Myo1g Mediates NKR‐P1C Expression in Primary Mouse NK Cells and Participates in Cytotoxicity

ABSTRACT

Myo1g, a short‐tail class I myosin, has been extensively studied for its roles in cellular adhesion, migration, cytokine secretion and receptor recycling in T and B lymphocytes. However, its involvement in other immune cell populations, particularly natural killer (NK) cells, remains an open question in immunology. NKR‐P1C is a type II integral membrane glycoprotein with a C‐type lectin domain; it also plays roles in NK cell activation and differentiation, IFN‐γ production, cytotoxic granule release. The expression of Myo1f and Myo1g mRNA in NK cells was previously found in the RIKEN database (https://gexc.riken.jp/). Thus, the present study began by confirming the presence of Myo1g protein in lysates from splenic NK cells of wild‐type (WT) mice. Subsequently, it was found that mice deficient in Myo1g (Myo1g−/−) exhibited decreased frequency and absolute numbers of NK cells (defined as CD3‐, CD19‐ and NKR‐P1C+ lymphocytes) in the bone marrow, blood and spleen. Then, NK cell development was evaluated (CD122, CD49b and NKp46), demonstrating that the absence of Myo1g does not affect overall NK cell generation but does influence NKR‐P1C expression. A functional analysis revealed reduced degranulation in NK cells from Myo1g−/− mice compared to WT controls. These findings were corroborated using purified NK cells, which demonstrated that NK cells from Myo1g−/− deficient mice exhibit reduced cytotoxicity. Collectively, these results suggest that Myo1g may serve as a potential target for modulating NK cell function and immune responses. To hypothesize potential mechanisms linking Myo1g to NKR‐P1C expression and cytotoxicity, it is possible that Myo1g influences the stability or surface trafficking of NKR‐P1C, given its known role in receptor recycling. Additionally, the absence of Myo1g could impair the cytoskeletal reorganisation required for NK cell function, affecting the formation of immunological synapses and the subsequent degranulation process.

In this study we analysed the participation of Myosin 1g in the development and functions of murine natural killer (NK) cells; we found that Myosin 1g is expressed in murine‐activated primary NK cells from the spleen. Additionally, in the absence of Myosin 1g, correlates with a reduced expression of NKR‐P1C but not of CD49b or NKp46. Finally, after purifying NK cells by negative selection, we found that the absence of Myosin 1g results in reduced degranulation and cytotoxicity. In conclusion, our results suggest that Myosin 1g could be a potential target for modulating NK cell function and immune responses.

1. Introduction

NK cell cytotoxicity is induced by the activation of multiple activating receptors encoded in the germ line resulting in the release of lytic molecules such as perforin and granzymes contained in secretory lysosomes [1] and/or in the expression of TRAIL and FASL that bind to their receptors expressed in the target cell TRAIL‐R and FAS (CD95) respectively [2]. Additionally, it has been reported that NK cells can induce pyroptosis by releasing granzyme A, which cleaves and activates gasdermin B, a pore‐forming protein whose expression is increased by IFN‐γ [3]. Cytotoxicity can be divided into three stages: recognition of the target cell, formation of the immunological synapse (SI) and death of the target cell. These processes require cytoskeleton reorganisation, such as actin polymerisation in SI and polarisation of the microtubule‐organising centre (MTOC) toward the target cell [4]. SI is generated after the target cell is recognised at the point of contact. The peripheral supramolecular activation group (pSMAC) contains adhesion molecules arranged in a ring at the point of contact. These adhesion molecules are mainly integrins, such as lymphocyte function‐associated antigen 1 (LFA‐1, CD11a/CD18) [1]; the binding of LFA‐1 with its ligand is insufficient for them to polarise from the granule to the SI [5]. The central SMAC (cSMAC), located within this ring, is the focal point of activation of signalling and exocytosis of secretory lysosomes. Actin polymerisation and cytoskeleton reorganisation are crucial for exocytosis, a process that involves the accumulation of filamentous actin in the plasma membrane‐associated cortical actin (pSMAC) region. In contrast, cSMAC is largely free of it, allowing direct contact with the secretory lysosomes and the plasma membrane of this region [1]. NKRP1C (Klrb1c, CD161c) receptor, recognised by the NK1.1 antibody (clone PK136), is an activating receptor on mouse NK cells signalling via DAP12 to trigger cytotoxicity and cytokine production. It plays an important role in antiviral immunity, tumour surveillance and immune regulation and is widely used as an NK cell marker in C57BL/6 mouse models [6]. Myosins are a family of ATP‐dependent motor proteins, grouped into 35 distinct classes; humans and mice have only 12 classes. Myosins comprise one or two heavy chains and several light chains. Heavy chains have three defined regions: the head, neck and tail. The head is the motor domain; its main function is to bind to actin [7]. The motor domain obtains energy through ATP hydrolysis to promote conformational changes that enable movement along actin filaments [8]. The neck has domains rich in isoleucine and glutamine (IQ) that facilitate interaction with light chains and enable the motor domain to move. Light chains are composed of calmodulin or calmodulin related proteins that promote neck domain stiffness and act as leverage that increases motor domain displacement since the actin translocation rate is directly proportional to the length of the lever arm [9]. The tail has fewer conserved domains that confer specific functions to each myosin, allowing interaction with proteins, lipids and even with itself [7]. Class I myosins (Myo1) are monomeric proteins with an approximate size of 110–140 kDa [10]. Humans and mice have eight genes that encode eight myosin heavy chains, from Myo1a to Myo1h. They are organised similarly in the head, neck and tail. In haematopoietic cells, class I myosins have been implicated in various functions, including intracellular transport, lamellipodia formation, cytoskeleton regulation, exocytosis regulation, endocytosis, phagocytosis, vesicular trafficking and adhesion [7]. Using the RIKEN database (https://gexc.riken.jp/), Cruz‐Zárate D. et al., determined the expression of Myo1f and Myo1g mRNA in NK cells, regardless of their maturation state and organ of origin [11]. Myo1f is a long‐tailed myosin expressed in neutrophils, macrophages, mast cells and T lymphocytes. It contains 1098 aa (124 KDa) in humans and 1099 aa (125 KDa) in mice. In addition to the TH1 domain, it contains TH2 and TH3 domains that enable it to interact with proteins such as 3BP2 and activate Cdc42, which can subsequently activate RhoA and Rac, thereby promoting cytoskeletal reorganisation and the generation of protrusions [11]. In contrast, Myo1g is a short‐tailed motor protein identified exclusively in haematopoietic cells [7]. It is primarily expressed in lymphocytes and mast cells, comprising 1018 amino acids (aa) (116 KDa) in humans and 1024 aa (117 KDa) in mice. Its PH domain binds to membrane phospholipids and lipid rafts; it differs from Myo1c in its dependence on phosphatidylinositol 4,5‐bisphosphate for membrane association [12]. Myo1g participates in cytokine secretion, cell migration, recycling of membrane molecules and cell adhesion [11]. While Myo1f influences immunological synapse formation through cytoskeletal rearrangement, Myo1g specifically affects NK cell adhesion and cytotoxic functions [11]. Activated B lymphocytes from Myo1g−/− mice show reduced adhesion to fibronectin, anti‐CD44 and anti‐LFA‐1, as well as a reduction in the expression of CD44 and LFA‐1 [13]. The presence of Myo1f and Myo1g mRNA in NK cells suggests they may have overlapping but distinct roles in modulating cytotoxicity, with Myo1g playing an important role in surface receptor expression and granule release crucial for NK cell function [11].

2. Materials and Methods

2.1. Mice

Six to twelve weeks old C57BL/6, wild‐type (WT), Myosin 1f (Myo1f−/−) and Myosin 1g (Myo1g−/−) deficient mice [14], were used for the experiments. The mice were bred at Cinvestav facilities. Cinvestav's Animal Care and Use Ethics Committee approved protocol 145‐15, which covered the experiments conducted in this project.

2.2. Genotyping of Myo 1f−/− and Myo1g−/− Mice

A small piece of tissue was obtained from the tail of mice, and DNA was extracted using the alkaline lysis method [15] and was used for genotyping by PCR, using primers described in Figure S1.

2.3. In Vivo NK Cell Activation

The mice were intraperitoneally injected with 300 μg of Poly‐I:C (Sigma) [16] After 24 h. Splenic cells were resuspended in 5 mL of PBS and carefully transferred to a tube containing 2.5 mL of Ficoll‐Paque PREMIUM (GE Healthcare). The tube was then centrifuged for 30 min at 400 g. The mononuclear cell ring was recovered and washed with PBS. Subsequently, erythrocytes were lysed with an isotonic ammonium chloride lysis buffer (NH₄Cl, 155 mM; NaHCO₃, 10 mM; EDTA, 0.1 mM), and the cells were then washed two times with PBS.

2.4. Enrichment of Primary NK Cells by Negative Selection

Mononuclear cells from mice previously treated with Poly‐IC were placed in 5 mL polyethylene tubes at a concentration of 1 × 10⁸ cells/mL in MojoSort buffer (0.5% BSA, 2 mM EDTA in PBS). NK cells were enriched using the MojoSort Mouse NK Cell Isolation Kit (BioLegend) by negative selection, as specified by the supplier.

2.5. Enrichment of Primary NK Cells by Panning

Polystyrene Petri dishes 90 × 15 mm (SPL Life sciences 10 090) were sensitised with anti‐CD90.1 (clone Thy1) (1:500) or anti‐IgM (Clone 7B6) (1:500) ascites for at least 1 h at 37°C at 5% CO₂. Mononuclear cells from mice previously treated with Poly‐IC were resuspended in 5 mL of PBS and placed in the previously sensitised Petri dishes for 30 min at 37°C in 5% CO₂. Finally, the cells that did not adhere to the plaque were recovered and counted.

2.6. Western Blot

Enriched NK cells were lysed with SDS buffer (25 mM HEPES, 2 mM EDTA, 25 mM NaF and 1% SDS) prepared with protease inhibitors (Sigma Fast), then sonicated by 3 cycles of 10 s ‘on’ and 15 s ‘off’ with an amplitude of 40% (SONICS, Vibracell VC130). Samples were run on 8% tris‐glycine SDS‐polyacrylamide gels and transferred to a nitrocellulose membrane. The membranes were blocked with 5% milk, washed three times with TBS‐Tween for 10 min and incubated overnight at 4°C with polyclonal anti‐Myo1g antibody (1:250) [12] or anti‐Myo1f antibody (Santacruz Biotechnology, 1:1000), diluted in 5% milk. After incubation, the membranes were washed five times and anti‐rabbit IgG‐HRP (Invitrogen) (1:40000) was applied. Beta‐actin was used as a loading control (anti‐beta actin hybridoma supernatant was kindly donated by Dr. José M. Hernández at Cinvestav) and anti‐IgG antimouse HRP (Jackson ImmunoResearch) (1:10000) was used. Immunoreactive bands were visualised with Clarity Western ECL.

2.7. Flow Cytometry

1 × 10⁶ total spleen cells were blocked with an Fc blocker (1:1000) for 30 min at 4°C. anti‐CD3e‐APC‐Cy7 (BioLegend), CD19‐PE (BioLegend), NKR‐P1C‐APC (BioLegend) antibodies and LIVE/DEAD Fixable Scarlet (Invitrogen) were used. The cells were washed twice with PBS and acquired using the Sysmex (XF‐1600) or BD FACS Canto II flow cytometer. The data were analysed using FlowJo version 10.10 software. The cells were fixed using fixation buffer (Sony). Afterward, the cells were washed with PBS and blocked as described for surface staining and resuspended in Intracellular Staining Perm Wash Buffer (BioLegend). The cells were blocked with 10% horse serum. Then, the cells were washed and incubated with the antibodies against Granzyme A‐PE (BioLegend), Granzyme B‐PE (BioLegend), CD3‐FITC (BioLegend), CD19‐APC‐Cy7 (BioLegend) and NKR‐P1C APC (BioLegend). Finaly the cells were washed twice with permeabilisation buffer, then twice with PBS. Likewise, 1 × 10⁶ bone marrow cells were incubated with anti‐CD3e‐PE (BioLegend), anti‐CD19‐APC‐Cy7 (BioLegend), Anti‐CD122‐Percp‐Cy5 (BioLegend), CD49b‐PECy7 (BioLegend), NKR‐P1C‐APC (BioLegend) and NKp46‐AF488 (BioLegend) antibodies, the cells were washed twice with PBS and acquired using BD FACSCanto II cytometer.

2.8. Culture of YAC‐1 Cells and Medium for Co‐Incubation

The YAC‐1 cells were cultured in RPMI‐1640 medium supplemented with 10% foetal bovine serum (FBS) (Gibco), 2 mM L‐glutamine (Gibco), 100 U/mL penicillin‐100 μg/mL streptomycin, 1 mM sodium pyruvate (Gibco) and 0.1 mM nonessential amino acids (Gibco) at 37°C in 5% CO₂. For co‐culture with NK‐enriched cells, 55 μM β‐mercaptoethanol (Gibco) and 30 ng/mL of recombinant murine IL‐15 (Preprotech) were added to the medium.

2.9. Degranulation Test

Panning‐enriched activated cells were co‐incubated with YAC‐1 target cells in an E:T ratio of 100:1, 50:1, 25:1, 12.5:1 in plates of 96 U‐bottom wells in a final volume of 200 μL. The plate was centrifuged at 200g for 3 min and then incubated for 4 h at 37°C in 5% CO₂. The cells were co‐incubated with the anti‐CD107a antibody (BD) after the 1st hour; it was then added to the Golgi Stop co‐culture (BD Golgi). NK cell degranulation was assessed by flow cytometry, analysing the CD3‐CD107a + cells.

2.10. Flow Cytometry‐Based Ex Vivo Murine NK Cell Cytotoxicity Assay

We based our approach on a protocol published previously [16]. Briefly, mice were inoculated with Poly(I:C) as previously described. The YAC‐1 cell line was harvested and stained with DiOC₁₈, a green fluorescent membrane stain from the Cell‐Mediated Cytotoxicity Kit (Invitrogen, L7010). NK cells were purified using the Mojo sort kit (BioLegend). Both cell types were co‐incubated at several effector‐target ratios (5:1, 2:1, 1:1, 0.5:1) for 4 h at 37°C, with 10 ng/mL of mIL‐15 and 55 μM of 2‐mercaptoethanol (Gibco) in a 96‐well culture plate with round‐bottom wells. Thereafter, the cells were stained with Zombie R685 fixable viability dye (1:1000) (BioLegend, 423 119). After that, the cells were washed with PBS containing 1% BSA, then fixed with 1% PFA. Finally, percent‐specific YAC‐1 killing was analyzed by flow cytometry. The analysis strategy is shown in Figure S2.

2.11. Statistical Analysis

The data are displayed as mean ± standard error (SEM) and were analysed using GraphPad Prism software 8.3.0 (San Diego, CA) with the Mann–Whitney U statistical test for nonparametric data. The values of *p < 0.05, **p < 0.001 were considered statistically significant.

3. Results

3.1. Myo1g Is Expressed in Murine‐Activated Primary NK Cells From the Spleen

The expression of Myo1f and Myo1g proteins in NK cells has not been reported in the literature; only mRNA expression for Myo1f and Myo1g in NK cells has been reported [11]. We decided to corroborate their expression using WB. NK cells were activated in vivo by injection of Poly‐IC and enriched by negative selection. The purification of NK cells was evaluated by flow cytometry, as shown in Figure 1a. After enrichment, 87% of NK cells were obtained, compared to 4% before the procedure. Since most cells that express Myo1f are neutrophils, which represent 34.0% ± 9.0% of the cells in the bone marrow (STEMCELL Technologies). Bone marrow lysates from WT (positive control) and Myo1f−/− (negative control) mice were used as antibody specificity controls. Myo1g is expressed by most are B and T lymphocytes, which are abundant in the spleen; therefore, total spleen lysate from a WT mouse (positive control) and a Myo1g−/− mouse (negative control) was used. Figure 1b shows a representative WB image of the expression of both myosins, demonstrating that the lysate from enriched NK cells shows bands corresponding to the molecular weights of these myosins; thus, we conclude that mouse NK cells express Myo1f and Myo1g.

FIGURE 1.

Myosin 1g and Myosin 1f are expressed in primary NK cells. Western blot of myosin 1g and myosin 1f. Spleen and bone marrow lysates from WT or myosin 1g−/− mice were used as positive and negative controls, respectively. (a) NK cells were enriched by mixing spleen or bone marrow cells from three mice treated with 300 μg poly‐IC for 24 h. and were identified by their size (FSC) and granularity (SSC), as they lack CD3 and CD19 expression but express NKR‐P1C. (b) The myosin 1f and the myosin 1g are shown with an approximate weight of 125 and 117 KDa, respectively. B‐actin was used as a loading control.

3.2. In the Absence of Myo1g, There Is a Decrease in the Frequency and Number of NK Cells

Given that the absence of these myosins can alter the frequencies of other cell lineages [17, 18], and since this effect had not been evaluated in NK cells, the present study assessed by flow cytometry whether the absence of Myo1f or Myo1g modifies the frequencies of NK cells in the bone marrow and spleen. Figure 2a illustrates the selection of the NK cell population, while Figure 2b,c display the NK cell populations in mice. In Myo1f−/− mice, no differences in frequencies or absolute numbers were observed in any organ analysed compared to WT mice. In contrast, Myo1g−/− mice exhibited a lower frequency and reduced numbers of NK cells relative to WT controls. This difference in Myo1g−/− mice suggested two possible explanations: a defect in NK cell development or a defect in the NKR‐P1C receptor. Given that Myo1g has been implicated in the recycling of membrane molecules [19], it was considered more likely that this effect is related to receptor localisation at the NK cell membrane, prompting further investigation of this hypothesis.

FIGURE 2.

Flow cytometry analysis of expression of NKR‐P1C receptor on CD3‐CD19‐ cells. NK cell populations in WT and myosin 1g−/−mice. (b) Frequency of NK cell populations in bone marrow (BM), blood and spleen from WT (grey), Myo1f(red) or Myo1g−/− (white) mice in 8‐week old mice. (c) The absolute number of NK cells per organ in the bone marrow and spleen in 8‐week old mice. Data are shown as mean ± SD of three independent experiments (n = 3) and were analysed using GraphPad Prism software version8.3.0 (San Diego, CA) with the t Student test for nonparametric data. *p < 0.05 and ***p < 0.001 were considered statistically significant.

3.3. Surface Expression of NKR‐P1C Is Reduced in CD3‐CD19‐ Cells From Myo1g−/−Mice

Myo1g may be involved in NK cell development. To determine whether the decrease in the CD3‐CD19‐NKR‐P1C+ population in Myo1g−/− mice was due to the absence of NKR‐P1C or reflected a more profound defect in NK cell generation, we performed intracellular staining to assess the receptor's absence or a decrease in NK cells. As shown in Figure 3a,b, in both the superficial and intracellular staining of CD3+ cells in Myo1g−/− mice, no positive population for NKR‐P1C is observed. This suggests that the defect observed in Myo1g−/− mice is not due to the intracellular localisation of the NKR‐P1C receptor; therefore, we decided to test the alternative hypothesis that the defect lies in cell development [20].

FIGURE 3.

The absence of Myo1g correlates with a reduced expression of NKR‐P1C on the membrane. (a) NKR‐P1C staining in WT mice without or with permeabilisation and Myo1g−/− mice without or with permeabilisation. (b) Bar graph showing percentage values of the lymphocyte region of CD3negNKR‐P1C+ cells of WT without or with permeabilisation and Myo1g−/− without or with permeabilisation (n = 6).

3.4. Characterisation of Granzyme A+ Cells and NK Development of Bone Marrow in WT and Myo1g−/− Mice

First, we decided to address it indirectly by counting CD3‐CD19‐ cells. As shown in Figure 4a, there is an increase in both the frequencies and absolute numbers of CD3‐CD19‐ cells in 8‐week‐old Myo1g−/− mice, which could indicate an attempt at compensation by the population of ‘immature’ NK cells or another double‐negative population. Additionally, we indirectly evaluated whether these CD3‐CD19‐ cells have intracellular expression of Granzyme A. Figure 4b,c show that there is an increased frequency of CD3‐Granzyme A+ cells in Myo1g−/− mice, and this frequency corresponds to the frequency obtained from CD3‐NK1.1+ cells fixed and permeabilised in WT mice, suggesting these cells are indeed NK cells. To enhance our understanding of the NK development we analysed CD122, CD49b and NKp46 markers [21]. The evaluation of two other molecules, NKp46 and CD49b, on spleen and bone marrow as shown in Figure 4d–f, demonstrates that the absence of Myo1g does not affect the overall generation of NK cells but does influence NKR‐P1C expression, except on progenitors NK cell subpopulation (Figure 4g).

FIGURE 4.

(a) Frequency of CD3neg CD19neg cells from bone marrow, blood and spleen of WT and Myo1g−/− mice. (b) Flow cytometry analysis of Granzyme A expression in permeabilised and non permeabilised cells from WT and Myo1g−/− mice. (c) Percentage of CD3‐GranzymeA+ cells in the lymphocyte region of WT and Myo1g−/−mice. Bars show SD(n = 2). (d) Expression of nkp46 and CD49b on the membrane of CD3‐CD19‐ spleen cells from WT (purple) or Myo1g−/− (red) mice. (e) Mean fluorescence intensity of nkp46 and CD49b expression from WT and Myo1g−/− mice, bars show SD (n = 2). (f) Flow cytometry analysis of NK development, (i) Progenitors, (ii) Immature and (iii) Mature based on NKR‐P1C and NKp46 expression. (g) Frequency of progenitors, immature and mature bone marrow NK cells of WT and Myo1g −/− mice, based on NKp46 expression, bars show SD (n = 3).

3.5. NK‐Cell Degranulation and Cytotoxicity Is Reduced on Myo1g−/− Mice

To evaluate whether Myo1g participates in the function of NK cells, the presence of Granzyme B in a population of enriched NK cells previously activated ‘in vivo’ by the injection of Poly I‐C [16, 22], demonstrating a similar level of activation among WT or Myo1g‐deficient mice (Figure 5a). We then co‐incubated NK cells with the YAC‐1 cell line to evaluate cytotoxicity. Figure 5b shows the expression of CD107a, a surrogate marker to evaluate degranulation. As shown, CD3‐, CD19‐cells from Myo1g−/− mice showed defective degranulation. Finally, we corroborated the previous observations using the protocol published by Wong et al. [16], in which purified NK cells are co‐incubated with DiOC₁₈ YAC1 cells for 4 h at different ratios, and YAC‐1 viability is evaluated with Zombie R685 fixable viability dye (as described in Section 2). The results demonstrated that NK cells from Myo1g‐deficient mice have reduced cytotoxicity (Figure 5e).

FIGURE 5.

Expression (a) and frequency (b) of CD107a + NK cells from WT (grey) and Myo1g−/− (white) mice. NK cells were enriched by mixing spleen cells from two mice treated with 300 μg poly‐IC for 24 h. NK cells were co‐cultivated with YAC‐1 cells for 4 h in the presence of anti‐CD107a and GolgiStop. (c) Verification of in vivo activation of NK cells by Granzyme B expression from WT and Myo1g−/− mice. (d) MFI of Granzyme B expression from WT and Myo1g−/− mice (n = 2). Data are shown as mean ± SD of three independent experiments (n = 3) and were analysed using GraphPad Prism software version 8.3.0 (San Diego, CA) with a t Student test for nonparametric data. *p < 0.05 was considered statistically significant. (e) Frequency of cytotoxicity of purified NK cells (E: Effector) on YAC‐1 cells (T: Target) of WT and Myo1g−/−. Comparisons were performed using a two‐way ANOVA (p < 0.0001); post hoc tests were conducted using Šidák's multiple comparisons analysis with a 95% confidence level. Comparisons were made for each dilution between the WT group and the Myo1g −/− group; all comparisons were statistically significant.

4. Discussion

NKR‐P1C is a type II, homodimer, disulfide‐linked, activating receptor primarily expressed on the cell surface of mouse NK cells. It is also found on subsets of T cells, including NKT cells, memory CD8+ T cells and some γδ T cells. Its expression is strain‐dependent, with high levels in C57BL/6, C57BL/10, CBA and 129 strains. It contains an extracellular C‐type lectin‐like domain and a short cytoplasmic tail that associates noncovalently with DAP12. Upon cross‐linking, the DAP12 ITAM motif is phosphorylated, recruiting Syk/ZAP70 kinases that trigger downstream signalling, including cytotoxicity and cytokine production [23]. On the other hand, class I myosin has become increasingly important in regulating various functions in immune cells. Thanks to their head and tail domains, class I myosins bridge membranes with the actin cytoskeleton, participating in functions such as cell adhesion, cell migration, recycling of membrane molecules, maintenance of membrane tension and cytokine secretion. Work from different laboratories (including ours) has shown that leukocytes express several class I myosins, including the short tail Myo1c and Myo1g and the long tail Myo1e and Myo1f. All these myosins bind to actin filaments, but their tail domains bind specifically to different phospholipids and other proteins. The absence of one of these myosins results in reduced cell adhesion, migration, secretion and other cellular processes [7, 24]. Still, the phenotype is not absolute, indicating that despite their specificity, there is some redundancy in their functions.

This study analysed the consequences of the absence of Myo1g and Myo1f on NK cell function. Initially, no information was available regarding the protein expression of both molecules, although mRNA for both has been reported in NK cells [11]. The results demonstrated that NK cells express both proteins. Myo1f, a class I long‐tail protein, is important for granulocyte adhesion, migration and secretion and is predominantly expressed by polymorphonuclear cells and macrophages. In contrast, Myo1g, a short‐tailed protein, is mainly expressed by lymphocytes, and its absence leads to defects in cell adhesion, migration, secretion and other cellular processes [19]. After confirming the presence of both myosins in NK cells, the effects of their absence were analysed in mice lacking either protein. The absence of Myo1f did not alter NK cell numbers; however, Myo1g‐deficient mice exhibited a marked reduction of both membrane and intracellular NKPR‐P1C molecules in NK cells. The search for intracellular NKR‐P1C was relevant because it has been reported that the absence of Myo1g decreases the surface expression of CD44, MHC‐II and LFA‐1, among other molecules. We did not detect changes in LFA‐1 and CD44 expression in NK cells from Myo1g‐deficient mice, as shown in Figure S3. The reduction of intracellular NKR‐P1C represents another, yet unexplored, mechanism in the expression of this protein. However, the absence of NKR‐P1C does not indicate a lack of NK cells; therefore, we next explored the presence of Granzyme A in CD3‐ cells as a surrogate marker to assess NK cell absence (RIKKEN database, https://gexc.riken.jp/). The results showed that CD3‐, Granzyme A+ cells were slightly increased in Myo1g‐deficient mice. We then assessed two additional molecules (Nkp46 and CD49b), which are useful for defining the NK cell phenotype [4]. The results demonstrated that both molecules were equally present in NK cells from WT and Myo1g‐deficient mice. Additionally, we explore deficiencies in NK cell development from precursors in bone marrow. Despite the dramatic reduction of NKR‐P1C, the results did not show major differences in NK cell development in the Myo1g‐deficient mice, except in the NK cell progenitor subpopulation. We hypothesize that Myosin‐1g deficiency could disrupt membrane organisation and receptor clustering, such as NKR‐P1C and others, in lymphoid progenitors, leading to impaired IL‐15R and CXCR4 signalling and defective NK‐cell differentiation [4]. Future studies should evaluate receptor dynamics and downstream signalling pathways to validate this mechanism.

In vivo activation by PolyI:C injection demonstrated that Granzyme B is equally induced in NK cells from WT and Myo1g‐deficient mice. Next, the cytotoxicity of these cells against the Yac‐1 cell line was assesed. Experiments evaluated degranulation by measuring surface expression of CD107a and residual viability of the Yac‐1 cell line in a co‐culture assay with enriched NK cells from WT or Myo1g‐deficient mice. The results demonstrated reduced degranulation and cytotoxicity of NK cells obtained from Myo1g‐deficient mice.

These results demonstrate that the absence of Myo1g affects NKR‐P1C expression via an unknown mechanism. Additionally, the lack of Myo1g reduces NK cell degranulation and cytotoxicity, which may be associated with reduced adhesion, defects in membrane tension required for secretion, and potentially other unexplored mechanisms, as observed in T and B lymphocytes. To further understand these mechanisms, future studies could investigate the role of Myo1g in receptor trafficking using live cell imaging and biochemical assays to assess the recycling and cell‐surface localisation of NKR‐P1C. Rescue experiments involving the reintroduction of Myo1g in deficient cells could also shed light on the underlying processes and verify its direct involvement in these functions. Exploring these avenues will not only deepen our understanding of Myo1g's role in NK cells but may also reveal broader implications for modulating immune functions. In summary, these findings highlight the importance of Myo1g in NK cell function. Although our results indicate that Myo1g plays a role in the expression of NKR‐P1C on NK cells, we cannot rule out other explanations, as the absence of Myo1g in other cells and tissues may modify the microenvironment on which NK cells develop. Then, further investigation is needed to clarify these mechanisms and their impact on NK cell‐mediated immune responses.

Funding

This work was supported by the Consejo Nacional de Ciencia y Tecnología (CF‐2023‐I‐741) and the Conahcyt fellowship (1144258).

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Figure S1: Schematic representation of the primer hybridisation position for Myo1g−/− and Myo1f −/− mouse genotyping. Exons are represented as blue rectangles and introns as solid lines. (a) Vertical dashed lines represent the deletion of exons 2 and 3 in Myo1g−/− mice. Representative image of a 1.5% agarose gel of endpoint PCR products for Myo1g−/− mouse genotyping. The letters A, B and A, C correspond to the primer mixtures used to amplify the WT and Myo1g−/− mouse fragments, respectively. Lane 1: base pair marker (100‐1000). Lanes 2 and 3: no DNA control. Lanes 4 and 5: WT control where only primer AB amplifies, with an expected size of 414 bp. Lanes 6 and 7: Myo1g−/− control where only the AC primers are amplified with an expected size of 424. Lanes 8–13: samples from the mice to be genotyped where it is shown that only the product corresponding to Myo1g−/− was amplified. (b) Exons are represented as blue rectangles and introns as solid lines. Vertical dashed lines represent the deletion of exons 5 and 6 in Myo1f −/− mice. Representative image of a 1.5% agarose gel of endpoint PCR products for genotyping Myo1f −/− mice. The letters A, B and A, C correspond to the primer mixtures used to amplify the WT and Myo1f−/− mouse fragments, respectively. Lane 1: base pair marker (100‐1000). Lanes 2 and 3: no DNA control. Lanes 4 and 5: WT control where only primer AB amplifies, with an expected size of 639 bp. Lanes 6 and 7: Myo1f−/− control where only the AC primers are amplified with an expected size of 563. Lanes 8–13: samples from the mice to be genotyped where it is shown that only the product corresponding to Myo1f−/− was amplified. (c) Expression of several receptors in the membrane of CD3‐CD19‐. NKR‐P1C, CD49b and nkp46 expression of the spleen from WT or Myo1g −/− mice.

Figure S2: Gating strategy for assessment of target cell killed by effectors. At the end of the 4‐h incubation, when analysed on flow cytometer cells that are double positive for DiO18 and Zombie 685 are the labelled‐target cells that had been killed. For each E:T ratio, Percent Specific Killing was calculated as (% Zombie+ of DiO₁₈+ cells) (Effector + Target) well—(%Zombie+ of DiO₁₈+ cells) Target only well. Data acquired on the BD FacsCanto II flow cytometer and analysed on Flowjo 10.8.

Figure S3: LFA‐1 and CD44 molecules are expressed on WT and Myo1g−/− NK cells. (a) Histograms obtained from flow cytometry analysis of spleen cells stained with LFA‐1 and CD44 from WT and Myo1g−/− mice. (b) Average of expression (MFI) of LFA‐1 and CD44 from WT and Myo1g−/− mice. bars with SD (n = 3).

Data Availability Statement

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