Cytokeratin-1 is essential for the detection of laminar shear stress in endothelial cells

ABSTRACT

Endothelial cells regulate diverse vascular functions by perceiving and reacting to laminar shear stress. In this study, a novel shear-sensing receptor was identified through the use of a pro-inflammatory protein, lysyl-tRNA synthetase (KARS), which is known to be secreted from endothelial cells via autophagy. Binding assays demonstrated that cytokeratin-1 (CK1) interacts with KARS at the endothelial cell surface. Additionally, CK1 was shown to be critical for ECM-cell adhesion and endothelial sensing of shear stress by mediating interactions with laminin and integrin α6. Overexpression of CK1 results in hyperactivation of endothelial nitric oxide synthase (eNOS) in response to laminar shear stress (LSS), potentially reducing the risk of atherosclerosis. Furthermore, elevated CK1 expression significantly decreases leukocyte adhesion to endothelial cells by modulating nitric oxide production stimulated by LSS. Conversely, CK1 knockdown leads to the formation of actin fibers and diminishes LSS-induced activation of several cell signaling components. These findings indicate that CK1 is a shear-sensing receptor, providing new perspectives on the close relationship between cell-to-matrix adhesion and mechanosensing.

Introduction

Hemodynamic shear stress regulates multiple endothelial functions, including vascular permeability, cell proliferation, and apoptosis Koskinas et al. (2009). Shear stress levels fluctuate at vessel branches due to significant alterations in blood flow in these regions. Atherosclerotic plaques predominantly develop in areas subjected to low shear stress and disturbed flow, indicating that shear stress serves as an anti-atherogenic factor (Davies 1995; Malek et al. 1999). Endothelial cells detect shear stress within blood vessels, adjusting their structural and functional properties to preserve vascular homeostasis. However, under conditions of low shear stress or disturbed flow, endothelial cells initiate pro-inflammatory, pro-oxidant, and vasoconstrictive responses. These vascular responses are governed by cell signaling pathways that are selectively activated by either elevated shear stress or irregular blood flows.

Shear stress stimulates endothelial cells to activate diverse signaling pathways, including Akt, ERK, and eNOS Lehoux et al. (2006), through recognized mechanosensors such as integrins Li et al. (2005), G-protein coupled receptors Jo et al. (1997), and platelet endothelial cell adhesion molecule-1 (PECAM-1) Li et al. (2005). Nitric oxide (NO) is a key vaso-regulator in vascular biology. Laminar shear stress (LSS) enhances NO production Cines et al. (1998). Shear-dependent NO synthesis has been extensively characterized, given that NO inhibits thrombosis, monocytic adhesion to the endothelium, and endothelial cell apoptosis Fleming and Busse (1999).

Lysyl tRNA synthetase (KARS) is an aminoacyl tRNA synthetase (ARS) that catalyzes the aminoacylation of its cognate tRNA with ATP hydrolysis Park et al. (2010). KARS is secreted into the extracellular environment by tumor necrosis factor (TNF)-α in HCT116 colon cancer cells, which leads to immune cell activation Park et al. (2005). Recent findings show that KARS can be secreted via exosomes after dissociating from the multi-synthetase complex and binding to syntenin in cancer cells (Kim et al. 2017, 2018). KARS promotes migration of immune cells and acts as a pro-inflammatory mediator (Park et al. 2005; Kim et al. 2017).

Our previous research demonstrated that KARS inhibits laminar shear stress (LSS)-induced cell signaling, including ERK activation Scientific Reports Yun et al. (2025) and that extracellular KARS disrupts vascular functions, such as shear stress-stimulated NO production Scientific Reports Yun et al. (2025). In this study, a pull-down binding assay using the shear-inhibitory protein KARS identified cytokeratin 1 (CK1) as a previously unrecognized shear-sensing receptor involved in modulating endothelial responses to shear stress.

Materials and methods

Cell culture

Bovine aortic endothelial cells (BAEC) derived from descending thoracic aortas were cultured at 37°C and 5% CO₂ in Dulbecco's Modified Eagle's Medium (1 g/liter glucose, Life Technologies, Inc.) supplemented with 20% fetal bovine serum (Wel GENE, Inc) and antibiotics. Cells between passages 3 and 10 were utilized.

Mouse aortic endothelial cells (MAECs) were isolated and maintained as previously described Kim et al. 2010. In summary, cells were kept at 37°C in a humidified atmosphere of 5% CO₂, using DMEM (Gibco, Grand Island, NY, USA) supplemented with 10% FBS, 1% bovine brain extract, 1% MEM non-essential amino acid solution (Gibco), and antibiotics (penicillin/streptomycin).

Oscillatory shear stress

BAECs or MAECs grown to 90% confluence in 10-cm culture dishes were subjected to OSS in the growth medium using a rotating Teflon cone (0.5° cone angle), as previously described Sorescu et al. (2003). Briefly, once BAECs or MAECs reached the appropriate confluence, the culture dishes were mounted in a rotating cone apparatus designed to induce directional flow changes at 1-Hz s (±5 dyn/cm²). OSS was produced by alternately rotating the cone with a stepping servo motor controlled by a computer program (DC Motor Company, Atlanta, GA, USA). Following the setup, the cells were subsequently exposed to OSS for designated durations.

Reverse transcription-polymerase chain reaction (RT–PCR): Total RNA was isolated using the QIAzol lysis reagent according to the manufacturer's protocol (QIAGEN), and 5 μg of purified RNA was reversely transcribed to generate cDNA as described previously Go et al. (2012). After cDNA synthesis, PCR amplification was performed under the following conditions: denaturation at 95°C for 30 s, annealing at 60°C for 30 s, and extension at 72°C for 30 s. The following specific primers were used for PCR amplification:

KARS, 5´ – ATT GAC TTC ACC CCG CCC TTC-3´ (forward)

  5´ – CAA GGA CCT TCA CTG CAC GTA-3´ (reverse)

GAPDH, 5´-CCAACGTGTCTGTTGTGGATCTGA-3´ (forward)

     5´-CAACCTGGTCCTCAGTGTAGCCTA-3´ (reverse)

GAPDH (glyceraldehyde-3-phosphate dehydrogenase) mRNA served as the internal control.

NO quantification

Cellular NO content was determined by measuring the fluorescence spectra of intracellular DAF-2 (Calbiochem). Cells were pre-incubated for 20 min in a HEPES buffer (5 mmol/L HEPES, 140 mmol/L NaCl, 5 mmol/L KCl, 2 mmol/L CaCl₂, 1 mmol/L MgCl₂, 5 mmol/L Glucose, pH 7.4) supplemented with 1 μmol/L Ca²⁺ ionophore A23187 (Sigma). Following this pre-incubation, cells were further incubated at 37°C in 5% CO₂ with 0.1 µmol/L DAF-2 for 15 min, harvested, and lysed by sonication. The supernatants were collected by centrifugal fractionation and appropriately diluted before being analyzed with a spectrofluorophotometer (RF 5301PC Shimadzu) at excitation and emission wavelengths of 495 and 515 nm (slit: 10 nm), respectively. NO concentrations were determined by assessing the DAF-2 fluorescence intensity Kim et al. (2008).

KARS expression and purification

KARS cDNAs were cloned into pGEX4T-1 (Amersham) or pET-28a (Novagen) vectors and subsequently transformed into E. coli (BL21-DE3). The recombinant cells harboring pGEX4T-1, pGEX4T-1-KARS, or pET-28a-KARS plasmids were cultured in Luria–Bertani (LB) medium and induced with 0.1 mM isopropyl β-D-thiogalactopyranoside (IPTG, Sigma Aldrich) for 8 h at 25°C. After induction, cells were collected by centrifugation, and the resulting pellets were resuspended in lysis buffer (20 mM KH₂PO₄, 500 mM NaCl, 2 mM β-mercaptoethanol) containing lysozyme (100 mg/ml). Cells were then subjected to sonication and centrifugation for lysis. The resulting soluble supernatants were applied to affinity chromatography using glutathione agarose beads (Phamacia Biotech) or Ni²⁺-NTA agarose beads (Qiagen) to purify the proteins.

In vitro KARS binding assay

Confluent BAECs were treated with 5 μg/ml chymotrypsin (Sigma Aldrich). Supernatants containing fragments of cell surface proteins were harvested after centrifugation, concentrated using Centricon devices (Millipore), and incubated with GST beads or GST-KARS beads at 4°C overnight. Bead-bound proteins were separated by SDS-PAGE, and bands specifically associated with KARS were identified by Q-TOF (in2gen).

Immunofluorescence

BAECs were fixed in 4% paraformaldehyde for 15 min, permeabilized in 0.5% Tween 20 for 10 min, and incubated in PBS containing 2% BSA and 0.1% Tween 20 for 1 h to block nonspecific binding. Cells were then incubated overnight at 4°C with anti-integrin α5 (Millipore), anti-integrin α6 (Abcam), anti-CK1 (Santa Cruz), or anti-His tag (Cell Signaling) antibodies, followed by incubation with TRITC-conjugated goat anti-rabbit (Zymed), FITC conjugated goat anti-mouse (Zymed), or FITC-conjugated Phalloidin (Sigma Aldrich) for 1 h. After mounting with Slow-Fade mounting medium, cells were visualized using a fluorescence microscope (Zeiss Autoplan2) or a confocal laser scanning microscope (Carl Zeiss-LSM 710). For the KARS-cell attachment assay, BAECs were pretreated with 500 nM His-KARS for 1 h at 37°C in 5% CO₂ and subsequently immunostained.

Transfection of plasmid

CK1 cDNAs were inserted into pEF6/V5-His TOPO plasmids (Invitrogen Life Technologies). BAECs grown to 90% confluency were transfected with 2 μg of plasmids encoding either empty vector, central domain, tail, Δtail, or full-length CK1 (wild-type) genes in the presence of 5 μl lipofectamine 2000 Transfection Reagent (Invitrogen Life Technologies), following the manufacturer's protocol. For Caco-2 cells, transfection was conducted using 2 μg of plasmids and 8 ml of FuGENE HD Transfection Reagent (Roche Applied Science) in accordance with the supplier's instructions.

Transfection of siRNA

BAECs cultured to 50% confluence were transfected with duplex oligonucleotide Stealth siRNA (300 pmol/ml, Invitrogen Life Technologies), using either control siRNA or siRNA targeting the CK1 gene (5′-CCACUUAUUCCGGAGUAACCAGAU A-3′), in the presence of Lipofectamine RNAi MAX Transfection Reagent (Invitrogen Life Technologies) following the manufacturer's instructions.

Preparation of cell lysates

Cells were washed with ice-cold PBS, then lysed for 30 min at 4 °C in a buffer (50 mM HEPES, pH 7.4, 150 mM NaCl, 1 mM vanadate, 1% Triton X-100, 10% glycerol, and 0.1 mM phenylmethylsulfonyl fluoride) containing a protease inhibitor cocktail. Lysates were centrifuged for 15 min at 4°C to separate the soluble supernatant from the insoluble pellet. Protein concentrations were quantified using a Micro BCA protein assay kit (Thermo Scientific).

Western blotting

Cell lysates were subjected to SDS-PAGE and transferred onto polyvinylidene difluoride (PVDF) membranes (Millipore) as described previously Lee et al. (2024). Membranes were blocked with PBS containing 5% BSA and 0.1% Tween 20 for 1 h, followed by overnight incubation at 4°C with antibodies specific for V5 (Invitrogen Life Technologies), actin, CK1, integrin b1 (Santa Cruz), peNOS, eNOS, pPI₃K, pAkt, Akt, pJNK, JNK, pERK, ERK, pp38, p38, caspase-3 (Cell Signaling), KARS, integrin α6, laminin (Abcam), integrin α5 (Millipore), or fibronectin (BD Bioscience). Afterwards, membranes were incubated with HRP-conjugated secondary antibodies and visualized using the enhanced chemiluminescence detection method (Amersham).

Endothelial cell adhesion

96-well plates were either left uncoated or coated overnight at 4°C with fibronectin (1 μg/well, Chemicon International) or laminin (2 μg/well, Sigma Aldrich), and then blocked for 1 h with PBS containing 1% BSA. Transfected cells were seeded into uncoated or pre-coated wells (1 × 10⁶ cells/well) and incubated for 1 h at 37°C in a CO₂ incubator. For Cytochalasin D experiments, transfected cells were pre-treated with 5 μM Cytochalasin D for 1 h, then plated onto ECM-coated wells with 5 μM Cytochalasin D. Non-adherent cells were gently removed by PBS washing, and adherent cells were stained with 0.2% crystal violet in PBS containing 10% ethanol for 7 min at room temperature. After additional PBS washes, stained cells were treated for 15 min with solubilization buffer (0.5 M NaH₂PO₄, pH 4.5, and 50% ethanol). Adhesion was assessed by measuring absorbance at 570 nm using a microplate reader (Bio-Rad, Model 550).

Immunoprecipitation

BAECs were rinsed with ice-cold PBS and then lysed in RIPA buffer (50 mM Tris, pH 7.4, 150 mM NaCl, 0.5% deoxycholate, 1% NP-40, 0.1% SDS, 0.1 mM phenylmethylsulfonyl fluoride, and a protease inhibitor cocktail). Protein concentrations in cell lysates were measured using a Micro BCA protein assay kit (Thermo Scientific). Lysates (700 μg) were incubated with 2.5 μg of either goat IgG or goat anti-CK1 antibody overnight at 4°C, followed by immunoprecipitation with 50 μl of Protein A/G PLUS Agarose beads (Santa Cruz) for 2 h at 4°C. Immunocomplexed beads were subsequently washed multiple times with RIPA buffer and subjected to western blot analysis.

CK1 expression and purification

CK1 cDNA fragments were cloned into pET-32a (Novagen) vectors and transformed into E. coli (BL21-DE3) cells. Cells harboring pET-32a or pET-32a-CK1 constructs were cultured in LB medium and induced with 0.5 mM IPTG for 4 h at 25°C to express proteins. After harvesting in PBS containing lysozyme (100 μg/ml), 1 mM phenylmethylsulfonyl fluoride, and 1.5% sarkosyl, cells were disrupted by sonication. The resulting soluble lysates were incubated with Ni²⁺-NTA agarose beads overnight at 4°C. Proteins immobilized on His-Trx tag beads and His-Trx-CK1 beads were identified via western blot.

In vitro CK1 binding assay

Confluent BAECs were lysed in RIPA buffer. Cell lysates (700 μg) were incubated overnight at 4°C with Trx beads or Trx-CK1 beads. After incubation, the beads with bound proteins were washed in RIPA buffer, subjected to SDS-PAGE, and analyzed by western blotting.

Laminar shear stress (LSS) studies

BAECs or MAECs were plated on glass slides, transfected, and serum-starved for 8 h. The slides with the confluent transfected cells were assembled in a parallel plate shear chamber. A flow channel (220-μm height x 25-mm width x 62-mm length) was created between the cells and a polycarbonate plate. Laminar shear stress was generated using a constant-head flow-loop by regulating the starvation medium flow rate as described previously Go et al. (1998). For antibody treatments, BAECs were pre-incubated for 1 h at the indicated concentrations with anti-CK antibodies (C1801, C2562, and C2931, Sigma Aldrich) or IgG before exposure to laminar shear.

Adhesion of THP-1

Transfected BAECs were starved for 12 h with or without 1 mM L-NAME (BIOMOL) and then incubated with LPS (1 μg/ml, Sigma Aldrich) for 6 h. Subsequently, some cells were maintained under static conditions and others were subjected to laminar shear stress for 5 min. THP-1 cells were stained with 10 mM Calcein AM (Sigma Aldrich) for 45 min at 37°C with 5% CO₂, followed by PBS washes. Labeled THP-1 cells (5 × 10⁶ cells/ml) were subsequently added to BAECs. After 1 h of incubation, non-adherent THP-1 cells were removed by washing with PBS, and adherent THP-1 cells were visualized using a fluorescence microscope (Zeiss Autoplan2) and quantified by counting.

Statistical analysis

Differences in data between the control and experimental samples were analyzed using a Student's t-test. Additionally, comparisons among multiple sample groups were performed using one way ANOVA.

Results

CK1 is a key surface molecule that interacts with the shear-inhibitory protein, KARS

To identify the binding partners of KARS, cellular surface protein fragments were prepared by treating intact BAECs with chymotrypsin and then separated by chromatography using GST beads or GST-KARS beads. As shown in Figure 1(A), two distinct bands were detected in the KARS-binding fraction by SDS-PAGE. Subsequent Q-TOF analysis revealed that the band fragments matched sequences from fibronectin-1 or CK1. Since fibronectin proteins have previously been established as important mediators of cellular responses to hemodynamic forces Urbich et al. (2002), these findings further validate the Q-TOF results. However, little was known about the function of CK1 in shear sensing and related cellular responses. As depicted in Figure 1(B), CK1 is composed of three domains: head, central, and tail. Q-TOF identified the binding fragments as being located in the central domain. To confirm the KARS-CK1 interaction, intact BAECs were incubated with His-tagged KARS proteins, followed by thorough washing to eliminate unbound KARS; subsequently, bound KARS and extracellular CK1 were visualized using antibodies specific to the His-tag and CK1, respectively. This double staining demonstrated the colocalization of His-tagged KARS and CK1 on the cell surface (Figure 1(C)). These results indicate that KARS binds to CK1 on the BAEC surface through the central domain of CK1.

Figure 1.

KARS interacts with CK1 at the endothelial cell surface. (A) BAECs were incubated with chymotrypsin. Subsequently, the supernatant containing surface protein fragments was collected and subjected to incubation with Bead-GST or Bead-GST-KARS. KARS-interacting proteins were identified by Q-TOF. (B) Alignment of amino acid sequences (red) between the CK1 fragment binding KARS and the central CK1 domain listed in the database. (C) BAECs were incubated with His-KARS and immunostained with anti-His tag (KARS) and anti-CK1 antibodies.

CK1 knockdown induces the formation of actin stress fibers

We subsequently examined whether altering CK1 expression by knockdown or overexpression impacts cellular morphology and function. CK1 knockdown led to a morphological change, producing elongated cell shapes, whereas CK1 overexpression had no observable effect (Supplementary Figure S1). To examine if these CK1 knockdown-induced morphological changes correlated with actin polymerization, CK1 silenced cells were stained using FITC conjugated Phalloidin. As presented in Figure 2(A, B), actin stress fibers were substantially formed in CK1 knockdown cells. To clarify whether actin fiber formation is responsible for the observed morphological changes, BAECs were analyzed in the presence or absence of cytochalasin D, an inhibitor of actin polymerization. As anticipated, cytochalasin D completely inhibited the formation of actin fibers induced by CK1 knockdown (Figure 2(B)). Collectively, these data suggest that CK1 knockdown causes cell elongation through actin fiber formation.

Figure 2.

CK1 knockdown increases actin stress fiber formation colocalized with integrin α6. (A) BAECs were transfected with scrambled control RNA or ck1 siRNA and subjected to immunoblotting with anti-CK1 and anti-actin antibodies. (B) Transfected BAECs were treated with or without cytochalasin D and stained using FITC-conjugated phalloidin. (C-D) BAECs were transfected with scrambled RNA control or ck1 siRNA, followed by immunostaining with anti-actin and anti-integrin α5 (C), or anti-integrin α6 antibodies (D). Stained cells were visualized using a confocal laser scanning microscope.

CK1 promotes endothelial cell adhesion to laminin through its head and central domains

Formation of actin fiber is recognized to be closely associated with ECM-cell adhesion (Hynes 1992; Kim et al. 2024). Two principal types of ECM-cell adhesions are focal adhesion and hemidesmosome, both mediated by integrin α5 and integrin α6, respectively Hynes (1992). Building on the observation that CK1 knockdown leads to actin fiber formation, it was proposed that CK1 knockdown influences ECM-cell adhesion. To examine this hypothesis, a study was initially conducted to evaluate whether CK1 knockdown alters the subcellular distribution of integrin α5, integrin α6, and actin. In quiescent cells, actin was observed to colocalize with both integrin α5 and integrin α6 (Figure 2(C, D)). Conversely, in cells where CK1 was knocked down, actin fibers exhibited colocalization with integrin α6 but were not associated with integrin α5 (Figure 2(C, D)). These observations suggest that endogenous CK1 levels play a critical role in facilitating actin integration into the integrin α5 complex, while actin association with integrin α6 remains unaffected by CK1 knockdown or actin fiber formation.

The subsequent analysis focused on whether CK1 overexpression or knockdown modulates cell adhesion. As illustrated in Figure 3(A), CK1 knockdown did not affect cell adhesion, whereas ectopic CK1 expression resulted in increased cell adhesion. To characterize CK1 functional regions, three specific domains (head, central, and tail) were generated using genetic approaches. Since the head domain was minimally expressed in BAEC, a Δtail construct containing both head and central domains was developed. Furthermore, it was determined that central, Δtail, and full-length CK1 are subject to endogenous cleavage within the central domain, although the mechanism underlying this cleavage remains unclear (Figure 3(B)). As depicted in Figure 3(C), cell adhesion was enhanced in cells overexpressing the central, Δtail, and full-length constructs. These findings demonstrate that CK1 facilitates cell adhesion largely through the head/central domain. Since the adherent effect of the Δtail was more pronounced than that of the central domain alone, this implies the head domain may possess functional significance. To identify the ECM molecule interacting with the head/central domain, adhesion assays were conducted using plates coated with fibronectin or laminin. Although fibronectin coating did not impact cell adhesion, the expression of central, Δtail, and full-length variants significantly increased adhesion to laminin-coated plates. This evidence supports that CK1 is a crucial factor for endothelial cell adhesion to laminin through its head/central domain.

Figure 3.

CK1 is essential for cell adhesion to laminin-coated plates. (A) Cell adhesion was evaluated in BAECs transfected with either a control vector or a vector encoding the ck1 gene (left panel), as well as in cells transfected with a scrambled RNA control or ck1 siRNA (right panel) according to the protocols detailed in Materials and Methods. Adhesive cells were stained with crystal violet and quantified by measuring absorbance at 570 nm (mean ± S.E., n = 5). *P < 0.05. (B) BAECs were transfected with ck1 and its deletion mutant DNAs. Diagrams of the generated mutants are presented in the right panel. Expression of CK1 and its mutants was confirmed by immunoblotting with anti-V5 antibodies. (C) For adhesion assays, cells overexpressing CK1 or its mutants were assessed on uncoated, fibronectin-coated, or laminin-coated plates (mean ± S.E., n = 5). *P < 0.05, **P < 0.03. (D) BAECs were transfected with either control or CK1 siRNA and subsequently treated with or without cytochalasin D. Cell adhesion was then evaluated as described above (mean ± S.E., n = 5). *P < 0.05.

It was observed that cytochalasin D treatment decreased cell adhesion of CK1 knocked-down cells by 15% on uncoated or fibronectin-coated plates, and by 25% on laminin-coated plates, while CK1 knockdown alone did not alter adhesion. This result indicates that maintenance of cell adhesion to laminin following CK1 knockdown is dependent on the formation of actin fibers.

To elucidate the molecular mechanisms underlying CK1 involvement in cell adhesion, in vitro binding assays were conducted using recombinant proteins (His-Trx and His-Trx-CK1). After pull-down of His-Trx or His-Trx-CK1 with BAEC lysates, immunoblotting was performed using anti-fibronectin or anti-laminin antibodies. As illustrated in Supplementary Figure S2A, CK1 specifically interacted with laminin but not with fibronectin. Further in vitro assays revealed an association between the laminin receptor (integrin α6β1) and CK1, while integrin α5 did not exhibit direct binding with CK1 (Supplementary Figure S2B). Given that integrin β1 (the fibronectin receptor) frequently forms heterodimers with integrin α5 Bordeleau et al. (2012), these findings indicate that CK1 preferentially associates with integrin α6 over integrin β1. Moreover, the colocalization of integrin α6 and CK1 was validated by immunostaining (Supplementary Figure S2C). Collectively, these data indicate that CK1 facilitates endothelial cell adhesion to laminin through its interaction with integrin α6.

CK1 is essential for the shear stress response in endothelial cells through its head/central domain

LSS regulates cell proliferation, migration, and diverse cell signaling pathways through mechanosensors located in integrin-associated adhesions Li et al. (2005). Given data indicating that CK1 plays a significant role in integrin-associated adhesions, it is highly likely that CK1 also modulates cell signaling in response to shear stress. Supporting this, inhibition of CK1 by specific antibodies was shown to abolish shear stress-induced activation of ERK and Akt (Figure 4(A)). The effects of CK1 overexpression or knockdown on signaling triggered by shear were evaluated. As illustrated in Figure 4(B, C), CK1 overexpression elevated the phosphorylation of signaling molecules, including ERK, Akt, and eNOS, in response to shear stress, while CK1 knockdown reduced the phosphorylation of ERK, Akt, and eNOS. These findings collectively suggest that CK1 is essential for shear stress-mediated signal transduction. Additional experiments employing CK1 deletion mutants demonstrated that overexpression of the central and Δtail domains enhanced LSS-induced phosphorylation of ERK, Akt, and eNOS, whereas overexpression of the tail domain alone had no influence (Figure 4(D)). These results suggest that CK1 mediates the LSS response through its head and central domains. To further substantiate CK1's importance for shear sensing across cell types, cell signaling was analyzed in various cells exhibiting differential CK1 expression. Notably, ERK activation by shear stress was absent exclusively in Caco-2 cells, which express low levels of CK1 (Figure 4(F) bottom panel). Experiments were then conducted to assess whether shear-induced signaling in Caco-2 cells could be modified by transfecting them with ck1 full-length and various deletion mutant constructs. As shown in the top panel of Figure 4(F), both full-length CK1 and its mutants were robustly expressed. Additionally, partial proteolytic cleavage of CK1 and its mutants in the central domain was observed in Caco-2 cells, as previously seen in BAECs (Figure 3(B)). In Caco-2 cells overexpressing central, Δtail, and full-length CK1, ERK activation in response to shear stress occurred, whereas Akt was not significantly activated (Figure 4(F)). Collectively, these results indicate that CK1 functions as a shear-sensing receptor, transducing shear stress signals primarily via its head and central domains.

Figure 4.

CK1 functions as a putative shear-sensing receptor. (A) BAECs were incubated with various concentrations of antibodies targeting members of the CK family and subjected to shear stress. Western blot analyses were conducted using anti-pAkt, anti-Akt, anti-pERK, and anti-ERK antibodies. (B-D) BAECs were transfected with ck1 siRNA (B), ck1 DNAs (C), or ck1 and its deletion mutant DNAs (D). Following exposure to shear stress, cell lysates were analyzed by immunoblotting with specific antibodies. (E) Cell lysates from several cell lines were subjected to western blotting using anti-CK1 or anti-actin antibodies (top panel). A subset of these cells underwent shear stress before being analyzed with anti-pERK or anti-ERK antibodies (bottom panel). (F) Caco-2 cells transfected with ck1 and its deletion mutant DNAs encoding V5 tagged proteins (top panel) were exposed to shear stress, and subsequently immunoblotted with anti-pAkt, anti-Akt, anti-pERK, or anti-ERK antibodies (bottom panel).

CK1 inhibits monocyte adhesion to endothelial cells through the induction of NO production

In the vascular system, LSS is widely recognized as an anti-atherogenic factor (Davies 1995; Malek et al. 1999). Given that CK1 functions as a shear-sensingk9 receptor, it was proposed that CK1 contributes to preventing the development of atherosclerosis. Previous reports have demonstrated that NO inhibits monocyte adhesion to endothelial cells, which represents an initial step in atherogenesis Tsao et al. (1996). Based on this, an experiment was conducted to evaluate whether CK1, which is necessary for shear-induced eNOS activation, influences monocyte binding to endothelial cells. Notably, CK1 knockdown inhibited LPS-induced THP-1 cell adhesion under LSS exposure but did not affect adhesion under static conditions (Figure 5). The addition of L-NAME appeared to reverse the inhibitory effects of CK1 overexpression on THP-1 cell adhesion, supporting the hypothesis that NO production mediated by CK1 is critical for suppressing monocyte adhesion.

Figure 5.

CK1 suppresses LSS-induced monocytic adhesion to BAECs through NO production. BAECs were transfected with either control or ck1 DNA and starved in the presence or absence of 1 mM L-NAME. The cells were treated with 1 μg/ml LPS, exposed to varying levels of shear stress (0 to 10 dyn/cm²), and co-incubated with Calcein AM labeled THP-1. After 1 h, non-adherent THP-1 cells were removed, and adherent cells were visualized using fluorescence microscopy (A) and quantified as depicted in the bar graph (mean ± S.E., n = 4). *P < 0.05, **P < 0.03 (B).

Discussion

It was determined that KARS interacts with CK1 on the endothelial cell surface. Nevertheless, the vascular functions of CK1 have not been well-characterized to date. CK1 is classified within the cytokeratin family of proteins. Cytokeratins are essential for maintaining cellular architecture by forming intermediate filaments (Pavalko and Otey 1994; Fuchs and Cleveland 1998). Previous investigations have shown that cytokeratins reinforce cellular structure and regulate diverse cell functions, including cell–cell adhesion, ECM-cell communication, cell migration, receptor–ligand binding, and receptor internalization (Pavalko and Otey 1994; Fuchs and Cleveland 1998). CK1 is also localized to the surface of endothelial cells, where it forms a receptor complex with the urokinase-like plasminogen activator receptor (UPAR) and the globular head of complement 1q protein receptor (gC1qR) (Hasan et al. 1998; Shariat-Madar et al. 1999). This receptor complex facilitates interactions between high-molecular-weight kininogen (HK) and prekalikrein. The formation of this complex activates prekalikrein, resulting in the generation of NO, a major vasodilator. Conversely, the association of CK1 with myeloperoxidase (MPO), released from neutrophils during inflammation, interrupts the interaction between HK and prekalikrein, thereby suppressing NO production Astern et al. (2007). Furthermore, CK1 expression is up-regulated in response to oxidative stress due to inflammation and hypoxia, whereupon it interacts with mannose-binding lectin (MBL) Collard et al. (2001). MBL binding via CK1 to endothelial cells triggers activation of the lectin complement pathway (LCP), contributing to immune responses and tissue repair. The immunological functions of CK1 described in previous studies may be related to the vascular roles of CK1 identified in the present study.

This study also demonstrated that CK1 knockdown alters cell morphology, resulting in elongated endothelial cell shape. Previous research indicates that knocking down CK8/CK18 disrupts the Rho-ROCK pathway, which regulates cell stiffness and actin organization, through mechanisms involving both actin stress fiber rearrangement and changes in ECM-cell interactions (Marceau et al. 2001; Bordeleau et al. 2012). Given the critical function of actin fibers in ECM-cell adhesion, there is a plausible link between CK1 knockdown-induced actin remodeling and modulation of ECM-cell interactions. Prior to discussing actin fiber involvement, it is essential to highlight that focal adhesions and hemidesmosomes are the two predominant ECM-cell adhesion structures in endothelial cells. Hemidesmosome integrity depends on the interaction between laminin, integrin α6, and cytokeratins, while focal adhesions comprise multiple proteins, including fibronectin, integrin α5, and actin (Hynes 1992). Consistently, our data demonstrate that under quiescent conditions, actin colocalizes with both integrin α5 and integrin α6, but after CK1 knockdown, actin stress fibers are found exclusively with integrin α6, not with integrin α5. Collectively, these findings implicate CK1 as a key regulator of cell-ECM adhesions mediated via hemidesmosomes or focal adhesions. Notably, CK1 knockdown shifts cell-ECM adhesions from being supported by both laminin and fibronectin to primarily laminin-mediated adhesion, further supporting the hypothesis that CK1 governs laminin-dependent hemidesmosome function through its interaction with integrin α6. This mechanism was validated by our CK1 overexpression experiments. CK1 overexpression enhanced cell adhesion via interaction with laminin through its head and central domains. In contrast, CK1 knockdown did not significantly affect cell-ECM adhesion (Figure 3(D), middle panel). These findings indicate the possible involvement of an additional regulatory factor controlling cell-ECM adhesion.

Additional experiments with knockdown confirm the involvement of actin fiber in laminin-mediated cell adhesion. Pretreatment of cells with cytochalasin D followed by CK1 knockdown resulted in reduced laminin-mediated cell adhesion (Figure 3(D), right panel). Consequently, these results reinforce the proposed concept that CK1 modulates ECM-cell adhesion via direct interaction with laminin, sustaining adhesion through formation of the laminin/actin-fiber/integrin α6 complex.

Shear stress promotes cytoskeletal reorganization via ECM-cell adhesion, leading to the alignment of endothelial cells in the direction of blood flow (Davies 1995; Malek et al. 1999). These responses are mediated by shear stress-induced cell signaling. Moreover, ECM-cell adhesion serves as a key component for detecting shear stress. This is supported by the current findings, which demonstrate that CK1 overexpression exaggerates shear stress-induced activation of ERK, Akt, and eNOS, while CK1 knockdown suppresses shear stress-mediated signaling. Through domain mapping, it was determined that the head/central domains of CK1 are essential for its shear-sensing function.

Extracellular KARS was found to interact with CK1, thereby inhibiting shear stress-induced activation of cell signaling. This work further demonstrates that CK1 is vital for counteracting the effects of KARS, suggesting that CK1 functions as a receptor for shear stress. This hypothesis was validated by the ectopic expression of CK1 in Caco-2 cells, enabling them to detect and respond to shear stress. Additional domain mapping in Caco-2 cells reinforces the importance of CK1's head/central domains in mediating shear-induced cell signaling.

NO is a key vasoregulatory molecule produced by endothelial cells in response to shear stress. Previous studies reported that LSS stimulates phosphorylation of eNOS, which then generates NO. NO inhibits the progression of atherogenesis (Cines et al. 1998; Fleming and Busse 1999). Notably, CK1 accentuates LSS-mediated eNOS phosphorylation, consequently enhancing NO production. Consistent with this, CK1 overexpression diminishes leukocyte-endothelial cell adhesion during shear stress, underscoring CK1's role in atherogenesis prevention. Analyses comparing normal human tissues to those from atherosclerotic patients revealed lower CK1 expression levels in the latter (Supplementary Figure S4).

In summary, CK1 serves as a potential mechanosensor that impedes the progression of atherosclerosis (Figure 6). These results highlight CK1 as a novel therapeutic target for the prevention and treatment of cardiovascular diseases.

Figure 6.

Schematic diagram of the mechanism of action of CK1.

Glossary

Abbreviations: KARS: lysyl-tRNA synthetase; CK-1: cytokeratin-1; LSS: laminar shear stress; OSS: oscillatory shear stress; NO: nitric oxide; eNOS: endothelial nitric oxide synthase; ERK: extracellular signal-regulated kinase; ARS: aminoacyl tRNA synthetase; MSC: multisynthetase complex; ECM: extracellular matrix; BAEC: bovine aortic endothelial cell; MAEC: Mouse aortic endothelial cell; TNF: tumor necrosis factor; LPS: lipopolysaccharide; HK: high-molecular-weight kininogen; MPO: myeloperoxidase; MBL: mannose-binding lectin; LCP: lectin complement pathway

Funding Statement

This work was supported by 2025 research funds from Dankook University, Republic of Korea.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplemental Material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/19768354.2025.2526426.