular matrix (ECM) and intercellular adhesion dynamics. Recent studies have revealed that, during collective migration, keratinocytes dynamically assemble suspended multicellular bridges when traversing non‐adherent regions (Jamora and Fuchs 2002; Raja et al. 2007; Vedula et al. 2012). These remarkable epithelial structures emerge through fibronectin‐mediated tensile forces, with adherens junctions playing a critical role in their formation and maintenance. These bridges exhibit elasticity, with keratinocytes displaying greater traction at the leading edge and evenly distributed force away from it. Fibronectin, a major non‐collagenous glycoprotein in the ECM and basement membrane, is widely involved in cell migration, adhesion, proliferation, and tissue repair (Pally and Naba 2024; Saraswathibhatla et al. 2023; Vedula et al. 2014). The collective migration of keratinocytes is closely related to fibronectin, and the process involves changes in the elasticity of skin cells.

Microtubules are essential for cell migration, orchestrating directional movement by transporting membrane components and signaling molecules to the leading edge of migrating cells (Garcin and Straube 2019; Kaverina and Straube 2011; Logan and Menko 2019). Asymmetrically organized in migrating cells, microtubules regulate adhesion dynamics and contractility to enable directional migration (Janke 2014; Morris et al. 2014). Microtubules undergo continuous cycles of growth and shrinkage, a phenomenon termed dynamic instability, with their stabilization mediated by post‐translational modifications and interactions with microtubule‐associated proteins (MAPs) (Mitchison and Kirschner 1984). Acetylated microtubules accumulate at the leading edge of actively migrating cells, promoting directional persistence (Akisaka et al. 2011; Gundersen and Bulinski 1988). This acetylation‐dependent migration mechanism, observed during embryonic skin development, relies on the increased strength and flexibility of acetylated microtubules to facilitate coordinated cell movement (Collins et al. 2021).

Recent studies have identified Enkurin domain‐containing protein 1 (ENKD1) as a novel microtubule‐associated protein that regulates astral microtubule stability during mitosis. This function modulates the division orientation of basal keratinocytes in the skin epidermis, influencing epidermal stratification (Zhong et al. 2022). Recent studies have demonstrated that proper spindle integrity in keratinocytes is essential for normal epidermal morphogenesis (Sun et al. 2025). Beyond its role in microtubule stabilization, emerging evidence demonstrates that ENKD1 also regulates cell migration velocity in non‐small cell lung cancer (NSCLC), highlighting its broader functional significance in cellular dynamics (Song et al. 2022).

2. Results

2.1. ENKD1 Deficiency Results in Reduced Cell Migration Speed and Compromised Skin Elasticity

Given that ENKD1, a microtubule‐binding protein, not only regulates microtubule stability and affects skin thickness and stratification but also influences the migration speed of NSCLC cells, we hypothesized that ENKD1 might also play a role in skin cell migration and skin elasticity.

To investigate this, we generated conditional knockout mice with targeted deletion of enkd1 in K14‐expressing tissues. Examining newborn (P0) mice revealed that heterozygous knockout (Enkd1 ^fl/+; K14‐Cre) mice exhibited increased epidermal wrinkling and rough skin texture compared to wild‐type (WT) mice. More strikingly, homozygous knockout (Enkd1 ^fl/fl; K14‐Cre) mice displayed even more pronounced epidermal wrinkling, greater skin roughness, and significant depressions on the abdomen, back, and lateral sides, along with visibly thinner skin that appeared partially translucent (Figure 1A,B). To assess skin elasticity, we performed skin extensibility assays using dorsal skin samples (1 cm × 1 cm squares) from WT, Enkd1 ^fl/+; K14‐Cre, and Enkd1 ^fl/fl; K14‐Cre mice. Under a stereomicroscope, connective tissue and adipose layers were carefully removed, and the samples were immersed in water to allow complete epidermal expansion. The expanded epidermal area was an indicator of skin elasticity (Figure 1C). Quantitative analysis revealed that both Enkd1 ^fl/+; K14‐Cre and Enkd1 ^fl/fl; K14‐Cre mice exhibited significantly reduced skin expansion compared to WT mice (Figure 1D). During collective migration, keratinocytes form epithelial bridges that exhibit elastic‐like behavior. Given the impaired skin extensibility observed in Enkd1 conditional knockout mice, we hypothesized a potential link between skin elasticity and keratinocyte migration. To test this, we designed and ordered ENKD1‐specific siRNAs to investigate whether ENKD1 regulates cell migration in cultured cells. Western blot analysis confirmed the knockdown efficiency, demonstrating significant reduction of ENKD1 protein levels in siRNA‐treated cells (Figure 1E). Subsequently, scratch wound healing assays were conducted to compare control and ENKD1‐knockdown groups. The wound closure was monitored at various time points, with the remaining wound area measured to evaluate cell migration rates. Control cells exhibited rapid wound closure, achieving nearly complete healing by 96 h. In contrast, ENKD1‐knockdown cells showed markedly delayed wound closure, with significant gaps remaining at the same time point (Figure 1F,G). These results clearly indicate that ENKD1 depletion substantially impairs cell migration. Taken together, our experimental data demonstrate that ENKD1 deficiency leads to both slowed cell migration and impaired skin extensibility, highlighting its critical role in maintaining skin elasticity and cellular dynamics.

FIGURE 1.

ENKD1 deficiency results in decreased cell migration speed and impaired skin elasticity. (A) From three P0 mice, WT, Enkd1 ^fl/+; K14‐Cre, and Enkd1 ^fl/fl; K14‐Cre, were collected and examined from three perspectives: Dorsal, lateral, and ventral regions. Compared to WT, Enkd1 ^fl/+; K14‐Cre exhibited increased epidermal folding and uneven skin texture. Enkd1 ^fl/fl; K14‐Cre mice displayed even more pronounced epidermal folding and further deterioration in skin smoothness. (B) Enlarged image of (A). (C) Skin elasticity was assessed in WT, Enkd1 ^fl/+; K14‐Cre, and Enkd1 ^fl/fl; K14‐Cre mice by measuring the extensibility of equal surface area (1 cm²) skin samples. (D) Quantitative analysis of skin surface area revealed a significant reduction in both Enkd1 ^fl/+; K14‐Cre and Enkd1 ^fl/fl; K14‐Cre mice compared to WT controls. n = 6. (***p < 0.001, ****p < 0.0001). (E) ENKD1 knockdown efficiency in cells was confirmed by Western blot. (F) Wound healing assay was performed in ENKD1‐depleted cells. Scratch wounds were monitored from Days 1 –4. While control wounds healed by Day 4, ENKD1‐deficient cells exhibited delayed wound closure. (G) Quantitative analysis of wound healing was performed using ImageJ. The scratch area was normalized to the field length to calculate the average cell migration distance.

2.2. ENKD1 Regulates Microtubule Stability

Previous studies have demonstrated that ENKD1 functions as a microtubule‐associated protein with centrosomal localization, which is regulated by HDAC6‐mediated deacetylation. Notably, the acetylation status of ENKD1 has been shown to influence corneal epithelial thickness (Song et al. 2025).

Building upon our previous findings that ENKD1 deficiency impairs cell migration and skin extensibility, we sought to elucidate the underlying mechanisms by investigating the role of ENKD1 in mitosis. Using human keratinocytes (HaCaT cells), we systematically analyzed the subcellular localization of ENKD1 throughout mitotic phases: prophase, prometaphase, metaphase, anaphase, and telophase/cytokinesis. HaCaT cells cultured in 24‐well plates were fixed with pre‐chilled methanol (−20°C) and co‐stained with antibodies against ENKD1 and α‐tubulin to examine ENKD1 distribution. ENKD1 exhibits cytoplasmic distribution in interphase, while it displays prominent accumulation at spindle poles in addition to its cytoplasmic presence during mitosis. Co‐staining with γ‐tubulin (a centrosomal marker), we further confirmed that ENKD1 strongly colocalized with γ‐tubulin throughout all mitotic phases (Figure 2A). These observations collectively indicate that ENKD1 primarily localizes to centrosomes during cell division. During cytokinesis, GFP‐tagged ENKD1 was observed at the midbody (Figure 2B). Furthermore, GFP‐ENKD1 overexpression demonstrated prominent accumulation at centrosomes (Figure 2C), strongly suggesting that ENKD1 likely regulates cell cycle processes through centrosomal interactions. These collective findings, centrosomal localization, midbody association, and microtubule binding, indicate that ENKD1 plays a regulatory role in cell division. We further observed that ENKD1 forms granular aggregates along the lateral walls of microtubules (Figure 2D), leading us to hypothesize that ENKD1 may mediate microtubule crosslinking by attaching to microtubule surfaces. Following the precise colocalization of ENKD1 and microtubules, we sought to determine whether ENKD1's microtubule‐interacting domain mediates microtubule stabilization. Functional validation was performed by overexpressing various ENKD1 truncation and deletion mutants in cells, followed by 30‐min ice treatment and immunofluorescence analysis. Our experiments revealed that truncated ENKD1 containing the microtubule‐binding domain (91–171 aa) retained microtubule‐stabilizing activity, whereas deletion mutants lacking this region completely lost their ability to stabilize microtubules (Figure 2E). These results demonstrate that the 91–171 amino acid segment is essential for ENKD1‐mediated microtubule stabilization. Overexpression of ENKD1 led to an increase in acetylated microtubules (Figure 2F). Given that acetylated microtubules represent the most stable microtubule population in cells, we conclude that ENKD1 overexpression enhances microtubule stability, suggesting microtubule acetylation as a potential mechanism underlying ENKD1‐mediated stabilization.

FIGURE 2.

ENKD1 regulates microtubule stability. (A) Subcellular localization of ENKD1 during mitosis in HaCaT cells. Scale bar: 6 μm. (B) Co‐localization of ENKD1 with midbody. Cells overexpressing GFP‐ENKD1 were immunostained, with microtubules in red and GFP‐ENKD1 in green. (C) ENKD1 accumulation at centrosomes. γ‐Tubulin staining (centrosomal marker) shows co‐localization with GFP‐ENKD1 in overexpression cells. (D) ENKD1 localizes along microtubule walls. Endogenous ENKD1 co‐staining with α‐tubulin in U2OS (left) and HeLa (right) cells, with insets showing magnified views of boxed regions. (E) Microtubule stabilization assay. Cells expressing GFP‐tagged ENKD1 truncation/deletion mutants were ice‐treated for 30 min before α‐tubulin staining (red). Green indicates GFP‐tagged constructs. (F) Increased microtubule acetylation upon ENKD1 overexpression. GFP‐ENKD1‐expressing cells (green) show enhanced acetylated microtubule signal (red).

2.3. ENKD1 Disrupts Cellular Localization of EB1 and NuMA

The regulation of spindle positioning is a crucial determinant for cell fate. During cell division, astral microtubules interact with cortical complexes, such as the evolutionarily conserved Gαi/LGN/NuMA complex in vertebrates, to orient the mitotic spindle (di Pietro et al. 2016; Yang et al. 2025; Zhong et al. 2025; Zhu et al. 2011). These microtubules anchor the spindle to the cell cortex, thereby governing spindle orientation. End‐binding protein 1 (EB1), a key regulator of microtubule dynamics, controls processes at growing microtubule plus‐ends and is essential for proper cell division. Studies have demonstrated that EB1 influences spindle orientation by modulating both astral microtubule stability and interactions with cortical complexes (Chen et al. 2018; Luo et al. 2016). Notably, EB1 phosphorylation plays a pivotal role in this regulatory mechanism (Chen et al. 2019). Our study revealed that ENKD1 overexpression significantly impairs EB1 accumulation at microtubule plus‐ends. Through co‐immunofluorescence staining in both HeLa and RPE1 cells, we observed a marked reduction in both the length and frequency of EB1‐decorated microtubule plus‐ends (Figure 3A–C). ENKD1 overexpression leads to the reduction of EB1 (Figure 3D). We further investigated the effect of ENKD1 on the localization of the key cortical protein NuMA. NuMA typically localizes to both spindle poles and their corresponding cortical regions during metaphase. Analysis of NuMA expression and localization in control and ENKD1‐knockdown HaCaT cells showed that ENKD1 depletion did not affect NuMA protein levels (Figure 3E). However, immunofluorescence revealed that ENKD1 knockdown significantly altered NuMA localization, impairing its proper cortical targeting (Figure 3F). We categorized NuMA localization patterns into three distinct phenotypes: (1) polar cortical localization at spindle pole‐facing cortex, (2) complete absence of cortical localization, and (3) pan‐cortical localization throughout the entire cell cortex (Figure 3G). Quantitative analysis demonstrated dramatic changes in these patterns following ENKD1 knockdown: cells with polar cortical localization decreased from 85% in controls to 44% (siENKD1#1) and 42% (siENKD1#2); those lacking cortical localization increased from 6% to 37% and 64%; while pan‐cortical localization remained relatively constant (16% control vs. 19% and 15% in knockdowns) (Figure 3H).

FIGURE 3.

ENKD1 disrupts the cellular localization of EB1 and NuMA. (A) ENKD1 overexpression reduces EB1 accumulation at microtubule plus‐ends across cell lines. EB1 (red); GFP‐ENKD1 (green). (B) Quantification of EB1 distribution patterns in HeLa. The fluorescence intensity of EB1 is quantified by the EB1 fluorescence signal in the control cells and GFP‐ENKD1‐overexpressing cells. (C) Quantification of EB1 distribution patterns in RPE‐1. The fluorescence intensity of EB1 is quantified by the EB1 fluorescence signal in the control cells and GFP‐ENKD1‐overexpressing cells. (D) ENKD1 overexpression affects EB1 protein level (Western blot). (E) ENKD1 knockdown does not affect NuMA protein levels (Western blot). (F) Abnormal cortical localization of NuMA (green) upon ENKD1 depletion, with microtubules (α‐tubulin, red). Scale bar: 6 μm. (G) Three phenotypes: Polar cortical localization (spindle pole‐associated), no cortical localization, and pan‐cortical distribution. (H) Statistical analysis of NuMA localization patterns in HaCaT cells after ENKD1 knockdown (n = 30 cells per group) (**p < 0.01, ***p < 0.001).

3. Discussion

Current research indicates that classical MAPs, including Tau, MAP2, and MAP1B, can negatively regulate EB1 localization at microtubule plus‐ends, albeit through distinct mechanisms. These proteins modulate microtubule dynamics by altering EB1 distribution patterns at growing microtubule ends (Nehlig et al. 2017; Ramirez‐Rios et al. 2016; Tortosa et al. 2013). Tau protein plays a crucial role in regulating microtubule dynamics by promoting both polymerization and stabilization of microtubules. Notably, elevated expression levels of tau induce microtubule bundling, which subsequently leads to the redistribution of EB1 along these bundled microtubule structures (Choi et al. 2009). MAP1B directly interacts with EB1 and modulates its binding to microtubules. Specifically, MAP1B overexpression significantly attenuates EB1 binding affinity to microtubule plus‐ends, whereas MAP1B knockdown conversely enhances EB1‐microtubule association (Tymanskyj et al. 2012). These observations collectively indicate that classical microtubule‐associated proteins (MAPs) function as master regulators that not only maintain microtubule stability but also precisely modulate microtubule dynamics. Through their distinct molecular mechanisms, these MAPs exert spatiotemporal control over EB1 localization patterns, thereby orchestrating the dynamic behavior of microtubule networks in a sophisticated manner. Our findings suggest ENKD1's effect on EB1 represents another potential mechanism through which it regulates microtubule dynamics. Although the precise molecular mechanism by which ENKD1 influences EB1 distribution remains elusive, its demonstrated impact on microtubule dynamics positions ENKD1 as both a potential prognostic marker for cancer progression and a promising therapeutic target in cancer treatment strategies.

4. Materials and Methods

4.1. Cell Culture and Transfection

HaCaT, HeLa, RPE‐1, U2OS, and HEK293T cells were obtained from the American Type Culture Collection. Cells were cultured in DMEM (Thermo Fisher Scientific, Waltham, MA, USA) supplemented with 10% fetal bovine serum. Plasmids were transfected into HEK293T cells using polyethyleneimine (Polysciences) for immunoblotting and transfected into HaCaT, HeLa, and RPE‐1 cells using Lipofectamine 3000 (Thermo Fisher Scientific) for immunostaining. siRNAs were transfected using Lipofectamine RNAiMAX (Thermo Fisher Scientific). All cell lines were assessed regularly for mycoplasma contamination using a MycoAlert PLUS Mycoplasma Detection Kit (Lonza, Basel, Switzerland). A mammalian plasmid expressing GFP‐ENKD1 was constructed by inserting the Enkd1 cDNA into the pEGFP‐C1 vectors. Mammalian plasmids expressing GFP‐ENKD1 truncations were constructed by PCR with pEGFP‐C1‐ENKD1 as a template. Human ENKD1 siRNAs (#1: 5′‐GGCCCAAAGUCUUCGUGAA‐3′; #2: 5′‐GUGGACUUCAUUCGUCACA‐3′) were synthesized by RiboBio (Guangzhou, China).

4.2. Animal Experiments

Enkd1 ^fl/fl mice were generated via the CRISPR/Cas9 system by creating loxP sites flanking exon 3 of the Enkd1 gene. Enkd1 ^fl/fl mice were crossed with K14‐Cre mice for the generation of Enkd1 conditional knockout mice. Experiments were performed according to the guidelines of the Animal Care and Use Committee of Nankai University.

4.3. Fluorescence Microscopy

Cells were fixed with ice methanol, permeabilized with 0.5% Triton X‐100, and blocked with 4% bovine serum albumin (BSA). Then, cells were sequentially probed with the primary antibodies, fluorophore‐conjugated secondary antibodies, and 4′,6‐diamidino‐2‐phenylindole (DAPI). DAPI was obtained from Sigma‐Aldrich. Antibodies targeting ENKD1 (#ab224560), α‐tubulin (#ab7291), NuMA (#ab97585), and Alexa Fluor‐conjugated secondary antibodies (#ab150081, #ab175694, #ab150117, #ab175699) were purchased from Abcam (Cambridge, UK). Antibodies targeting acetylated α‐tubulin (#T7451), γ‐tubulin (#T6557) were purchased from Sigma‐Aldrich (St Louis, MO, USA). Antibodies targeting EB1 (#610534, BD Bioscience, San Jose, CA, USA) were purchased from the indicated sources.

4.4. Immunoblotting

Cells were lysed using cell lysis buffer (Beyotime). Proteins were loaded into SDS‐polyacrylamide gel, resolved by electrophoresis, and transferred onto polyvinylidene difluoride membranes (Millipore, Burlington, MA, USA). The membranes were blocked in 5% fat‐free milk and incubated with primary antibodies and horseradish peroxidase‐conjugated secondary antibodies (#sc2004, #sc2005, Santa Cruz, CA, USA). Targeted proteins were visualized with an enhanced chemiluminescence substrate kit (Millipore).

4.5. Quantification and Statistical Analysis

All experiments were biologically repeated at least three times. The relative intensity of proteins on immunostaining images was quantified using the ImageJ software. Quantitative analysis of wound healing was performed using ImageJ. All statistical data were analyzed with GraphPad Prism 8 and presented as mean ± SEM. Unpaired two‐tailed Student's t‐tests were used for comparison between the two groups. Differences were considered significant when the p value was less than 0.05.

Author Contributions

Dengwen Li, Tao Zhong, and Dan Dong designed the experiment and wrote the manuscript. Tao Zhong, Xiaofan Wu, Mingzheng Hu, Ruming Liu, Ying Shan, and Dan Dong performed experiments and analyzed data.

Conflicts of Interest

The authors declare no conflicts of interest.

Dong, D. , Hu M., Wu X., et al. 2026. “ ENKD1 Modulates Skin Elasticity Through Microtubule Stability Regulation.” Cytoskeleton 83, no. 8: 539–546. 10.1002/cm.70016.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.