Leptin induces phagocytosis of apoptotic bodies by hepatic stellate cells via a Rho guanosine triphosphatase–dependent mechanism

Authors


  • Potential conflict of interest: Nothing to report.

Abstract

Leptin, a profibrogenic cytokine, plays an important role in the development of non-alcoholic steatohepatitis. Leptin also regulates immune responses, including macrophage phagocytic activity. Stellate cells are key elements in liver fibrogenesis, and previously we have demonstrated that phagocytosis of apoptotic bodies by stellate cells is profibrogenic. To study the effects of leptin on the phagocytic activity of hepatic stellate cells, we exposed both LX-2 cells and primary stellate cells to leptin, and we have observed increased phagocytic activity. In stellate cells isolated from Zucker (fa/fa) rats, the rate of phagocytosis was significantly decreased. To investigate the mechanism by which leptin induces phagocytosis, we focused on the role of Rho-guanosine triphosphate (GTP)-ases. We found that leptin induced the PI3K-dependent activation of Rac1, and that nicotinamide adenine dinucleotide phosphate, reduced form (NADPH) oxidase activation was also implicated in this process. Leptin also induced RhoA activation and translocation to the phagosomes. Expression of the constitutive active Rac1 and RhoA both increased the phagocytic rate, whereas inhibition of the Rho-dependent kinase decreased the phagocytic activity. Conclusion: We describe a novel role of leptin in the fibrogenic process, the induction of phagocytosis of apoptotic bodies by hepatic stellate cells. The data provide strong evidence of a Rho-GTPase–mediated regulation of the cytoskeleton during stellate cell phagocytosis. Leptin-mediated phagocytic activity of stellate cells therefore could be an important mechanism responsible for progression of fibrosis in non-alcoholic steatohepatitis. (HEPATOLOGY 2008.)

Leptin, a 16-kDa peptide, plays a key role in the regulation of body weight.1 It also has an important role in the development of non-alcoholic steatohepatitis, which is characterized by the presence of the metabolic syndrome and progressive steatosis, inflammation, and fibrosis in the liver. Hepatic stellate cells (HSC) are key elements of the fibrogenic process. They are normally quiescent and produce only small amounts of extracellular matrix components. After activation by cytokines or reactive oxygen species, they undergo a morphological and functional transition to myofibroblast-like cells, with the subsequent production of transforming growth factor-β1 (TGF-β1) and extracellular matrix.2 Leptin has been shown to be critical in the development of hepatic fibrosis, because leptin-deficient ob/ob mice do not develop liver fibrosis after CCl4 injury.3 Furthermore, lack of the leptin receptor, Ob-R, prevents fibrogenesis induced by bile duct ligation in fa/fa (Zucker) rats.4 Because leptin is essential in the induction of liver fibrosis, it is important to better understand the mechanism(s) by which leptin regulates this process. Although the fibrogenic properties of leptin are postulated to be from the up-regulation of procollagen α1(I),3, 5 mitogenesis and the inhibition of HSC apoptosis,6 the precise mechanism by which leptin promotes liver fibrosis remains undetermined.

Recently, we have shown that HSC phagocytosing apoptotic bodies (ABs) from dying hepatocytes undergo activation and a major fibrogenic response with the production of oxidative radicals, as well as the up-regulation of procollagen α 1(I) and TGF-β1 gene expression.7 Because phagocytosis by HSC is likely to be a common pathway leading to fibrosis, independent of the cause of the liver disease, factors enhancing the phagocytic process could further trigger activation of HSC and consequently liver fibrogenesis. Leptin-induced phagocytosis requires binding of leptin to the leptin receptor, as leptin replacement in ob/ob mice normalized phagocytic activity.8 These observations led us to postulate that leptin may promote fibrogenesis through enhanced phagocytosis of ABs by HSCs. How leptin plays a role in phagocytosis is still unclear. Our speculation is that it may involve recruitment or activation of cytoskeleton-regulating elements, such as Rho guanosine triphosphatases (GTPases), which play an active role in actin cytoskeleton remodeling, a requisite step in phagocytosis. Interest in the Rho family of small GTPases is compounded by a recent work demonstrating that they play an integral role in HSC activation and the fibrogenic process.9

In the current study, we have described a novel profibrogenic activity of leptin; the induction of phagocytosis of ABs by HSCs. We found that leptin elicits its prophagocytic activity by the regulation of small GTP-binding proteins. Leptin activated Rac1 via the phosphatidylinositol-3-kinase, and also induced activation and translocation of RhoA to the phagosomes, and subsequent activation of the Rho-associated kinase (ROCK). These changes translated into increased phagocytic activity of HSC.

Abbreviations

AB, apoptotic body; GTP, guanosine triphosphate; HSC, hepatic stellate cell; JAK, Janus kinase; NADPH, nicotinamide adenine dinucleotide phosphate, reduced form; Ob-R, leptin receptor; PBS, phosphate-buffered saline; ROCK, Rho-associated kinase; SDS, sodium dodecyl sulfate; SE, standard error; STAT, signal transducer and activator of transcription protein; TAMRA, carboxytetramethyl rhodamine succinimidyl ester; TGF-β, transforming growth factor beta; WT, wild-type; SOC3, suppressor of cytokine signaling.

Materials and Methods

Animals.

Male Zucker lpar;fa/fa) and wild-type (WT) rats from Charles River Laboratories Inc. (Wilmington, MA) were used for primary HSC isolation. The animals were housed in facilities approved by the National Institutes of Health. All procedures were reviewed and approved by the Animal Welfare Committee of the University of California Davis.

Cell Culture and Preparation of Apoptotic Bodies.

The human immortalized HSC line LX-2 (provided by Dr. S.L. Friedman, Mount Sinai Medical School, New York, NY)10 and primary rat HSCs isolated from Zucker and WT rats were used. Primary HSCs were isolated from rats as previously described7 and used 2 days after isolation. The purity of isolated HSCs was assessed by the autofluorescence of vitamin A droplets, and was 95% or greater. LX-2 cells exhibit typical features of HSCs in primary culture, such as expression of desmin and glial acidic fibrillary protein, and responsiveness to TGF-β. HSCs were maintained in Dulbecco's modified Eagle's medium supplemented with 5% fetal bovine serum, 100 U/mL penicillin, and 100 g/mL streptomycin and incubated at 37°C in a humidified atmosphere with 5% CO2.

For preparation of ABs, the human hepatoma cell line HepG2 was used, cultured in Dulbecco's modified Eagle's medium and 10% fetal bovine serum. ABs were generated by exposing the cells to ultraviolet irradiation (100 mJ/cm2, 142 seconds), as described preciously.7 ABs were then collected 48 hours after the ultraviolet irradiation. To generate labeled ABs, HepG2 cells were incubated with carboxytetramethyl rhodamine succinimidyl ester (TAMRA) (Invitrogen, Carlsbad, CA) before the ultraviolet irradiation, as described previously.7

Phagocytosis Assay.

LX-2 or primary rat HSCs were seeded in six-well dishes at a density of 2.5 × 105/mL. The cells were then cultured in serum-free medium for 16 hours and then treated with the following reagents: leptin (100 ng/mL, R&D Systems Inc., Minneapolis, MN), leptin plus AB, the PI3 kinase inhibitor LY294002 (20 μM, EMD Chemicals Inc., Gibbstown, NJ) plus AB for 6 hours, or the ROCK inhibitor Y27632 (10 μM, Sigma, St. Louis, MO) plus AB. The concentration of leptin used was based on prior rodent studies on liver fibrogenesis.6 TAMRA-labeled ABs (approximately 105/mL were added to HSCs at 37°C). Intracellular ABs were detected by TAMRA labeling after 48 hours, by fluorescent microscopy. The rate of phagocytosis was assessed by dividing the number of TAMRA-positive HSCs by the total number of HSCs counted.

Complementary DNA Transfection.

Complementary DNA encoding constitutive active Rho (RhoQ63L) and constitutive active and dominant-negative Rac (RacQ61L and RacT17N, respectively), originally generated by Dr. J. Silvio Gutkind (National Institute of Dental and Craniofacial Research, National Institutes of Health, Bethesda, MD), were used. pCMV6-XL6-SOCS3 (Origene, Rockville, MD), RacQ61L, RacT17N, RhoQ63L, and control vector were transfected into LX-2 cells (transfection efficiency, 60%-70%), using the lipofectamine reagent (Invitrogen, Carlsbad, CA) according to the manufacturer's recommendation.

Rac1- and Rho GTP Pull Down Assays.

LX-2 cells were exposed to ABs in the presence or absence of LY294002; then an affinity precipitation assay was performed to measure the active GTP-bound Rac1, according to the manufacturer's instructions (Millipore, Billerica, MA). Briefly, the cells were lysed with the buffer multilayered buffer containing 10% glycerol and 1 mM phenylmethylsulphonyl fluoride and agitated at 4°C for 15 minutes followed by centrifugation at 14,000g at 4°C for 5 minutes. The supernatant containing 20 μg protein was mixed with beads coated with p21-activated kinase containing a Rac-binding domain (p21-binding domain). The mixture was rotated for 45 minutes at 4°C, and the agarose beads were washed and collected by brief centrifugation. The beads were then resuspended in the sample buffer with 50 mM dithiothreitol and separated by sodium dodecyl sulfate (SDS)-polyacrylamide gel, and Rac1 was detected by a specific antibody provided by the kit. For the Rho-GTP pull-down assay, LX-2 cells and HSCs from WT and Zucker rats were exposed to ABs for different times as described previously; in the case of LX-2 cells in the presence or absence of the Janus kinase (JAK) inhibitor AG490 (50 μM, EMD Chemicals Inc., Gibbstown, NY), or LY294002, as described previously. For the Rho activity assay, the supernatant was collected and mixed with glutathione-agarose beads bound with Rhotekin (Millipore, Billerica, MA), containing a Rho-binding domain. The mixture was rotated for 45 minutes at 4°C, and the agarose beads were washed and collected by centrifugation, then resuspended in the sample buffer and separated by SDS-polyacrylamide gel. The GTP-bound RhoA was detected by a specific antibody provided by the kit.

Western Blot Analysis.

To detect JAK/signal transducer and activator of transcription protein (STAT) phosphorylation, LX-2 cells were incubated in serum-free medium then either mock transfected with pCMV6-XL6 or with pCMV6-XL6-SOC3. They were exposed to ABs for 0.5 hour then collected. In brief, the cells were washed with 1 mM Na3VO4/phosphate-buffered saline (PBS) and collected into the lysis buffer containing 50 mM dithiothreitol, 1 mM Na3VO4, 1 mM NaF (pH 8.0), 5 mM phenylmethylsulphonyl fluoride, and 1% SDS. The lysates were centrifuged at 12,000g for 5 minutes at 4°C. The supernatant was collected, and the protein concentration was determined with the Bio-Rad protein assay kit (Bio-Rad Laboratories, Hercules, CA) following the instruction from the manufacturer, and 10 to 20 μg protein was separated by SDS-polyacrylamide gel electrophoresis. The proteins from the gel were then electroblotted (Bio-Rad Mini-Protean II transblot system) onto nitrocellulose paper (Bio-Rad) and blocked by 5% nonfat milk powder in Tris-buffered saline-Tween (20 mM tris-base, 137 mM NaCl, 0.1% Tween 20, pH 7.6), followed by incubation with the appropriate antibodies for 16 hours at 4°C: anti-phospho-JAK1 polyclonal antibody (1:500, EMD-Calbiochem), JAK1 antibody (1:300; Santa Cruz Biotechnology, Santa Cruz, CA), anti-phospho-STAT3 monoclonal antibody (1:250; Cell Signaling Technology, Danvers, MA), and anti-STAT3 polyclonal antibodies (Santa Cruz, CA).

RhoA Translocation Assay.

To determine the membrane-bound (GTP-ase active form) of RhoA, LX-2 cells were serum starved for 16 hours and exposed to leptin or ABs. To obtain membrane and cytosolic fractions, the cells were washed and homogenized in a buffer containing 50 mM Tris-HCl, pH 7.4, 1 mM ethylene glycol tetraacetic acid, 1 mM ethylene diamine tetraacetic acid, 10 μg/mL leupeptin, 10 μg/mL aprotinin, 5 mM benzamidine HCl, 10 μg/mL soybean trypsin inhibitor, and 1 mM phenylmethylsulfonyl fluoride by ultrasonic polytron. The cell extract was centrifuged at 800g for 5 minutes. The supernatant was then collected and further centrifuged at 15,000g for 15 minutes at 4°C. The supernatant was centrifuged again at 100,000g for 1 hour at 4°C, and then the supernatant was collected as the cytosolic fraction, and the pellet, as the membrane fraction, was then resuspended. The protein concentration was measured; and 10 μg cytosol and 20 μg membrane fractions were used to perform western blot analysis for RhoA.

Immunohistochemistry.

LX-2 cells grown in chamber slides were exposed to TAMRA-labeled ABs for 5 hours in the presence or absence of myosin light chain kinase inhibitor ML-7 (5 μM, EMD Chemicals Inc., Gibbstown, NJ) for 30 minutes. The cells were fixed in 4% paraformaldehyde/PBS for 15 minutes at room temperature, then were washed in PBS and blocked with 2% bovine serum albumin in PBS for 1 hour followed by incubating with anti-RhoA monoclonal antibody (1:40) for 16 hours at 4°C, (Santa Cruz Biotechnology, Santa Cruz, CA). After washing with PBS, the secondary, Alex Fluor 488–labeled anti-mouse immunoglobulin G (1:1000) (Invitrogen) was applied. The images were analyzed by fluorescence microscopy.

Results

Leptin Increases Phagocytosis of Apoptotic Bodies by HSC.

To test whether leptin exposure induces phagocytosis of ABs by HSCs, and to ascertain that the effect on phagocytosis is leptin receptor specific, HSCs were isolated from WT and Zucker rats. In HSCs from Zucker rats, the rate of phagocytosis was significantly lower compared with WT HSCs (Fig. 1). Leptin treatment significantly increased the rate of phagocytosis in HSC from WT rats, whereas no increase was detected in HSCs from Zucker rats. This confirms that leptin receptor–mediated signaling is required for the induction of phagocytosis. Next, to elucidate the mechanism responsible for induction of phagocytosis by leptin, we focused on the role of Rho-GTPases.

Figure 1.

Primary HSCs from WT and Zucker rats were treated with AB, leptin plus AB. The phagocytosis of WT HSCs was induced by ABs to 15.1-fold (±2.41)-fold, and leptin increased this response to 20.3-fold (±1.91). HSCs from Zucker rats exhibited only a 6.04-fold (±1.89) increase in phagocytic activity, and leptin did not significantly change the phagocytic rate. Data are expressed as mean ± standard error (SE), N = 4. *P < 0.01. Inset depicts control HSC (WT) and HSC phagocytosing TAMRA-labeled ABs.

Leptin Induces PI3K-Dependent Activation of Rac1, and Induction of Phagocytosis.

Because Rac1 is a crucial element of the phagocytic machinery,11 we first focused on the role of this GTPase in leptin-induced phagocytosis. Because leptin may mediate its effects on the cytoskeleton through the PI3 kinase, we exposed LX-2 cells to leptin (100 ng/mL) or ABs, in the presence or absence of the PI3 kinase inhibitor LY294002 (20 μM). Because Rac1 is a major subunit of the reduced nicotinamide adenine dinucleotide phosphate (NADPH) oxidase complex, and NADPH oxidase activation occurs during phagocytosis of ABs by HSCs,7 we also tested whether using the NADPH oxidase inhibitor apocynin had an effect on Rac1 GTPase activity. Leptin induced activation of Rac1, and this was inhibited by the PI3 kinase inhibitor LY294002 (Fig. 2A). Inhibiting NADPH oxidase with apocynin (100 μM) also partially inhibited leptin-induced Rac1 activation, suggesting that Rac1 activation during phagocytosis occurs in concert with NADPH oxidase activation. Because we demonstrated that leptin induces Rac1 GTPase activity through the PI3 kinase, next we tested whether inhibition of the PI3 kinase has an effect on the phagocytic rate. In leptin-treated LX-2 cells, inhibition of PI3 kinase indeed decreased the rate of phagocytosis by 0.28-fold; inhibiting the NADPH oxidase by apocynin caused a further decrease by 0.53-fold. This suggests that leptin mediates its effects on phagocytosis by the PI3 kinase-dependent activation of Rac1, and that NADPH oxidase activity is also required for Rac1 induction during phagocytosis (Fig. 2B).

Figure 2.

Leptin induces the PI3K-dependent activation of Rac1 and increase in phagocytosis. (A) LX-2 cells were exposed to leptin (100 ng/mL) or ABs, in the presence or absence of the PI3 kinase inhibitor LY294002 (20 μM), and the NADPH oxidase inhibitor apocynin (100 μM). Western blot and densitometry showed that leptin induced activation of Rac1, and this was inhibited by the PI3 kinase inhibitor LY294002. Inhibiting NADPH oxidase with apocynin also partially inhibited leptin-induced Rac1 activation. N = 4, *P < 0.01. (B) Inhibition of PI3K decreased the rate of phagocytosis by 0.28-fold (±0.09) in leptin-treated LX-2 cells, and inhibiting NADPH oxidase by apocynin caused a 0.53-fold (±0.14) decrease compared with leptin-treated cells. Mean ± SE, N = 4, *P < 0.01, **P = 0.009.

Overexpression of Constitutively Active Rac1 Induces Phagocytosis of Apoptotic Bodies by HSC.

Because of the key role of the Rho family of small GTPases in actin remodeling, we next tested the role of this family of proteins in leptin-based phagocytosis by using complementary gain and loss of function approaches. Overexpression of the constitutive active Rac1 (RacQ61L) indeed was sufficient to induce phagocytosis of ABs by HSCs. LX-2 cells transfected with RacQ61L engulfed more ABs, whereas transfection of the dominant negative mutant (RacT17N) resulted in a decrease in phagocytosis compared with the RacQ61L-transfected cells; however, still it was above the baseline level (only empty vector-transfected cells, without ABs), suggesting that Rac1 inhibition alone is not sufficient to block phagocytosis of ABs (Fig. 3). This suggests that there are other signaling pathways induced by phagocytosis, likely involving the activation of RhoA.

Figure 3.

Overexpression of constitutive active Rac1 induces phagocytosis of apoptotic bodies by HSCs. LX-2 cells transfected with Rac1 Q61L engulfed more apoptotic bodies (1.97-fold, ±0.17), compared with mock-transfected cells exposed to ABs. Transfection of the dominant negative mutant (T17N) resulted in decrease of phagocytosis compared with the Rac1Q61L-transfected cells, however, still above the control level (control, only pcDNA3.1-transfected cells). Mean ± SE, N = 4, *P < 0.01.

Leptin Treatment Induces RhoA Activation in Primary HSCs Phagocytosing ABs.

RhoA, another actin-cytoskeleton regulating GTPase, is also known to be involved in phagocytosis. To study its activation by leptin, pull-down of active GTP-RhoA by the Rhotekin assay was performed. In addition to LX-2 cells, HSCs from leptin receptor (Ob-RL)–deficient Zucker rats (fa/fa) and WT rats were used to monitor for leptin specificity of RhoA activation. In HSCs from WT rats, leptin induced activation of RhoA, whereas in HSC from Zucker rats exposed to leptin, no increase in activation of RhoA was detected (Fig. 4A). In addition, RhoA GTPase activity was dependent on JAK signaling, because the JAK inhibitor AG490 inhibited leptin-induced RhoA activity; however, inhibition of the PI3 kinase did not affect it (Fig. 4B). To confirm that engulfment of ABs induced JAK1 and STAT3 phosphorylation, western blot analyses were performed. Transfection with a SOCS3 expressing plasmid diminished JAK phosphorylation and consequently STAT3 phosphorylation (Fig. 5A). To confirm that RhoA activation is indeed reduced by JAK inhibition, RhoA pull-down assay was performed in SOCS3 expressing HSCs after exposure to either leptin or ABs. We found that both leptin and AB-induced activation of RhoA have decreased in SOCS3 expressing stellate cells, corroborating our previous data (Fig. 5B).

Figure 4.

Leptin induces RhoA GTPase activity in primary HSCs and in LX-2 cells. (A) HSCs were isolated from WT and Zucker rats, cultured for 2 days, and then exposed to leptin (100 ng/mL) or ABs. Rho GTP pull-down assays and western blot analyses were performed to detect active RhoA. In HSCs isolated from WT rats, leptin induced activation of RhoA, whereas in HSCs from Zucker rats exposed to leptin, no increase in activation of RhoA was detected. N = 3. (B) LX-2 cells were exposed to leptin in the presence or absence of JAK inhibitor AG490, or PI3 kinase inhibitor LY294002, and RhoA GTPase activity was studied. Western blot and densitometry showed that RhoA GTPase activity was dependent on JAK signaling because AG490 inhibited leptin-induced RhoA activity; however, inhibition of the PI3K did not affect it. Mean ± SE, N = 4, *P < 0.01.

Figure 5.

Phagocytosis of apoptotic bodies by HSCs induces JAK1 and STAT3 phosphorylation and RhoA GTPase activity. (A) LX-2 cells were cultured in serum-free medium, mock-transfected (only pCMV-XL6), or transfected with the SOCS3 construct, and 48 hours later exposed to ABs for half an hour. Western blot analyses were performed to detect JAK1 and STAT3 phosphorylation. Transfection of LX-2 cells with the SOCS3 construct inhibited JAK1 and consequently STAT3 phosphorylation. (B) RhoA pull-down assay was performed in SOCS3 or mock-transfected LX-2 cells after exposure to either leptin or ABs. RhoA activity decreased in SOCS3-transfected cells compared with control (mock transfected cells) after leptin, and especially after AB treatment.

Overexpression of Constitutive Active RhoA Induces Phagocytosis of Apoptotic Bodies by HSC.

To establish whether expression of the constitutive active RhoA had an effect on the phagocytic rate, we transfected LX-2 cells with RhoQ63L and exposed them to ABs. An increase in phagocytosis was detected, over empty vector-transfected cells exposed to AB (Fig. 6A). Treating HSCs with C3, which inhibits adenosine diphosphate ribosylation of RhoA, caused a significant decrease in the engulfment of AB, compared with leptin and AB-exposed cells (Fig. 6B). Thus, RhoA in our system is a positive regulator of phagocytosis.

Figure 6.

Overexpression of constitutive active RhoA induces phagocytosis of apoptotic bodies by HSCs, whereas inhibition of RhoA GTPase activity decreases it. (A) In LX-2 cells transfected with the constitutively active RhoA (Q63L), an increase in phagocytosis was detected (1.75-fold), compared with empty-vector–transfected cells exposed to ABs. Mean ± SE, N = 4, *P < 0.01. (B) HSCs were exposed to ABs and leptin in the presence or absence of C3, an inhibitor of adenosine diphosphate ribosylation of RhoA. A 0.49-fold (±0.14) decrease in phagocytosis was detected, compared with leptin and AB-exposed cells. Mean ± SE, N = 4, *P < 0.01

Leptin Treatment and Phagocytosis of AB Induces RhoA Translocation to Phagosomes and Activation of ROCK.

Because RhoA is known to play a key role in phagocytosis, next we tested whether RhoA translocation to the phagocytic cup occurs on leptin stimulation. In these experiments, LX-2 cells were exposed to TAMRA-labeled ABs, and immunohistochemistry to detect RhoA was performed. In control cells, RhoA is cytoplasmic (Fig. 7A). In HSCs phagocytosing ABs, RhoA translocates to the phagosomal membrane (Fig. 7B, arrow), the AB labeled with TAMRA (inset). On leptin stimulation, there is an increase in phagocytosis of ABs, and again, RhoA mainly localizes to the phagosomal membrane (Fig. 7C). ML-7, an inhibitor of the myosin light chain kinase, inhibits the translocation of RhoA to the membrane; therefore, the signal is cytoplasmic (Fig. 7D). To confirm these data, membrane fractions were extracted from leptin, or AB-treated HSCs, and western blot analysis was performed to detect RhoA translocation. Leptin induced translocation of RhoA to the membrane fraction, as did exposure to ABs (Fig. 7E). This suggests that, besides activation of the RhoA GTPase, its translocation to the phagocytic cup also occurs during leptin-induced phagocytosis.

Figure 7.

Leptin induces translocation of RhoA to phagosomes and ROCK activation. LX-2 cells were exposed to ABs, leptin plus AB in the presence or absence of ML-7 (ABs were labeled previously with TAMRA, red). (A) Immunofluorescence was performed, and RhoA (green) was seen in the cytoplasm of control cells, and (B) translocating to the phagosomes (inset) in AB, or (C) L+AB treated cells (arrowheads). In the presence of leptin, more RhoA-positive phagosomes were seen. (D) After treatment of a myosin light chain kinase inhibitor ML-7, less phagocytosis was observed, and RhoA failed to localize to the phagosomal membrane around AB. (E) Western blot assay was performed to assess membrane-bound RhoA. RhoA protein was increased in the membrane fraction and decreased in the cytosol after exposure to either AB or leptin. GAPDH served as an equal loading control. (N = 3.) (F) Phagocytosing LX-2 cells were exposed to the ROCK inhibitor Y27632 (10 μM) before exposing them to ABs, and this decreased the phagocytic rate by 0.53-fold when compared with that without the inhibitor. Mean ± SE, N = 3, *P < 0.05.

To mechanistically extend the role of RhoA in our model, next we determined whether its downstream target ROCK plays a role in the phagocytosis of ABs. HSCs were exposed to the ROCK inhibitor Y27632, before phagocytosis, and this indeed decreased the phagocytic rate by 0.53-fold (Fig. 7F), suggesting that RhoA mediates its effects on phagocytosis through ROCK activation.

Discussion

Phagocytosis of AB by HSC is an important fibrogenic mechanism in the liver.7, 12 We have previously shown that phagocytosis in HSC induces the activation of NADPH oxidase and up-regulation of procollagen α 1(I) and TGF-β expression. The mechanism of HSC activation by phagocytosis is likely to be a common pathway linking chronic liver injury, apoptosis, and fibrogenesis in the liver, independent of the cause of the liver disease. Therefore, studying factors that regulate phagocytosis is of great importance, because enhanced phagocytic activity of HSCs could lead to accelerated liver fibrogenesis. Leptin has been postulated to play an important role in the phagocytic process, because leptin-deficient ob/ob mice exhibited impaired response to Gram-negative pneumonia,13 and exogenous leptin was shown to up-regulate both phagocytosis by macrophages and production of inflammatory cytokines.8 Thus, it was plausible that leptin, one of the major profibrogenic cytokines, may play a role in the phagocytic process in HSCs, thereby further enhancing their fibrogenic activity. Indeed, we found that leptin induced the phagocytic activity of HSCs, and that the phagocytic activity was dependent on signaling through the Ob-R, because HSCs from Zucker rats deficient in this receptor failed to have an increase in the phagocytic rate. Although the increase in the phagocytic activity by leptin at first sight seems to be moderate, induction of different signaling pathways simultaneously could translate into an amplified fibrogenic response [NADPH oxidase activation, procollagen α1 (I) up-regulation]. The next question is how leptin regulates phagocytosis. Because Rho-GTP-ases (Rac and Rho) are known to be important regulators of the phagocytic process,11, 14 we focused on how leptin may modulate their GTPase activity. Leptin induces the GTPase activity of Rho proteins in migratory cells15; therefore, we postulated that leptin may stimulate phagocytosis in HSC through activation of the Rho-GTPases. Rac GTPases are known to control actin polymerization into lamellipodial and filopodial membrane protrusions.11 Rac1 is ubiquitously expressed and is likely to be the main Rac GTPase in nonhematopoietic cells.16 In addition to regulating cytoskeletal function during phagocytosis, Rac GTPases are also important elements of the NADPH oxidase complex by participating in the recruitment of the p67phox subunit to the enzyme complex on activation during phagocytosis.16 Based on these concepts, we first determined whether leptin has an effect on Rac1 GTPase activity. We found that Rac1 activity was induced by leptin by the PI3 kinase, and using a PI3 kinase inhibitor decreased the phagocytic rate. In addition, Rac1 activity was also dependent on the NADPH oxidase, because inhibition of the enzyme by apocynin caused decreased GTPase activity of Rac1, suggesting that a leptin/NADPH oxidase/Rac1 pathway is also involved in phagocytosis. This is in keeping with a recent report confirming that NADPH oxidase is a downstream mediator of leptin signaling in HSCs.17, 18 Apocynin also can directly affect phagocytosis in different systems (albeit at a much higher concentration), where the production of reactive oxygen species was required for phagocytosis to occur.19 Expressing the constitutive active form of Rac1 in HSCs increased phagocytosis of ABs, whereas the dominant negative form decreased it, indicating that Rac1 is indeed a positive regulator of phagocytosis in HSCs. The observation that the phagocytic rate was not reduced to the baseline by the dominant negative Rac1 (Fig. 3) suggested the existence of additional pathway(s) by which phagocytosis of ABs was regulated. To examine this possibility, we studied the role of RhoA. RhoA is known to facilitate the assembly of contractile actomyosin filaments by activation of ROCK.20–22 The substrate of ROCK, myosin light chain phosphatase, is involved in the regulation and assembly of actin-myosin filament bundles during phagocytosis.23 We found that leptin induced GTPase activity and also translocation of RhoA to the phagosomes. The expression of the constitutive active RhoA in HSCs increased the phagocytic rate, whereas inhibiting adenosine diphosphate ribosylation with C3 decreased it. In previously published studies, active Rac1 and RhoA played opposite roles in regulating phagocytosis in macrophages, Rac 1 stimulating and RhoA inhibiting it through ROCK.14 However, in our studies, ROCK inhibition resulted in a decreased phagocytic rate, which may translate to decreased fibrogenic activity based on our previous study. This is supported by earlier findings in which ROCK inhibition decreased HSC activation24 and was antifibrogenic in the liver and kidney.25, 26 The difference concerning the role of active RhoA on phagocytosis may result also from the fact that HSCs are considered to be nonprofessional phagocytes, and as such, their mechanism of phagocytosis is not well characterized. For instance, RhoA was shown to positively regulate integrin-mediated phagocytosis in macrophages, and inhibition of RhoA by C3 resulted in decreased uptake of serum-opsonized zymosan particles.27, 28 Whether integrins, especially the β2 subclass, play a role in phagocytosis by HSCs remains to be determined. It is known, however, that integrin-mediated signaling is a crucial element in HSC activation and liver fibrogenesis.29, 30 Thus, it is plausible that phagocytosis-induced signaling events in HSCs are integrin-mediated and result in RhoA/ROCK activation, actin cytoskeleton rearrangement, and phagocytic cup formation. Interestingly, we also found that RhoA translocated to the phagosomes, where it also may play a role in NADPH oxidase-induced superoxide formation during phagocytosis.31

In conclusion, our data suggest a new model in which intact leptin signaling is required to maintain and augment the phagocytic process in HSCs and the clearance of AB during chronic liver injury. It is plausible that leptin deficiency in ob/ob mice could result in a decrease in phagocytosis of apoptotic cells and a consequent increase in inflammatory activity in the liver.

In our study, leptin-induced Rac1 and RhoA activation are responsible for the increase in the phagocytic rate (Fig. 8). Because previously we have shown that phagocytosis of AB by HSC induces their activation and fibrogenesis,7 further induction of phagocytosis by leptin and activation of Rac1 and RhoA could exacerbate the fibrogenic process. Indeed, a recent study showed that sustained Rac1 activation leads to accelerated liver fibrosis in mice.9 Thus, regulation of Rac1 and RhoA seem to be critical steps in the fibrogenic process. Because leptin is a major profibrogenic cytokine, induction of the phagocytic activity by HSCs could be an important mechanism of liver fibrosis in nonalcoholic steatohepatitis.

Figure 8.

Mechanisms of leptin-induced phagocytosis in HSCs. Leptin induces Rac1 GTPase activity through the PI3 kinase with the involvement of NADPH oxidase. In addition, activation of RhoA GTPase by JAK also occurs, and this results in translocation of RhoA to the phagosome and activation of ROCK.

Ancillary