TRPV1-dependent NKCC1 Activation in Mouse Lens Involves Integrin and the Tubulin Cytoskeleton

Abstract

Previously we showed hyperosmotic solution caused TRPV1-dependent NKCC1 activation in the lens by a mechanism that involved ERK1/2 signaling. In various tissues, integrins and the cytoskeletal network play a role in responses to osmotic stress. Here, we examined the association between integrins and TRPV1-dependent activation of NKCC1 in mouse lens epithelium. Wild-type (WT) lenses exposed to the integrin agonist leukadherin-1 (LA-1) for 10 min displayed a ~33% increase in the bumetanide-sensitive rate of Rb uptake indicating NKCC activation. Paclitaxel, a microtubule stabilizing agent, abolished the Rb uptake response. In primary cultured lens epithelium LA-1 caused a robust ERK1/2 activation response that was almost fully suppressed by paclitaxel. The TRPV1 agonist capsaicin caused a similar ERK1/2 activation response. Consistent with an association between integrins and TRPV1, the TRPV1 antagonist A889425 prevented the Rb uptake response to LA-1 as did the ERK inhibitor U0126. LA-1 did not increase Rb uptake by lenses from TRPV1 knockout mice. In cells exposed to a hyperosmotic stimulus, both the ERK1/2 activation and Rb uptake responses were prevented by paclitaxel. Taken together, the findings suggest TRPV1 activation is associated with integrins and the tubulin cytoskeleton. This aligned with the observation that LA-1 elicited a robust cytoplasmic calcium rise in cells from wild-type lenses but failed to increase calcium in cells from TRPV1 knockout lenses. The results are consistent with the notion that integrin activation by LA-1, or a hyperosmotic stimulus, causes TRPV1 channel opening and the consequent downstream activation of the ERK1/2 and NKCC1 responses.

Introduction

Transparency and focusing power of the eye lens are dependent on homeostatic maintenance of the water content of the tightly packed and extensively coupled cells (Bassnett, Kuszak, Reinisch, Brown, & Beebe, 1994). The lens is transparent because it is formed by fiber cells that are structurally and functionally differentiated. Most fibers lack organelles that would scatter light and their cytoplasm is transparent due to abundant expression of soluble proteins known as crystallins. Due to their high cytoplasmic concentration of crystallin proteins, fiber cells that make up the lens cortex and nucleus require active control of their hydration (Quinlan & Clark, 2022). Because fibers have limited ion transport capability, ion and water homeostasis of the entire lens relies on the monolayer of epithelial cells at the anterior surface. By regulating the activity of ion transport mechanisms in the epithelium, the lens can compensate, to a degree, for osmotic swelling or shrinkage. Hypoosmotic stimuli cause TRPV4-dependent Na,K-ATPase activation (Delamere et al., 2020). The shrinkage response is entirely different. Previously we showed that mouse lenses exposed to hyperosmotic solution caused TRPV1-dependent NKCC1 activation by a mechanism that involved ERK1/2 signaling (Shahidullah, Mandal, & Delamere, 2018; Shahidullah, Mandal, et al., 2020). In various tissues, investigators have suggested the integrins and the cytoskeletal network play a role in responses to osmotic stress (Jiao, Cui, Wang, Li, & Wang, 2017).

Integrins influence and are influenced by dynamic cell-cell and cell-matrix adhesion (Bachmann, Kukkurainen, Hytönen, & Wehrle-Haller, 2019). They play a role in activation and integration of a variety of cellular responses, often in conjunction with the microtubule cytoskeleton (LaFlamme, Mathew-Steiner, Singh, Colello-Borges, & Nieves, 2018). In collaboration with other receptors, integrins can activate signal transduction cascades that regulate many cellular activities, including changes in gene expression, cytoskeletal assembly, proliferation, survival, migration, and tissue homeostasis (Danen & Sonnenberg, 2003; Hynes, 2002; Streuli & Akhtar, 2009). Because integrin receptors can function as bi-directional signal distribution hubs (Hu & Luo, 2013; Hynes, 2002), mechanical regulation of integrin adapter conformations can play a part in mechanosensing (Hytönen & Wehrle-Haller, 2016). For example, integrins are affected when osmotic shrinkage causes the plasma membrane to slacken inwardly. There are various reports on the interaction of mechanosensitive channels and integrins (Jiao et al., 2017) as well as interactions between tubulin and TRPV1 (Prager-Khoutorsky, Khoutorsky, & Bourque, 2014).

In the present study, we considered whether integrins are associated with the TRPV1-dependent activation of NKCC1 in mouse lens epithelium. We used a Rb uptake approach to examine the response to an integrin agonist, leukadherin-1 (LA-1), and also probed for ERK1/2 activation.

Materials and Methods

Chemicals and Antibodies

Bumetanide, adenosine 5′-triphosphate disodium salt hydrate, rubidium chloride and chemicals for preparing Krebs solution were purchased from Sigma (St. Louis MO). TRPV1 antagonist, A889425 was purchased from alomone Labs (Jerusalem, Israel). Fura2-AM was purchased from Thermo Fisher Scientific (Waltham, MA). Paclitaxel and Leukadherin-1 were obtained from Selleckchem (Selleck Chemical LLC, Houston, TX). Rabbit polyclonal anti p44/42 MAP kinase antibody, mouse monoclonal anti phospho-p44/42 MAP kinase (Thr-202/Tyr-204) antibody were obtained from Cell Signaling Technology (Danvers, MA, USA). Mouse monoclonal β-actin antibody was purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Goat anti-rabbit secondary antibody conjugated with IRDye 680 and goat anti-mouse secondary antibody conjugated with IR Dye 800 were obtained from LI-COR Biosciences (Lincoln, NE, USA). The Micro BCA Protein Assay Kit was purchased from Thermo Scientific Pierce (Rockford, Illinois, USA).

Krebs Solution

All experiments, including Rb uptake studies, Western blot and intracellular calcium measurements were conducted using bicarbonate-buffered Krebs solution that contained (in mM) 119 NaCl, 4.7 KCl, 1.2 KH2PO4, 25 NaHCO3, 2.0 CaCl2, 1 MgCl2, and 5.5 glucose. For Rb uptake experiments, Rb-containing Krebs solutions was used in which KCl and KH₂PO4 were replaced by adding an equivalent concentration of RbCl and NaH₂PO₄, respectively. The pH for all solutions was adjusted to 7.4. Krebs solution was bubbled with a mixture of 5% CO₂ and 95% air for 45 minutes before pH adjustment to 7.4.

Lenses and cultured lens epithelium

Lenses were obtained from adult (18–20 weeks) male and female wild type C57BL/6J mice or TRPV1 KO mice (B6.129X1-Trpv1<tm1Jul>/J) (Jackson Laboratory, Maine). The use of animals was conformed to the Universal Declaration on Animal Welfare (UDAW) and approved by the Institutional Animal Care and Use Committee (IACUC) of the University of Arizona. Following CO₂ euthanasia, eyes were removed and each lens was isolated by careful dissection and transferred to Krebs solution. Rb uptake experiments were performed using intact lenses and Western blot studies were performed using confluent primary cultured lens epithelial cells. Intracellular calcium measurements were performed using semiconfluent primary cultured lens epithelial cells.

Lens epithelium culture.

Lens epithelial cells were obtained from freshly isolated lenses obtained from male or female mice and established in primary culture using an approach described previously (Shahidullah, Mandal, et al., 2020). The capsule-epithelium was isolated from the lens using two pairs of fine forceps, one to tear open the capsule, the another to hold the capsule. Each capsule-epithelium explant was unfolded and spread out on the surface of a 60-mm collagen-coated tissue culture dish (Corning). Samples from 4–6 lenses were spaced at equal distances on each dish. A small amount (~0.5 ml) of complete culture medium was placed along the margin of the dish to prevent dehydration and the lid was positioned, then the dish was placed in a CO₂ incubator at 37°C. To facilitate cell spreading from the capsule-epithelium explants, two to three scratches were made on each capsule/epithelium after spreading it on the dish. When the explants had firmly attached to the bottom of the dish (usually after 60 min), 4 mL of complete medium was added to flood the dish and cover the explants. The complete culture medium was prepared by mixing the three components of an epithelial cell medium kit (Sciencell Research Laboratories, Carlsbad, CA): 500 mL of basal EpiCM medium, 10 mL of fetal bovine serum (FBS), and 5 mL of a mixture of penicillin and streptomycin. To allow sufficient time for the cells to recover and spread to the area surrounding the explants, the first medium change was done after 3–4 days. Then the medium was changed every alternate day. After 6–8 days, cells had sprouted out of the explants and there were enough to be trypsinized and propagated. For trypsinization, the medium was removed, cells were washed twice with HBSS without calcium or magnesium and then incubated at 37°C with 4.0 mL of 0.25% Trypsin EDTA for 3 min under constant shaking at low speed. The trypsin was neutralized by adding an equal volume of a mixture of FBS and newborn calf serum (1:1). The cell suspension was then placed in 15-ml tubes, triturated 3–4 times using a sterile disposable plastic transfer pipette and centrifuged at 167 g for 10 min. The supernatant was discarded, and the pellet was resuspended in 4–5 mL of complete medium and seeded in 25 cm² flask at a density of 10,000–15,000 cells/cm². The medium was changed the next day and then every alternate day. When the cells became confluent, usually in 5–6 days, they were propagated to the next passage. Only third-passage cells were used in this study.

Rubidium uptake

Rubidium (Rb) uptake studies were carried out using freshly isolated intact mouse lenses using an approach based on a method described previously (Shahidullah et al., 2018; Shahidullah, Mandal, et al., 2020). Briefly, each lens was carefully dissected from the eye under visualization with a dissecting microscope to make sure the capsule remained intact. Damaged lenses were discarded. Lenses were immediately incubated in Rb-containing Krebs solution at 37°C with or without test agents. Rb uptake time was 10 min. In some cases, lenses were preincubated with an antagonist or inhibitor in normal Krebs solution for a specified time prior to the uptake period in Rb-containing Krebs solution. At the end of the 10 min uptake period, lenses were washed in plentiful ice-cold isotonic MgCl₂ solution (5 ml for each lens). After the wash, the lenses were cleaned of any attached vitreous or ciliary epithelium by gently rolling on a filter paper moistened with isotonic MgCl₂ solution. Each lens was then placed in a pre-weighed 12×75 mm glass tube and weighed again to obtain lens wet weight. The lenses were then dried for 7 days in an oven at 70°C to remove water. The weight of each tube with dry lens was recorded and used to calculate lens dry weight. Dry lenses were transferred to 5 ml Eppendorf tubes and digested overnight at 65°C in 200 µl of 30% nitric acid. Then 1.8 ml of double distilled water was added to each tube to make a final volume of 2 ml. Rb concentration in the solution was determined by atomic absorption spectrophotometry using an AAnalyst 100 (PerkinElmer, Waltham, MA). Rb uptake results were expressed as mmol/kg lens dry weight/10 min. NKCC-mediated Rb uptake was teased out by using bumetanide, a recognized NKCC inhibitor. We cannot rule out a possible minor contribution of channel-mediated Rb⁺ uptake, either via lens K⁺ channels or less abundant TRPV1 channels. However, this contribution would be extremely negligible since (1) electrochemical gradient will oppose Rb entry into the cells, (2) bumetanide completely blocked the LA-1 mediated increase in Rb uptake which has no effect on TRPV1 channel opening.

Western Blot

Western blot studies were used to measure ERK1/2 phosphorylation using an approach described earlier (Shahidullah, Wei, & Delamere, 2013). After treating confluent monolayers of primary cultured lens epithelium with test agents for a specified period in Krebs solution at 37°C, the cells were homogenized in iced-cold RIPA buffer [50 mM HEPES, 150 mM NaCl, 1 mM EDTA, 10 mM sodium fluoride, 10 mM sodium pyrophosphate, 2 mM sodium orthovanadate, 10% glycerol, 1% Triton X-100, 1% sodium deoxycholate, plus a protease inhibitor cocktail (Thermo Fisher Scientific, IL) and phosphatase inhibitor cocktails 1 and 2 (EMD Millipore) at pH 7.5. To obtain an adequate protein concentration, confluent cells from each 60 mm dish were homogenized in 250 µl of RIPA buffer. The homogenate was centrifuged for 30 min at 14,000 g, then the supernatant was collected. Protein concentration in the supernatant was measured using a BCA protein assay kit (Pierce). An aliquot of supernatant containing 20 µg protein was mixed with Laemmli buffer and subjected to SDS-PAGE gel (7.5%) electrophoresis to separate the proteins. The proteins in the gel were then transferred by electrophoresis to nitrocellulose membrane which was kept in a blocking buffer overnight at 4°C (AquaBlock; East Coast Biologics, Inc.). The nitrocellulose membrane was then treated overnight at 4°C with primary antibodies: rabbit polyclonal anti p44/42 MAP kinase antibody (1;1000), mouse monoclonal anti phospho-p44/42 MAP kinase (Thr-202/Tyr-204) antibody (1:2000). After treatment with primary antibodies, the membrane was washed 3 times with Tween 20 in Tris-buffered saline (TTBS) (30 mM Tris, 150 mM NaCl and 0.5% (vol/vol) Tween 20; pH 7.4) and incubated for 60–90 min at room temperature with goat anti-rabbit secondary antibody conjugated with IRDye 680 or goat anti-mouse secondary antibody conjugated with IR Dye 800 (1:20,000; LI-COR, Lincoln, NE). Then the membrane was washed twice (5 min each) with TTBS at room temperature and twice (5 min each) with PBS at room temperature. Protein band densities were then measured using an Odyssey infrared scanner (LI-COR). By using secondary antibodies at two different infrared wavelengths, both the total and phospho-ERK1/2 were quantified simultaneously, permitting accurate calculation of a band density ratio.

Intracellular calcium

Single cell cytoplasmic calcium concentration was measured in Fura-2 AM-loaded cells using an microscope-base imaging system (Intracellular Imaging Inc., Cincinnati, OH) and as described earlier (Shahidullah, Rosales, & Delamere, 2022; Shahidullah, Wilson, et al., 2020). Semiconfluent cells (<50% confluence, 48h culture) plated on a 35 mm culture dish were loaded by incubation at 37°C in 1 ml of normal Krebs solution, containing 10 µM Fura-2 AM for 20 min prepared from a Fura-2 AM stock solution (1 mM) in DMSO with 20% pluronic. At the end of the loading period, cells were washed 3–5 times (1 ml each time) then incubated further at 37°C for 15–20 min for de-esterification. Then the cells were again washed 3–5 times with normal Krebs solution and the dish was mounted on an open perfusion mini-incubator (Harvard Apparatus, Model PDMI-2) attached to the stage of an inverted microscope (Nikon Eclipse TS100). The cells were continuously superperfused (1.0 ml per min) with prewarmed Krebs’ solution equilibrated with 95% air/5% CO₂ and the fluorescence intensities were measured at alternating excitation wavelengths of 340 nm and 380 nm, and the emission collected at 510 nm. After 5 min of control measurement, the superfusate was switched to a Krebs solution containing the test agent. The temperature of the perfusate was maintained at 37°C using a temperature controller (Harvard Apparatus, Holliston, MA, USA, Model TC. 202A). The fluorescence ratios were converted into free calcium concentration by the system-integrated software, with a calibration curve obtained according to the manufacturer’s protocol, using a Fura-2 calcium calibration kit supplied by Invitrogen. In each experiment, data from 15 to 30 individual cells were averaged and considered as n = 1.

Statistical Analysis

A two-sample t test was used to compare the difference between two groups of data. One way analysis of variance (ANOVA) followed by Šídák’s multiple comparison tests was used to compare differences for more than two groups of data. A probability (P) value of <0.05 was considered significant. The data from each experiment were also tested for normality by using the Kolmogorov-Smirnov test and Shapiro-Wilk test and by constructing QQ plots (predicted v actual). The data showed a normal Gaussian distribution.

Results.

Intact wild-type mouse lenses were exposed to the integrin agonist leukadherin-1 (LA-1) and Rb uptake was measured as an index of inwardly directed potassium transport. Lenses exposed to 25 µM LA-1 for 10 min displayed a ~33% increase in the rate of Rb uptake (Fig. 1). NKCC-mediated Rb uptake was teased out by using bumetanide (Fig 1A), a selective NKCC inhibitor and which has no effect on TRPV1-mediated calcium entry. The increase in Rb uptake was absent when lenses were exposed to LA-1 in the presence of bumetanide (1 µM), (Fig. 1A). The ability of bumetanide to prevent the Rb uptake response is consistent with stimulation of NKCC activity by LA-1. When lenses were exposed to LA-1 in the presence of paclitaxel (100 nM), a microtubule stabilizing agent, the Rb uptake response to LA-1 was abolished (Fig. 1B).

Fig. 1.

The influence of bumetanide or paclitaxel on leukadherin-1 (LA-1)-induced Rb uptake response by intact wild-type mouse lenses. Freshly isolated lenses were incubated for 10 min in Rb-containing Krebs solution in the presence or absence (Control) of LA-1 (25 µM). Some lenses also were exposed to the NKCC inhibitor bumetanide (Bum) (1 µM, 20 min preincubation) or microtubule stabilizer paclitaxel (100 nM, 6h preincubation) and then exposed to LA-1 in the continued presence of either bumetanide (Panel A) or paclitaxel (Panel B). The values are the mean ± SEM of results from 7–14 lenses. *** (p<0.001) Indicates a significant difference from control.

To keep animal usage to a minimum, some studies on the response to LA-1 were carried out using primary cultured mouse lens epithelium. Using this approach, LA-1 was found to cause a robust ERK1/2 activation (phosphorylation) response, revealed by Western blot analysis (Fig. 2A). ERK1/2 activation was transient, reaching a maximum at 5 min. Paclitaxel almost fully suppressed the ERK1/2 response to LA-1 (Fig. 2B).

Fig. 2.

Western blot showing the time-course of ERK1/2 activation in cultured wild-type mouse lens epithelium exposed to LA-1 (Panel A). Samples isolated from cultured lens epithelium, exposed to LA-1 (25 µM) for 2 −15 min, were separated by electrophoresis, then probed for phospho-ERK1/2 (pERK1/2, Upper bands) and total ERK1/2 (tERK1/2, Lower bands). pERK1/2 band density displayed a transient increase that reached a maximum at 5 min. Paclitaxel prevented the ERK1/2 activation response in cultured wild-type mouse lens epithelial cells exposed to LA-1 (Panel B). Cells were exposed to LA-1 (25 µM) for 5 min in the presence or absence (Control) of paclitaxel (100 nM, 6h preincubation) were separated by electrophoresis and then probed for phospho-ERK1/2 (pERK1/2, Upper bands) and total ERK1/2 (tERK1/2, Lower bands). In each panel a representative Western blot was shown on the left and the bar graph for the band density ratio of phospho-ERK/total ERK (pERK/tERK) was shown on the right. The results are the mean ± SEM of 3 independent experiments. * (p < 0.05) and ** (p<0.01) indicate significant differences from control.

While peak ERK1/2 activation occurred at 5 min of LA-1 treatment (Fig. 2), significant activation was detectable at 2 min (the shortest time-point examined) and remained significantly elevated for at least 15 min. The time course of the Western blot ERK1/2 activation study is consistent with the Rb uptake study results (Fig. 1) that showed increased Rb uptake in lenses exposed to LA-1. The Rb uptake study was based on a 10 min LA-1 exposure period and ERK1/2 was activated for most of this time.

To examine the possible association between integrins and TRPV1, Rb uptake was measured in intact wild-type lenses exposed to LA-1 in the presence of the TRPV1 antagonist A889425 (1.0 μM). The increase of Rb uptake caused by LA-1 was eliminated by A889425 (Fig. 3A). The findings suggest TRPV1 might be required for the ion transport response to integrin activation. In keeping with this notion, LA-1 failed to increase Rb uptake by lenses from TRPV1 knockout mice (Fig. 3B).

Fig. 3.

The TRPV1 inhibitor A889425 prevented the Rb uptake response in wild-type mouse lenses exposed to LA-1 (Panel A). Rb uptake was measured in intact wild-type lenses exposed to LA-1 (25 µM) for 10 min in the presence or absence of A889425 (1 µM) that was added 20 min earlier. Control lenses received neither LA-1 nor A889425. The values are the mean ± SEM of results from 5–6 lenses. ** (p<0.01) Indicates a significant difference from control. LA-1 did not increase Rb uptake by intact lenses obtained fromTRPV1 knockout mice (Panel B). TRPV1 KO lenses were incubated for 10 min in Rb-containing Krebs solution in the presence or absence (Control) of LA-1 (25 µM). The values are the mean ± SEM of results from 7 lenses.

Our earlier studies on porcine lens epithelium showed TRPV1-dependent ERK1/2 phosphorylation when cells were exposed to either capsaicin or hyperosmotic solution (350 mOsm) (Mandal, Shahidullah, & Delamere, 2018). To determine whether hyperosmotic solution-induced ERK1/2 activation in mouse lens epithelium is linked to integrin-tubulin interaction, we examined the effect of paclitaxel on hyperosmotic-induced ERK1/2 activation and Rb uptake responses. Hyperosmotic solution elicited a pattern of ERK1/2 activation like that of LA-1 response with a peak at 5 min (Fig. 4). Importantly, the hyperosmotic solution induced ERK1/2 activation and Rb uptake responses were completely prevented by paclitaxel (Fig. 5).

Fig. 4.

Western blot showing the time course of ERK1/2 activation in cultured wild-type mouse lens epithelium exposed to hyperosmotic solution. Samples isolated from cells exposed to hyperosmotic solution (350 mOsm) for 2 −15 min were separated by electrophoresis then probed for phospho-ERK1/2 (pERK1/2, Upper bands) and total ERK1/2 (tERK1/2, Lower bands). pERK1/2 band density displayed a transient increase reaching maximum at 5 min. The Left panel shows a representative Western blot. The bar graph in the Right panel shows mean ± SEM of 3 independent experiments. * (p < 0.05) and ** (p<0.01) indicate significant differences from control.

Fig. 5.

Paclitaxel prevented the ERK1/2 activation response in cultured wild-type mouse epithelial cells exposed to hyperosmotic solution (Panel A). Cells were exposed to hyperosmotic solution (350 mOsm) for 5 min in the presence or absence (Control) of paclitaxel (100 nM) that was added 6h earlier (preincubation time). A representative Western blot is shown on the Left. The bar chart on the Right shows the band density ratio of phospho-ERK/total ERK (pERK/tERK). The results are the mean ± SEM of 3 independent experiments. *** (p<0.001) Indicates a significant difference from control. Paclitaxel also prevented the increase of Rb uptake in intact wild-type lenses exposed to hyperosmotic solution (Panel B). Rb uptake was measured in lenses exposed to hyperosmotic solution for 10 min in the presence or absence (Control) of paclitaxel (100 nM) added 6h earlier (preincubation time). The results are the mean ± SEM of 6 – 7 independent experiments. *** (p<0.001) Indicates a significant difference from control.

Because TRPV1 activation was previously found to increase NKCC1 activity by a mechanism that involved ERK1/2 signaling (Shahidullah et al., 2018), we compared ERK1/2 activation in cultured lens epithelial cells exposed to LA-1 or capsaicin, a TRPV1 agonist. Like LA-1 (see Fig. 3) capsaicin (1.0 μM) caused transient ERK1/2 activation (Fig. 6). The ERK1/2 activation time course was similar in cells treated with LA-1 or capsaicin (Fig. 7).

Fig. 6.

Western blot showing the time course of ERK1/2 activation in cultured wild-type mouse lens epithelium exposed to the TRPV1 agonist capsaicin. Samples isolated from cells exposed to capsaicin (1 µM) for 2 −15 min were separated by electrophoresis then probed for phospho-ERK (Upper bands) and total ERK1/2 (Lower bands). pERK1/2 band density displayed a transient increase with a maximum at 5 min. The Left panel shows a representative Western blot. The bar chart in the Right panel shows the band density ratio of phospho-ERK/total ERK (pERK/tERK). The results are the mean ± SEM of 3 independent experiments. * (p<0.05) Indicates a significant difference from control.

Fig. 7.

The time course of ERK1/2 activation was similar in cultured wild-type mouse lens epithelial cells exposed to LA-1 or capsaicin. As described above (Figs 3, and 9), samples isolated from cells exposed to LA-1 (25 µM) or capsaicin (1 µM) for 2 −15 min were probed by Western blot for phospho-ERK1/2 and total ERK1/2. The line graph shows the band density ratios, pERK/total ERK, at the time points shown on the x-axis. The results are the mean ± SEM of 3 independent experiments. * (p<0.05 and ** (p<0.01) indicate a significant difference from control for each treatment. There is no statistical significance between capsaicin and LA-1 treatment at any of the time points.

In studies on porcine lens, TRPV1 in the epithelium was found to be the sensor in a complex feedback mechanism that responds by activating ERK1/2, WNK and NKCC1 when the lens is subjected to a hyperosmotic stimulus (Shahidullah et al., 2018; Shahidullah, Mandal, et al., 2020). Consistent with the role of ERK1/2 in the response mechanism, the ERK inhibitor U0126 (10 µM) abolished LA-1-induced increase of Rb uptake in mouse lens (Fig. 8).

Fig. 8.

A highly selective ERK1/2/ MEK inhibitor, U0126, prevented the Rb uptake response in wild-type mouse lenses exposed to LA-1. Rb uptake was measured in intact wild-type lenses exposed to LA-1 (25 µM) for 10 min in the presence or absence of U0126 (10 µM) that was added 20 min earlier (preincubation time). Control lenses received neither LA-1 nor U0126. The values are the mean ± SEM of results from 6 lenses. *** (p<0.001) Indicates a significant difference from control.

Taken together, the findings are consistent with TRPV1 activation being associated with integrins and the tubulin cytoskeleton. To examine this further, we used Fura-2 to measure cytoplasmic calcium in cultured mouse lens epithelial cells exposed to LA-1. The integrin agonist elicited a robust increase of cytoplasmic calcium in cells from wild-type lenses (Fig. 9, left panel). In contrast, LA-1 failed to increase calcium in cells from TRPV1 knockout lenses (Fig. 9, right panel). As a positive control, a calcium increase caused by ATP was confirmed in TRPV1 knockout cells. A schematic diagram for the signaling cascade on LA-1 induced integrin activation leading to NKCC1 activation and Rb uptake increase is shown (Fig. 10).

Fig. 9.

LA-1 caused an increase of cytoplasmic calcium in cultured lens epithelial cells from wild type but not TRPV1 knockout (KO) mice. The results were obtained in cells loaded with Fura-2 AM. LA-1 (25 µM) added as shown (↑) caused an immediate increase of cytoplasmic calcium concentration in wild-type cells (Left panel) that was absent in TRPV1 KO cells (Right panel). The calcium response of TRPV1 KO cells to ATP (100 µM) was recorded as a positive control. Data from 15–30 cells were averaged and considered as n=1. Results are means ± SEM of 5 independent experiments.

Fig. 10.

Schematic diagram showing the signaling cascade and its possible components involved in Rb uptake increase due to integrin activation by LA1 or by hyperosmotic solution. Note that signaling components following TRPV1 activation were published earlier (Shahidullah, et al 2018; 2020).

Discussion

The present findings point to a functional link between lens integrin activation and tubulin cytoskeleton linked TRPV1 activation. TRPV1 activation is known to cause a characteristic signaling response in the lens that stimulates NKCC1 (Mandal et al., 2018; Shahidullah et al., 2018; Shahidullah, Mandal, et al., 2020). Because integrin activation appears to cause TRPV1 channel opening, integrin activation also triggers an NKCC1 response. The link between integrin, the tubulin cytoskeleton and TRPV1 activation might be an important part of the mechanism that causes lens TRPV1 activation in response to osmotic shrinkage. In support of this notion, the integrin activator LA-1 was found to elicit a variety of responses consistent with TRPV1 channel opening, including calcium entry, ERK1/2 activation and stimulation of NKCC activity evident as increased bumetanide-sensitive Rb uptake. LA-1 caused a calcium entry response in wild-type lens epithelium but not epithelium from TRPV1 knockout lenses. Moreover, the Rb uptake response to integrin activation by LA-1 was inhibited by an ERK inhibitor U0126 and a TRPV1 antagonist, A889425, and the response was absent in lenses from TRPV1 knockout mice. The results are consistent with the notion that integrin activation causes TRPV1 channel opening and the consequent downstream activation of the ERK1/2 and NKCC1 responses.

Integrins are αβ–heterodimers that act as cell-extracellular matrix and cell-cell adhesion molecules mediating cell-cell and cell-extracellular matrix adhesive connections and signal transduction across the plasma membrane (Springer, 1990). Integrins are the principal receptors used by animal cells to bind to the extracellular matrix. Leukadherin-1 (LA-1) is a small molecule agonist that works primarily by binding to and influencing beta-2 (β−2) integrin receptor (CD11b/CD18) which is expressed in lens epithelium (Mathew et al., 2003; McLean et al., 2005; Zhang et al., 2000). CD11 is defined as the α component (Integrin α chain) of various integrins, especially ones in which the β component is CD18 (integrin β2 chain).

It was observed that the microtubule stabilizer paclitaxel inhibited the Rb uptake and ERK1/2 responses to LA-1. This is consistent with crosstalk that can exist between integrins and microtubules (LaFlamme et al., 2018). Microtubules are polymerized α- and β-tubulin heterodimers that play fundamental roles in vital cellular processes. Microtubules exhibit special dynamic properties. In a population of microtubules, at any point in time a subset of microtubules is rapidly growing while others are quickly shrinking. Individual microtubules switch randomly between growing and shrinking states, sometimes changing back and forth multiple times during their lifetime. This combination of growth, shrinkage, and rapid transitions between the two is defined as dynamic instability. We suggest that LA-1 binding to integrin receptor stimulates microtubule dynamic instability and subsequently activates TRPV1 and NKCC leading to increase in bumetanide-sensitive Rb uptake. How this dynamic instability of microtubules involves and activates TRPV1 is beyond the scope of this study. However, it is known that dynamic instability is suppressed when microtubule associated proteins (MAPs) or other stabilizing agents, such as taxol or glycerol, are present (Drechsel, Hyman, Cobb, & Kirschner, 1992; Hotani & Horio, 1988; Kowalski & Williams, 1993; Schilstra, Bayley, & Martin, 1991). Previously it was also reported that paclitaxel (a microtubule stabilizer) disrupts intracellular NKCC1 trafficking by interfering with microtubule dynamics and associated motor proteins (Chen et al., 2014). While TRPV1 may be intrinsically sensitive to distortion of the lipid bilayer (Argudo, Bethel, Marcoline, Wolgemuth, & Grabe, 2017), there are various reports on the interaction of mechanosensitive channels and integrins (Jiao et al., 2017) as well as interactions between tubulin and the TRPV1 cytoplasmic terminus (Prager-Khoutorsky et al., 2014). Tubulin binding motifs on TRPV1 have been reported (Goswami, Hucho, & Hucho, 2007). However, there is variation between different cell types and no apparent consensus on mechanism, though interaction of TRPV1 with the actin cytoskeleton has not been documented (Goswami et al., 2004; Goswami et al., 2007). If, as our data suggest, TRPV1 activation occurs through a functional link with the tubulin cytoskeleton, such activation might occur when integrin activation imposes rearrangement of the tubulin cytoskeleton, or when osmotic shrinkage distorts the plasma membrane inward against the cytoskeleton. This may explain why paclitaxel is able to suppress TRPV1-dependent ERK1/2 activation that occurs in cultured lens epithelial cells exposed to hyperosmotic stimulus.

TRPV1 is a polymodal ion channel that is activated by a variety of stimuli including heat, low pH and capsaicin (Luu, Owens, Mebrat, & Van Horn, 2023). TRPV1 is a nonselective channel that allows permeation of monovalent cations although has a tenfold preference for calcium ions (Caterina & Julius, 2001). TRPV1 activation plays an important role in the lens. The TRPV1 feedback loop mechanism increases NKCC1 activity in the lens epithelium to compensate for osmotic shrinkage of the fiber mass and decrease of cellular hydrostatic pressure (Delamere et al., 2020). TRPV1 acts as a sensor and its activation triggers a signaling response that involves ERK1/2, PI3 kinase, Akt, WNK. Ultimately, NKCC1 activity in the epithelium is activated (Shahidullah et al., 2018; Shahidullah, Mandal, et al., 2020). Remarkably, the TRPV1 feedback loop works only to correct for osmotic shrinkage. It is unaffected by osmotic swelling which instead activates TRPV4 and a different feedback loop (Delamere et al., 2020; Shahidullah, Mandal, & Delamere, 2012). It remains to be determined whether the activation of TRPV1 by shrinkage but not swelling is related to the association of TRPV1 activation with integrin and the tubulin cytoskeleton.

The findings are consistent with lens integrin expression. In humans, there are 24 members of the integrin family, α/β heterodimers formed by various combinations of 18 α subunits and 8 β subunits. They can be categorized into subfamilies based on ligand or cell specificity including leucocyte-specific receptors, RGD (Arg-Gly-Asp) receptors, laminin receptors and collagen receptors. The lens expresses several α and β integrin subunits including β1 (CD29) and β2 (CD18) (Mathew et al., 2003; McLean et al., 2005; Menko & Philip, 1995; Scheiblin et al., 2014; Walker, Zhang, Zhou, Woolkalis, & Menko, 2002; Wang et al., 2017) that are involved in various functions including regulation of differentiation, fiber cell homeostasis, and responding to cellular stress. LA-1 interacts with β−2 integrin (CD18) which is expressed in lens epithelium (Mathew et al., 2003; McLean et al., 2005). There have been several studies on lens TRPV1 expression but not, as far as we know not, on TRPV1 and integrin interaction. In trigeminal ganglia, TRPV1 is expressed together with integrin subunits that can bind fibronectin (Jeske, Patwardhan, Henry, & Milam, 2009) while integrin-TRPV1 coordination is critical for osmosensation (Jiao et al., 2017).

The arrangement of lens ion transport mechanisms is highly unusual and TRPV1 channels are required for osmotic homeostasis. Almost the entire structure is formed by fiber cells that have negligible Na,K-ATPase activity and thus very limited capacity for primary or secondary active transport. Na,K-ATPase specific activity is high only the epithelial monolayer at the anterior surface (Delamere & Dean, 1993; Tamiya, Dean, Paterson, & Delamere, 2003). Fibers are unable to function on their own and homeostasis of the entire lens relies on the relatively small number of epithelial cells. There are two opposing feedback loops in the epithelium. One feedback loop is activated by TRPV4 and corrects for lens swelling by activating Na,K-ATPase. The opposing feedback loop is activated by TRPV1 and corrects for lens shrinkage by activating NKCC1 (Shahidullah, Mandal, et al., 2020). Because they activate the feedback loops that regulate homeostasis, TRPV4 and TRPV1 are master controllers of lens function. Findings from the present study suggest TRPV1 activation might occur because it has a functional link with the tubulin cytoskeleton, arranged such that channel opening occurs when integrin activation rearranges the tubulin cytoskeleton, or when osmotic cell shrinkage distorts the cytoskeleton. Then, as previously reported (Shahidullah et al., 2018; Shahidullah, Mandal, et al., 2020), TRPV1 triggers ERK1/2, WNK and SPAK signaling that leads to an increase in NKCC1 activity that attempts to compensate for the osmotic imbalance.

Data availability statements:

Generated and analyzed data for this study can be found in our data repository at: https://arizona.box.com/s/33jz95xz7k0tvetkyhoyjap86v1ju362