Estrogen modulation of MgATPase activity of nonmuscle myosin-II-B filaments

Abstract

The study tested the hypothesis that estrogen controls epithelial paracellular resistance through modulation of myosin. The objective was to understand how estrogen modulates nonmuscle myosin-II-B (NMM-II-B), the main component of the cortical actomyosin in human epithelial cervical cells. Experiments used human cervical epithelial cells CaSki as a model, and end points were NMM-II-B phosphorylation, filamentation, and MgATPase activity. The results were as follows: 1) treatment with estrogen increased phosphorylation and MgATPase activity and decreased NMM-II-B filamentation; 2) estrogen effects could be blocked by antisense nucleotides for the estrogen receptor-alpha and by ICI-182,780, tamoxifen, and the casein kinase-II (CK2) inhibitor, 5,6-dichloro-1-beta-(D)-ribofuranosylbenzimidazole and attenuated by AG1478 and PD98059 (inhibitors of epithelial growth factor receptor and ERK/MAPK) but not staurosporine [blocker of protein kinase C (PKC)]; 3) treatments with the PKC activator sn-1,2-dioctanoyl diglyceride induced biphasic effect on NMM-II-B MgATPase activity: an increase at 1 nm to 1 microM and a decrease in activity at more than 1 microM; 4) sn-1,2-dioctanoyl diglyceride also decreased NMM-II-B filamentation in a monophasic and saturable dose dependence (EC(50) 1-10 microM); 5) when coincubated directly with purified NMM-II-B filaments, both CK2 and PKC decreased filamentation and increased MgATPase activity; 6) assays done on disassembled NMM-II-B filaments showed MgATPase activity in filaments obtained from estrogen-treated cells but not estrogen-depleted cells; and 7) incubations in vitro with CK2, but not PKC, facilitated MgATPase activity, even in disassembled NMM-II-B filaments. The results suggest that estrogen, in an effect mediated by estrogen receptor-alpha and CK2 and involving the epithelial growth factor receptor and ERK/MAPK cascades, increases NMM-II-B MgATPase activity independent of NMM-II-B filamentation status.

FIG. 1

Phosphorylation of NMM-II-B filaments. A–D, Experiments in vivo (intact cells). CaSki cells were shifted for 3 d to steroid-free medium (SFM) and treated before assays with the vehicle (C, control) or 10 nM 17β-estradiol (E, 48 h), 10 μM ICI-182,780 (ICI, 24 h), DRB (6 h), or 10 μM staurosporin (STA, 8 h), alone or in combination. Cells were labeled with [³²P]orthophosphate; lysates were immunoprecipitated (IP) with the anti-NMMHC-II-B antibody, fractionated on 6% PAGE, and autoradiographed (upper panels) or were immunoblotted only with the antibody (IB) (lower panels). The experiments were repeated three times with similar trends. D shows the densitometry analysis [means ±SD, arbitrary units (A.U.)] of the data in A–C as well as experiments using cells treated with 10 μM ASO-ERα(48 h) or tamoxifen (TMX, 24 h), alone or in combination with estrogen (Est). Data were analyzed in terms of phosphorylated NMM-II-B per total NMM-II-B and normalized to densitometry of the control band (10 U). a, P < 0.01–0.05, compared with SFM, C; b, P < 0.05–0.01, compared with SFM, C, and + Est. E–I, Phosphorylation by CK2 and PKC in vitro, determined in terms of ATP hydrolysis and [³²P]Pi accumulation. E and F, Effects in vitro of CK2 (E) and PKC (F) using RRREEETEEE or the MARCKS-PSD, respectively, as substrates (added at concentrations of 1 and 5 μM, respectively). C, Control (no added kinase); HI, heat-inactivated kinases (5 min at 60 C). Experiments also involved coincubations with 50 μM DRB or 50 μM staurosporin (STA), respectively. Shown are means (± SD) of three experiments. a, P < 0.01. G–I, Effects of CK2 and PKC on phosphorylation of purified NMM-II-B filaments in vitro. G and H show dependence of CK2 (G) and PKC (H) on NMM-II-B concentration, in the absence (filled circles) and presence of inhibitors (empty circles) (DRB or staurosporin, respectively, both added at a concentration of 50 μM). Data were normalized to zero in control samples (C, no added kinase). I, Purified NMM-II-B filaments were coincubated with CK2 or PKC, in the absence or presence of 50 μM DRB or 50 μM staurosporin, respectively. Shown are means (±SD) of three experiments. a, P < 0.01.

FIG. 2

Assembly of purified NMM-II-B filaments determined in terms of the NMM-II-B critical concentration (CC). A, Experiments in vivo (intact CaSki cells, as in Fig. 1, A–C). B, Experiments in vitro, using purified NMM-II-B filaments (described in Fig. 1I). a, P < 0.02–0.01, compared with control (C).

FIG. 3

A, Steady-state ATP-dependence of ATP hydrolysis by purified NMM-II-B filaments. Insert, Time course of ATP hydrolysis. The ATP dose-related increase in [³²P]Pi release was calculated from the slower steady-state increases in [³²P]Pi release (>5 sec, insert). Data were fitted to the Michaelis-Menten equation: V = (ATP) · Vmax/KATPase + (ATP), P < 0.01 for two experiments. In the example, the calculated Vmax was 1.89 mmol Pi per (mole NMM-II-B per minute) and K_ATPase 67 μM ATP. B, Steady-state F-actin-dependent ATP hydrolysis by purified NMM-II-B filaments. Data were fit to the Michaelis-Menten equation: V = (actin) · Vmax/KATPase + (actin), P < 0.01 for two experiments. In the specific example, the calculated Vmax was 0.115/sec and the K_ATPase 37 μM F-actin. Experiments in A and B were done in low ionic (Na⁺ and Ca²⁺) strength to maintain NMM-II-B filaments in homodimerized state. C, Steady-state NaCl-dependent disassembly of NMM-II-B filaments, determined in terms of percent NMM-II-B recovered in the supernatant after incubation at 0–4 C with increasing concentrations of NaCl. The curve is means of two experiments.

FIG. 4

Steady-state F-actin-dependent ATP hydrolysis by purified NMM-II-B filaments. Vmax (filled bars) and K_ATPase (empty bars) were determined from fitting the results in each category to the Michaelis-Menten equation. A, NMM-II-B filaments were obtained from cells treated in vivo with one or more of the indicated drugs as in Fig. 1, A–C. B, Purified NMM-II-B filaments were coincubated with one or more of the indicated kinases and agents in vitro as described in Fig. 1I. Shown are means (±SD) of three to four experiments. a and b, P < 0.05–0.01, compared with control (C).

FIG. 5

NaCl-dependence of steady-state F-actin-induced ATP hydrolysis by purified NMM-II-B filaments. A, Vmax. B, K_ATPase. Experiments used NMM-II-B filaments obtained from cells treated in vivo with 17β-estradiol (E) or NMM-II-B filaments obtained from estrogen-depleted cells (C, control) that were coincubated in vitro with CK2 or PKC. Experiments were conducted in the presence of 125 mM NaCl (filled bars) or 200 mM NaCl (empty bars). Shown are means (±SD) of three experiments. a, P < 0.01, compared with C; b, P < 0.05 [125 mM NaCl vs. 200 mM and compared with control (125 mM) conditions]. UD, Undeterminable.

FIG. 6

Pharmacology and mechanisms of estrogen effects. CaSki cells were shifted for 3 d to steroid-free medium and treated with the vehicle (C, control) or one of the indicated hormones/drugs/agents, alone or in combination, and purified NMM-II-B filaments were used for the experiments. Estrogen effects on NMM-II-B filamentation were determined in terms of NMM-II-B critical concentration (CC, filled circles or bars). Estrogen effects on F-actin-dependent ATP hydrolysis were determined in terms of Vmax (empty circles or bars). A, Dose-response reactions. Cells were treated with one of the indicated concentrations of 17β-estradiol (17β-E₂) for 48 h. B, Time-response reactions. Cells were treated with 10 nM 17β-E₂ for the indicated durations of time. C, Specificity of estrogen effect. Cells were treated with 10 nM 17β-E₂ or one of the following estrogens (all for 48 h): diethylstilbestrol (DES, 10 nM); estrone (E₁, 100 nM); 17β-estradiol (17βE₂, 10 nM); and 17β-estradiol-6-[o-carboxymethyl] oxime-BSA (17β-E₂-BSA, 1 μM). D and E, Effects of ASO-ERα on ERα, ERβ, GPR30, and GAPDH steady-state mRNA levels (determined by real-time PCR) (D) and ERα, ERβ, and tubulin protein steady-state levels (determined by Western blotting) (E). Cells were treated for 48 h with 10 nM 17β-E₂ alone (C, control) or in combination with the CLO-ERα or ASO-ERα(both used at 10 μM for 48 h). In E, after blotting with the anti-ERα antibody, gels were reprobed and blotted with the antitubulin antibody. The experiments were repeated twice with similar trends. F, Effects of inhibitors on NMM-II-B filamentation and F-actin-dependent ATP hydrolysis. Cells were treated for 48 h with 10 nM 17β-E₂ alone, or plus one of the following: ASO-ERα(10 μM, 48 h); ICI-182,780 (ICI, 10 μM, 24 h); tamoxifen (TMX, 10 μM, 24 h), AG1478 (5 μM, 12 h), or PD98059 (10 μM, 6 h). Shown are means ±SD of three to six repeats. a, P < 0.01, compared with control, C (serum-free conditions); b, P < 0.05–0.01, compared with control and 17β-E₂. G and H, Estrogen effects on CK2 activity (G) and ERK1–2 activation (H). Estrogen-depleted cells were treated with 10 nM 17β-E₂. At the indicated times after start of treatment, dishes with cells were used for the assays. Shown are means ±SD of three repeats; trends were significant at P < 0.01 in both cases. mgPr, Milligram (total cellular) protein (G). In H, data were normalized to densitometry of the control band [10 arbitrary units (AU)].

FIG. 7

Effects of diC8 on PKC translocation (A) and NMM-II-B filamentation (B). A, Total homogenates (TH) or plasma membrane-enriched fractions (PM) were obtained from estrogen-depleted CaSki cells that were treated with 10 μM diC8 (30 min), 10 nM 17β-estradiol (17β-E₂) (48 h), or 15 μM PP1 (12 h), alone or in combination. Western blots were done using the specified primary antibodies against PKCα or the phosphorylated PKC (phospho-pan PKC). Experiments were repeated twice with similar trends. B, Purified NMM-II-B filaments were obtained from estrogen-depleted CaSki cells that were treated 15 min before experiments with one of the indicated concentrations of dic8. NMM-II-B filamentation (filled circles) was determined in terms of NMM-II-B critical concentration (CC), and F-actin-dependent hydrolysis of ATP (filled circles) was determined in terms of Vmax. Shown are means ±SD of three to six repeats.