July 2004
Volume 45, Issue 7
Free
Glaucoma  |   July 2004
Role of Lysophospholipid Growth Factors in the Modulation of Aqueous Humor Outflow Facility
Author Affiliations
  • Priyatham Sai Mettu
    From the Departments of Ophthalmology,
  • Pei-Feng Deng
    From the Departments of Ophthalmology,
  • Uma K. Misra
    Pathology, and
  • Govind Gawdi
    Pathology, and
  • David L. Epstein
    From the Departments of Ophthalmology,
  • P. Vasantha Rao
    From the Departments of Ophthalmology,
    Pharmacology and Cancer Biology, Duke University Medical Center, Durham, North Carolina.
Investigative Ophthalmology & Visual Science July 2004, Vol.45, 2263-2271. doi:10.1167/iovs.03-0960
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Priyatham Sai Mettu, Pei-Feng Deng, Uma K. Misra, Govind Gawdi, David L. Epstein, P. Vasantha Rao; Role of Lysophospholipid Growth Factors in the Modulation of Aqueous Humor Outflow Facility. Invest. Ophthalmol. Vis. Sci. 2004;45(7):2263-2271. doi: 10.1167/iovs.03-0960.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

purpose. To investigate the role of lysophospholipid growth factors in the regulation of aqueous humor outflow in the trabecular meshwork (TM).

methods. The expression profile of the endothelial differentiation gene (Edg) family of G-protein coupled receptors was determined by RT-PCR of human TM (HTM) cell–derived total RNA and by PCR amplification of HTM cell–derived and tissue–derived cDNA libraries. The effects of lysophosphatidic acid (LPA) and sphingosine-1-phosphate (S1P) on actin cytoskeleton and focal adhesions and on myosin light-chain (MLC) phosphorylation in HTM cells were evaluated by immunofluorescence microscopy and Western blot analysis, respectively. Activation of Rho GTPase in HTM cells was quantified by “pull-down” assays. Mobilization of intracellular calcium in HTM cells was determined using spectrofluorometric digital-imaging microscopy. The effects of LPA and S1P on aqueous humor outflow facility were evaluated by perfusion of enucleated porcine eyes.

results. Each of the receptor isoforms Edg1, -2, -3, and -4 was readily detectable in three of four HTM cell–derived libraries, whereas Edg2 was detectable in the HTM tissue library. LPA (20 μM) and S1P (1 μM) stimulated actin stress fiber and focal adhesion formation, increased MLC phosphorylation, and induced marked activation of Rho GTPase in HTM cells. Both LPA (20 μM) and S1P (10 μM) also stimulated increases in intracellular calcium concentration in HTM cells. LPA- and S1P-induced effects on MLC phosphorylation in HTM cells were markedly inhibited by pretreatment with the Rho kinase–specific inhibitor Y-27632 (5 μM). Perfusion of LPA (50 μM) and S1P (5 μM) in enucleated porcine eyes produced a significant decrease in aqueous humor outflow facility from baseline of 37% (n = 6) and 31% (n = 5), respectively.

conclusions. These studies demonstrate that LPA and S1P, the physiological agonists of Edg receptors, decrease outflow facility in perfused porcine eyes in association with increased MLC phosphorylation and Rho guanosine triphosphatase (GTPase) activation. These data provide evidence for a novel mechanism for negative regulation of outflow facility, which may contribute to overall physiological homeostasis of aqueous humor outflow facility.

Understanding of the physiological regulation of aqueous humor outflow facility through the trabecular meshwork (TM) is fundamental to the investigation of glaucoma. Such knowledge could provide valuable insight into the pathophysiological abnormalities that lead to impaired outflow and the consequent elevation in intraocular pressure (IOP) that commonly characterizes glaucomatous disease. 1 2 Numerous studies have centered on understanding how cells of the outflow pathway interact with one another and with the surrounding extracellular matrix (ECM) substratum to generate resistance to aqueous humor outflow. 3 4 5 6 7 8 9 10 11 12 In understanding TM cellular behavior, it has become apparent that the interplay of the cell cytoskeleton together with the ECM provides a structural framework for the outflow pathway and thus influences the filtering capacity of the TM. 3 4 13 Thus, elucidating the mechanisms that regulate these features has become an integral focus for efforts directed toward both glaucoma pathophysiology and medical treatment. 14 15 16  
Despite efforts devoted to identifying and characterizing the nature of signaling molecules at the TM that may serve as potential pharmacological targets, 5 6 9 11 the precise roles of intracellular signaling pathways and extracellular mediators of such pathways in the modulation of aqueous outflow facility have not been well defined. However, several investigations have focused on the identification of secretory and regulatory proteins that might be implicated in some manner in glaucomatous disease. 17 18 In an effort to identify extracellular factors in the aqueous that might influence outflow facility, we fractionated and characterized samples of porcine aqueous humor, screening these fractions for bioactivity in the modulation of porcine TM cell morphology and cytoskeletal organization (Epstein DL, et al. IOVS 2001;42:ARVO Abstract 746). One fraction that displayed such activity was both heat stable and protein bound, suggesting that lipid factors may well be involved in influencing outflow facility. 
In keeping with our previous and ongoing work to explore the signaling mechanisms integral to the functional regulation of the outflow pathway, we have postulated that such lipid-based molecules might act as extracellular first messengers. One such class of molecules is the lysophospholipid growth factors, which have been well characterized as physiological agonists of the endothelial differentiation gene (Edg) family of G-protein coupled receptors. Indeed, there has been a growing interest in exploring the role of such factors in normal physiology as well as within the context of disease processes. 19 20 21 22 23 The two most well-characterized factors among this class are lysophosphatidic acid (LPA) and sphingosine-1-phosphate (S1P). Although each of these mediators was initially considered only as an intracellular second messenger, more recent work has identified each of them as a ligand (LPA: Edg2, -4, and -7; S1P: Edg1, -3, -5, -6, and -8) of Edg receptors, which are widely distributed throughout a number of tissues. 19 21 LPA and S1P have been shown to elicit and modify a diverse array of cellular responses, including cell migration and proliferation, protection from apoptosis, and, notably, alterations in cell morphology and cytoskeletal architecture. 19 20 Although lysophospholipids have been detected in a number of biological fluids, including aqueous humor, 24 25 there is currently little information on the significance of lysophospholipids/Edg signaling in the TM. Therefore, the goal of this study was to investigate the effects of the lysophospholipids LPA and S1P in the TM and to elucidate the nature of the signaling mechanisms that are integral to the regulation of TM function with respect to outflow facility. 
Materials and Methods
Human TM (HTM) cell and tissue cDNA libraries were a gift of Pedro Gonzalez of Duke University (Durham, NC). Rabbit polyclonal antibody directed against myosin light-chain (MLC) was a gift of Joe G. Garcia of Johns Hopkins University (Baltimore, MD). LPA, fatty acid-free bovine serum albumin (BSA), fura-2/acetoxymethyl ester (fura-2/AM), thapsigargin, rhodamine-phalloidin, anti-vinculin monoclonal antibody, and TRITC-conjugated goat anti-mouse IgG antibody were purchased from Sigma-Aldrich (St. Louis, MO). GF-109203X (protein kinase C [PKC] inhibitor) was from Alexis Biochemicals (Lausen, Switzerland). S1P was from Biomol Research Laboratories (Plymouth Meeting, PA). ML-7 (MLC kinase inhibitor) was purchased from Calbiochem (San Diego, CA). The Rho kinase inhibitor Y-27632 was obtained from Welfide Corp. (Osaka, Japan). Rabbit polyclonal antibody directed against di-phospho-MLC (Thr18/Ser19) was from Cell Signaling Technology (Beverly, MA). Rho GTPase activation assay kit was obtained from Upstate Biotechnology (Lake Placid, NY). Oligonucleotide primers, for Edg1 (5′-ACG TCA ACT ATG ATA TCA TCG TCC G; 3′-CAT TTT CAG CAT TGT GAT ATA GCG C); Edg2 (5′-TCC ATT GCC TTC TTT TAT AAC CGA A; 3′-TGC TGA CAG TCA GTC TCC GAG TAT T); Edg3 (5′-ACC ATC GTG ATC CTC TAC GCA C; 3′-CTT GAT TTA CTT CTG CTT GGG TCG); Edg4 (5′-ATG GGC CAG TGC TAC TAC AAC G; 3′-AGG AAG ACA AGC AGG CTC GAC); and Edg5 (5′-ACT GTC CTG CCT CTC TAC GCC; 3′-GTC TTG AGC AGG GCT AGC GTC), were synthesized by the Duke University Medical Center Primer Core Facility (Durham, NC). All other chemicals were of analytical grade. 
Cell Cultures
HTM cells from cadaveric eyes were isolated as described previously. 5 Cells were cultured at 37°C under 5% CO2, in Dulbecco’s modified Eagle medium (DMEM) containing 10% fetal bovine serum (FBS) and penicillin (100 U/mL)-streptomycin (100 μg/mL). Multiple primary cell cultures from distinct donor eye tissues were used. All experiments were conducted using confluent cell cultures, and cells were serum starved for a period of 48 hours, unless otherwise indicated. Cells were used at passages 2 to 7. 
Polymerase Chain Reaction Assays
To determine the expression pattern of Edg receptor isoforms in TM, reverse-transcribed total RNA of HTM cells (two sets, each derived from eyes of distinct donors, ages 26 and 54), HTM cell plasmid-derived cDNA libraries (two sets, each derived from distinct donors, donor ages uncertain) and HTM tissue plasmid-derived cDNA library (one set, donor age 69) were amplified by PCR, using sequence-specific forward (5′) and reverse (3′) oligonucleotide primers for each of the receptor isoforms Edg1 to 5. An RT null sample (−RT) served as the negative control for RT-PCR. PCR products were then subjected to agarose gel electrophoresis with ethidium bromide for UV visualization. PCR-generated DNA products were also sequenced to confirm the specific sequence of individual isoforms of Edg receptors. 
Cytoskeletal Staining
HTM cells were grown to confluence on gelatin (2%)-coated glass coverslips. Cells were subjected to serum starvation before treatment with either LPA (20 μM) or S1P (1 μM) for 1 hour at 37°C. Treatment with fatty acid-free BSA (8 μg/mL) served as the negative control. In addition, cultures (without serum starvation) of HTM cells were pretreated under serum-free conditions with one of the following pharmacologic inhibitors for 20 minutes before growth factor treatment: Y-27632 (5 μM, Rho kinase inhibitor), GF-109203X (10 μM, PKC inhibitor), or ML-7 (25 μM, MLC kinase inhibitor). 26 27 28 Treated cells were fixed and stained for actin stress fibers and focal adhesions, as described previously. 5  
Myosin Light-Chain Phosphorylation
Myosin light-chain (MLC) phosphorylation status in HTM cells was determined with the procedure described by Garcia et al. 29 Briefly, serum-starved cultures of HTM cells were treated with LPA (20 μM), S1P (1 μM), or fatty acid-free BSA for 1 hour at 37°C and were then extracted with 10% cold trichloroacetic acid. Precipitates obtained after centrifugation at 13,000 rpm were dissolved in 8 M urea buffer containing 20 mM Tris, 23 mM glycine, 10 mM dithiothreitol (DTT), saturated sucrose, and 0.004% bromophenol, in a sonicator. For qualitative analysis of total MLC in treated samples, equal volumes of sample lysates (derived from equal numbers of cells) were subjected to urea/glycerol-polyacrylamide gel electrophoresis (PAGE) and Western blot analysis with rabbit polyclonal antibody directed against MLC, as described previously. 5 For quantitative analysis of di-phosphorylated MLC in treated samples, protein concentrations of precipitates were first determined by the Bradford method, 30 and equal amounts of protein (75 μg/sample) were subjected to urea/glycerol-PAGE and Western blot analysis with rabbit polyclonal antibody directed against di-phospho-MLC (Thr18/Ser19), as described previously. 5 For quantitative analysis, equal protein loading was further confirmed by Western blot analysis with monoclonal antibody directed against actin. In an additional set of experiments, cultures of HTM cells (without serum starvation) were pretreated under serum-free conditions with one of the three inhibitors listed earlier for cytoskeletal staining or DMSO control for 20 minutes before treatment with LPA (20 μM) and S1P (1 μM) and subsequent cell extraction. Quantitative analysis of di-phosphorylated MLC (Thr18/Ser19) in these pretreated samples was then performed with equal amounts of protein (60 μg/sample), as described earlier. 
Rho GTPase Activation
Activation of Rho (GTP-bound Rho) in HTM cells was determined using immobilized rhotekin Rho-binding domain slurry as described in the manufacturer’s protocol (Upstate Biotechnology). Briefly, serum-starved HTM cell cultures were treated with LPA (20 μM), S1P (1 μM), or fatty acid–free BSA for 1 hour at 37°C, and cells were then extracted and lysed at 4°C in lysis buffer consisting of 25 mM HEPES buffer (pH 7.5) containing 25 mM NaF, 1 mM sodium orthovanadate, 0.15 M NaCl, 1% Igepal, CA-630, 10 mM MgCl2, 1 mM EDTA, 10% glycerol, aprotinin (10 μg/mL), and leupeptin (10 μg/mL). Cell lysates preincubated with GTPγS and GDP served as positive and negative controls, respectively. Protein concentration was estimated by the Bradford method. 30 Equal amounts of protein (300 μg/sample) were then incubated with agarose-rhotekin Rho binding domain at 4°C to isolate or “pull-down” activated GTP-Rho, and agarose beads were centrifuged and washed in lysis buffer. Proteins released from the agarose beads were then subjected to SDS-PAGE (12.5%) and Western blot analysis using rabbit polyclonal antibody directed against Rho, as described previously. 31 For measurement of total Rho, protein concentration of total cell lysates (nonincubated) was estimated by the Bradford method, 30 and equal amounts of protein (25 μg/sample) were subjected to SDS-PAGE (12.5%) and Western blot analysis. 
Measurement of Intracellular Calcium Levels
Intracellular calcium in lysophospholipid-treated single HTM cells was quantified using digital-imaging microscopy as described previously by Odom et al. 32 and Misra et al. 33 Briefly, HTM cells were plated on glass coverslips placed in 35-mm Petri dishes at a density of 106 cells/cm2 and allowed to adhere for 12 hours at 37°C before serum starvation (24 hours). Cells were preincubated with the fluorescent Ca2+ indicator fura-2/AM (4 μM) for 30 minutes in the dark at 37°C and were then washed twice with Hanks’ balanced salt solution containing 10 mM HEPES and 3.5 mM NaHCO3 (pH 7.4). Intracellular calcium was subsequently measured using digital-imaging microscopy as previously described. 32 33 After obtaining baseline measurements for 2 to 3 minutes, LPA (2 and 20 μM), S1P (1 and 10 μM), or fatty acid-free BSA was added and multiple [Ca2+]i measurements were taken. To analyze intracellular calcium mobilization quantitatively, peak calcium signal values, expressed as the mean ± SE, were determined for small clusters (numbering 9–15) of treated HTM cells. Treatment with thapsigargin served as a positive control for intracellular calcium release. 
Aqueous Humor Outflow Facility
Porcine eyes (obtained freshly from a local abattoir) were perfused with either LPA (50 μM) or S1P (5 μM) by the standard constant pressure technique using a Grant stainless steel corneal fitting. 14 Initial baseline outflow measurements were established at 15 mm Hg and 33°C with perfusion medium containing Dulbecco’s phosphate-buffered saline (DPBS; pH 7.4), and 5.5 mM d-glucose. After this, the anterior chamber aqueous of one eye of each pair was exchanged either with LPA (50 μM) or S1P (5 μM) dissolved in perfusion medium containing 0.2% fatty acid-free BSA and perfused with the same concentration of lipid factors for a period of 5 hours. The contralateral fellow eye was perfused with medium containing 0.2% fatty acid-free BSA alone. Outflow measurements were recorded at hourly intervals. Effects of each factor are expressed as the percentage change in outflow facility (compared with baseline values) over 5 hours, in factor-treated versus sham-treated paired controls. Values are expressed as mean ± SE. Data was analyzed by a paired two-tailed Student’s t-test to determine significance. 
Outflow Pathway Morphology
At the end of a 5-hour perfusion period, sham control and factor-treated fellow eyes were fixed for histologic examination, by perfusing them with 2.5% glutaraldehyde and 2% formaldehyde at 15 mm Hg pressure. Tissue quadrants obtained from factor-treated and control eyes were fixed in 1.0% osmium tetroxide in 0.1 M sodium cacodylate buffer and then stained with 1% uranyl acetate. Finally, sections obtained by microtomy (70 nm) were stained sequentially with KMnO4 and Sato’s stain and photographed using an electron microscope (Jem-1200 EX; JEOL, Tokyo, Japan) and a light microscope (Carl Zeiss Meditec, Thornwood, NY). 
Results
Expression of Edg Receptors in HTM Cells
To determine the expression profile of Edg receptors in HTM, we used PCR of reverse-transcribed total RNA extracted from two HTM cell cultures (derived from different donor eyes, ages 26 and 54) and PCR of plasmid-derived cDNA libraries from HTM cells and from HTM tissue (donor age 69) to amplify specific DNA fragments of individual Edg receptor coding regions. Expression of Edg receptor isoforms 1 to 4 was apparent in both RT-PCR HTM cell libraries (lane 1: age 26; lane 2: age 54) and in one of the plasmid-derived cDNA HTM cell libraries, whereas in the second cDNA HTM cell library and in the cDNA HTM tissue library, only Edg2 was detectable (Fig. 1) . Expression of Edg5 was not detected in any of the HTM cell or tissue libraries (data not shown). Human Schlemm’s canal (SC) cells were also tested for expression of Edg receptors by RT-PCR amplification, and as was seen in HTM cells by RT-PCR, SC cells also exhibited the expression of Edg 1-4 receptors (data not shown). 
Effects of LPA and S1P on HTM Cell Cytoskeletal Organization
Treatment of serum-starved HTM cells (grown to confluence on gelatin-coated glass coverslips) with either LPA (20 μM) or S1P (1 μM) for 1 hour induced alterations in staining patterns for both F-actin (phalloidin) and focal adhesions (vinculin). Both LPA and S1P stimulated formation of actin stress fibers and focal adhesions in treated cells, compared with BSA-treated control cells (Fig. 2) . In addition, treatment with S1P appeared to promote formation of cortical actin fibers in a limited number of treated cells. Assessment of HTM cell morphology by phase-contrast microscopy did not reveal detectable changes in cell shape in LPA- or S1P-treated cells (data not shown). 
Effects of LPA and S1P on MLC Phosphorylation in HTM Cells
To investigate the effects of LPA and S1P on HTM cell MLC phosphorylation, total cell lysates from serum-starved HTM cell cultures treated with LPA (20 μM), S1P (1 μM), or BSA control were subjected to either (Fig. 3A) quantitative analysis with urea-glycerol/PAGE of equal amounts of protein followed by Western blot with polyclonal antibody directed against di-phospho-MLC (Thr18/Ser19) or (Fig. 3B) qualitative analysis with urea-glycerol/PAGE of equal volumes of sample lysates (derived from equal numbers of cells) followed by Western blot with polyclonal antibody directed against MLC (Fig. 3B) . Treatment of HTM cells with LPA or S1P promoted a robust increase in di-phosphorylation of MLC, compared with control-treated cells. Qualitative analysis of total MLC phosphorylation (Fig. 3B) was consistent with this observed effect, as LPA- and S1P-treated cells exhibited di-phospho-MLC immunoreactive bands with greater intensity relative to control-treated cells, in which very little MLC existed in the di-phospho form. 
LPA- and S1P-Promoted Activation of Rho GTPase in HTM Cells
To explore the potential involvement of Rho GTPase in LPA- and S1P-mediated cellular effects, we investigated activation of Rho GTPase in lysophospholipid-treated HTM cells. Equivalent amounts of protein precipitates prepared from serum-starved HTM cells treated with either LPA (20 μM) or S1P (1 μM) were incubated with agarose-rhotekin Rho binding domain to selectively “pull-down” the activated Rho (GTP-Rho). Figure 4 depicts changes in Rho GTPase activation in treated HTM cells. Both LPA and S1P stimulated Rho GTPase activation in HTM cells. Cell lysates preincubated with either GDP or GTPγs were subjected to the same protocol along with LPA- and S1P-treated samples, and these samples demonstrated null and detectable immunoreactive bands of Rho GTPase, respectively (data not shown). Western blot analysis of total Rho in nonincubated total cell lysates revealed that levels of Rho in HTM cells were not influenced by treatment with either LPA or S1P (Fig. 4)
Intracellular Calcium Concentration in HTM Cells Treated with LPA and S1P
To investigate the effects of LPA and S1P on intracellular calcium concentration, we measured intracellular calcium levels in lysophospholipid-stimulated, serum-starved HTM cells by digital-imaging microscopy, using the fluorescent Ca2+-sensitive indicator fura-2/AM. After baseline values were obtained, cells were treated with either LPA (2 and 20 μM) or S1P (1 and 10 μM), and intracellular calcium values were recorded. Transient intracellular calcium signals were evident in only a limited number of cells in response to 2 μM LPA (data not shown), but treatment with 20 μM LPA consistently elicited calcium signals in HTM cells (Fig. 5A) . In contrast, intracellular calcium signals were not observed in 1 μM S1P-treated cells (data not shown), whereas 10 μM S1P elicited transient calcium signals in slightly less than one-half of treated HTM cells (Fig. 5A) . To compare relative magnitudes of elicited calcium signals, mean signal values for small clusters (numbering 9 to 15) of cells treated with either 20 μM LPA or 10 μM S1P were calculated (Fig. 5B) . Treatment with 20 μM LPA elicited nearly a two-fold higher calcium signal than treatment with 10 μM S1P. 
Effects of Specific Pharmacologic Inhibitors on LPA- and S1P-Induced Changes in HTM Cell MLC Phosphorylation
To determine the relative contributions of both calcium-independent and calcium-dependent intracellular signaling pathways to lysophospholipid-induced MLC phosphorylation in HTM cells, sets of three HTM cell cultures underwent pretreatment with one of the following in serum-free conditions: DMSO control, Y-27632 (5 μM, Rho kinase inhibitor), GF-109203X (10 μM, PKC inhibitor), or ML-7 (25 μM, MLC kinase inhibitor) for 20 minutes. Each set of cultures was subsequently treated with BSA control, LPA (20 μM), and S1P (1 μM) for 1 hour. Cell lysates containing equal amounts of protein were subjected to urea/glycerol-PAGE and Western blot analysis using polyclonal antibody directed against di-phospho-MLC (Fig. 6) . Pretreatment of HTM cells with ML-7 had a marginal effect on LPA- and S1P-induced increases in MLC phosphorylation relative to control cells. While pretreatment of HTM cells with GF-109203X also had little effect on the increase in MLC phosphorylation in cells treated with LPA, GF-109203X–pretreated cells that were treated with S1P exhibited partial attenuation of MLC phosphorylation compared with control cells, suggesting a modest involvement of PKC in S1P-mediated effects on HTM cell MLC phosphorylation. However, pretreatment of HTM cells with Y-27632 almost completely inhibited LPA- and S1P-mediated effects on MLC phosphorylation (Fig. 6) . Study of actin cytoskeletal organization in pharmacological inhibitor-pretreated HTM cells, which were then treated with either LPA or S1P, was similar to these findings. Inhibition of Rho kinase completely attenuated the effects of lysophospholipid treatment on actin stress fiber formation compared with a partial inhibition observed with inhibition of either PKC or MLC kinase (data not shown). These data suggest that Rho kinase-mediated signaling is the predominant pathway by which LPA and S1P exert their effects on MLC phosphorylation and the actin cytoskeleton in HTM cells. 
Modulation of Aqueous Humor Outflow Facility in Enucleated Porcine Eyes Perfused with LPA and S1P
Freshly enucleated porcine eyes were perfused with either 50 μM LPA (Fig. 7A) or 5 μM S1P (Fig. 7B) at a constant pressure of 15 mm Hg, after establishing the baseline outflow facility with DPBS buffer containing glucose at 33°C. Basal rates of outflow facility (in microliters per minute per mm Hg) in the control (LPA series)-, LPA-, control (S1P series)-, and S1P-perfused groups were 0.326 ± 0.059 (n = 6), 0.310 ± 0.052 (n = 6), 0.409 ± 0.070 (n = 5), and 0.485 ± 0.082 (n = 5), respectively (mean ± SE). Outflow facility was observed to decrease significantly from baseline after 1 hour of either LPA (P < 0.04) or S1P (P < 0.03) treatment. Outflow facility continued to decrease significantly at each of the four remaining time points over the perfusion time course with maximum decreases in outflow facility from baseline of 37% and 31%, for LPA and S1P, respectively. Fellow-paired control eyes perfused with PBS buffer plus 0.2% fatty acid-free BSA in the LPA and S1P series showed 14% and 34% increase in outflow facility, respectively, over the corresponding initial baseline outflow facility values after 3 hours of perfusion. 
Histologic examination of LPA- and S1P-perfused porcine eyes performed by light microscopy (LM) and electron microscopy (EM) demonstrates compact trabecular beams and juxtacanalicular tissue, which was not remarkably different from examination of sham-perfused eyes (Fig. 8 , S1P-perfused eyes; data for LPA-perfused eyes not shown). Different quadrants of sham- and factor-treated eyes were used in this analysis. No cell loss or accumulation of cell debris was evident in the TM of eyes perfused with LPA (50 μM), S1P (5 μM), or 0.2% fatty acid-free BSA control solution (Fig. 8) . Specifically, we noted the lack of accumulation of material in the extracellular space, which may have otherwise accounted for the observed decrease in outflow facility during perfusion. 
Discussion
In this study, we have attempted to elucidate a potential role for lysophosphatidic acid (LPA) and sphingosine-1-phosphate (S1P) in the modulation of aqueous humor outflow facility. Collectively, data presented herein suggest that these lysophospholipid growth factors promote a decrease in outflow facility in perfused enucleated porcine eyes, and this apparent increase in resistance to outflow correlates with increased MLC phosphorylation and increased formation of actin stress fibers and focal adhesions in HTM cells, primarily by activation of the Rho/Rho kinase signaling pathway. 
We have hypothesized that although specific inhibition of Rho kinase might increase outflow facility, 5 conversely, factors that promote activation of Rho/Rho kinase might then decrease outflow facility. To explore this, we have chosen to study, from among various known activators of the cellular actomyosin apparatus, lysophospholipid growth factors, based on preliminary studies suggesting the possible involvement of lipid-based molecules in the modulation of TM cell cytoskeletal organization (Epstein DL, et al. IOVS 2001;42:ARVO Abstract 746) and based on a vastly expanding literature exploring the roles of these factors in various physiological processes. Among its effects, LPA is known to influence cell morphology and smooth muscle contractility. 19 34 S1P has been implicated in modulating endothelial cell migration and adherens junction assembly, as well as smooth muscle cell migration and contractility. 20 35 In addition, various intracellular signaling molecules, including Rho GTPase (through activation of Gα12/13), as well as PKC and MLC kinase (through activation of Gαq/11), have been shown to serve as downstream effectors for the lysophospholipids. 23 In this study, we took a mechanistic approach to study the effects of LPA and S1P on TM cells and on aqueous humor outflow facility. 
In our study of the expression profile of Edg receptors, we detected expression of Edg receptor isoforms 1 to 4 in two HTM cell cultures (cultured from two different donors, ages 26 and 54) by RT-PCR and in one plasmid-derived cDNA library by PCR. However, only Edg2 was detectable in a second cDNA library and in an HTM tissue (donor age 69) cDNA library. The differential expression that is seen among these libraries suggests that Edg receptor expression in the TM may be inducible or that receptor expression in the TM may be up- or downregulated. It has been demonstrated that Edg receptor expression level may be up- or downregulated under different physiological conditions. 36 37 38 Particularly of note, certain growth factors, such as vascular endothelial growth factor (VEGF), have been shown to upregulate expression of Edg receptor isoforms, 39 suggesting that cross-talk between lysophospholipid-mediated signaling and other signaling pathways may play a role in the regulation of Edg receptor expression. In addition, although it is notable that there was no qualitative difference in expression profile by RT-PCR in HTM cells taken from donors of different ages (26 and 54), we cannot exclude the possibility that levels of receptor expression may vary with age in TM. However, further studies are needed to investigate this possibility and to understand the cellular mechanisms involved in the regulation of Edg receptor expression. 
Treatment of serum-starved HTM cells with either LPA or S1P promoted formation of actin stress fibers and focal adhesions (Fig. 2) , and these changes were associated with increased di-phosphorylation of MLC in HTM cells, as assessed by Western blot analyses for di-phospho-MLC and total MLC (Figs. 3A 3B) . In addition, increased MLC phosphorylation observed in LPA- and S1P-treated HTM cell lysates correlates well with the strong activation of Rho GTPase observed in treated HTM cells (Fig. 4) . Measurement of intracellular calcium levels in treated HTM cells demonstrates a consistent increase in calcium levels in LPA-treated cells, whereas S1P-treated cells did not display similar consistency (Figs. 5A 5B) . It is possible that this contrast between LPA- and S1P-treated cells was related to differential levels of expression of the various Edg receptor isoforms at the cellular level, since certain isoforms elicit more robust increases in intracellular calcium through coupling to Gαq/11 and its downstream signaling pathways. 34 Nevertheless, taken together, the data on Rho activation and intracellular calcium mobilization in HTM cells suggest the possibility that calcium-independent and calcium-dependent signaling mechanisms activated by LPA and S1P both contribute to the observed increases in HTM cell MLC phosphorylation. However, the presence of a dramatically attenuated di-phospho-MLC immunoreactive band in Y-27632–pretreated HTM cells treated with LPA and S1P suggests that calcium-independent signaling is the primary contributor to lysophospholipid-mediated changes in MLC phosphorylation (Fig. 6) . Notably, this effect was observed using an inhibitor concentration of 5 μM, which is well within the specificity threshold for inhibition of Rho kinase. 26 Although pretreatment with the MLC kinase inhibitor ML-7 produced only a marginal effect on lysophospholipid-mediated increases in HTM cell MLC phosphorylation, it was interesting that pretreatment with GF-109203X partially inhibited MLC phosphorylation in S1P-treated cells, suggesting that PKC signaling may supplement Rho kinase in S1P-elicited HTM cellular effects. Furthermore, this finding indicates that LPA and S1P activate differential signaling pathways. Taken together, these data suggest that activation of Rho kinase plays a major role in lysophospholipid-induced MLC phosphorylation in HTM cells, whereas PKC and MLC kinase make minor contributions to increases in HTM cell MLC phosphorylation mediated by lysophospholipids. 
Porcine eye perfusion data for both LPA and S1P were similar with progressive decreases in outflow facility from baseline throughout the 5-hour time course, resulting in maximum decreases in outflow facility from baseline of 37% and 31%, respectively (Fig. 7) . Notably, the lack of accumulated material in the extracellular spaces in the outflow pathway on morphologic examination (Fig. 8) suggests that decreases were not attributable to physical blockage of paracellular fluid flow by lipids. Higher concentrations of LPA (50 μM) and S1P (5 μM) have been used in perfusion studies to compensate for the more limited accessibility of perfused factors to TM cells in intact tissue, relative to the accessibility to TM cells in culture conditions. However, we have observed that treatment of TM cells in culture conditions with either 50 μM LPA or 5 μM S1P does not result in opposing effects on cytoskeletal reorganization but, rather, produces similar effects to those observed with concentrations used for cytoskeletal assays in this study (data not shown). 
In total, these data suggest that the effects of LPA and S1P on TM cellular “tone” (contraction/relaxation) may contribute to an increased resistance to outflow, as evident in perfusion studies. LPA- and S1P-stimulated increases in MLC phosphorylation, mediated primarily through activation of the Rho/Rho kinase signaling pathway, promote actomyosin-based cellular contraction. Enhanced cellular contraction in turn may then influence cell–cell and cell–ECM interactions, 40 and such a response may then cause a reduction in the filtering space for fluid flow through the TM. However, such a change in filtering space along the outflow pathway was not apparent by either LM or EM (Fig. 8) . One potential explanation for these findings is that the effects of these factors on outflow facility may not be readily evident by histologic examination of the outflow pathway, given that LPA and S1P promote only a modest decrease (30%–40%) in outflow facility. We have observed in previous studies in our laboratory that SC cell monolayers treated with LPA (20 μM) exhibit a decreased permeability relative to control-treated cell monolayers (Kumar J, Rao PV, Epstein DL, unpublished data, September 1999), suggesting that altered paracellular permeability may contribute to the reduction in outflow facility that is seen in the absence of observable histologic changes in the outflow pathway. 
Although, the mechanism for extracellular release of lysophospholipids is unknown, 24 25 within the context of ocular biology, it is vital to determine which cell types possess the ability to synthesize such factors and secrete them into the aqueous. LPA and S1P have been implicated in several other disease processes, particularly various types of vasoconstriction, 41 and enhanced production of sphingomyelin metabolites and increased synthesis and activation of Rho GTPase have been observed in certain types of vascular smooth muscle in response to hypoxic stress. 42 43 44 In endothelial cells of blood vessels subjected to shear stress, expression levels of Edg receptor subtypes and activation of Rho GTPase have been found to be increased, 45 46 suggesting the possibility that either the Edg receptor system and/or the cellular actomyosin apparatus is either integral to, or may dynamically altered in response to, the chronic mechanical stress of increased IOP in primary open-angle glaucoma (POAG). These possibilities warrant further investigation within the context of glaucomatous disease based on the demonstrable influences of LPA and S1P in increasing resistance to aqueous humor outflow. 
In conclusion, the data presented herein indicate that LPA and S1P, through the Rho/Rho kinase signaling pathway, promote activation of the TM cell actomyosin apparatus and decrease outflow facility in porcine eyes, providing evidence for a novel mechanism for negative regulation of conventional aqueous humor outflow facility. The bioactivity of the lysophospholipids LPA and S1P in modulating TM function and the potential involvement of these factors in the regulation of the conventional outflow pathway may open new approaches to develop a better understanding, not only of the physiology and pathophysiology of POAG but also of potential novel therapeutic interventions for glaucomatous disease. 
 
Figure 1.
 
Expression of Edg receptors in human TM cells and TM tissue was determined by RT-PCR of total RNA from two primary cultures of cells from different donors, age 26 (lane 1) and age 54 (lane 2) and by PCR amplification of two HTM cell plasmid-derived cDNA libraries and one HTM tissue cDNA library, using sequence-specific oligonucleotide primers. Expression of Edg receptors 1 to 4 isoforms was readily detectable in both cell RT-PCR libraries and in one of the cell cDNA libraries, whereas Edg2 was the only isoform amplified from one of the HTM cell-derived cDNA libraries and from the HTM tissue-derived cDNA library.
Figure 1.
 
Expression of Edg receptors in human TM cells and TM tissue was determined by RT-PCR of total RNA from two primary cultures of cells from different donors, age 26 (lane 1) and age 54 (lane 2) and by PCR amplification of two HTM cell plasmid-derived cDNA libraries and one HTM tissue cDNA library, using sequence-specific oligonucleotide primers. Expression of Edg receptors 1 to 4 isoforms was readily detectable in both cell RT-PCR libraries and in one of the cell cDNA libraries, whereas Edg2 was the only isoform amplified from one of the HTM cell-derived cDNA libraries and from the HTM tissue-derived cDNA library.
Figure 2.
 
LPA- and S1P-induced actin cytoskeletal reorganization and formation of focal adhesions in human TM cells. Treatment of serum-starved human TM cells with either LPA (20 μM) or S1P (1 μM) for 1 hour induced formation of actin stress fibers (phalloidin staining) and focal adhesions (vinculin staining) compared with BSA-treated control cells. S1P-treated cells exhibited extensive cortical actin stress fibers with obvious cortical rigidity compared with LPA-treated cells.
Figure 2.
 
LPA- and S1P-induced actin cytoskeletal reorganization and formation of focal adhesions in human TM cells. Treatment of serum-starved human TM cells with either LPA (20 μM) or S1P (1 μM) for 1 hour induced formation of actin stress fibers (phalloidin staining) and focal adhesions (vinculin staining) compared with BSA-treated control cells. S1P-treated cells exhibited extensive cortical actin stress fibers with obvious cortical rigidity compared with LPA-treated cells.
Figure 3.
 
Increased MLC phosphorylation in LPA- and S1P-treated human TM cells. Serum-starved human TM cells treated with either LPA (20 μM) or S1P (1 μM) for 1 hour exhibited increased MLC phosphorylation (di-phosphoform) relative to BSA-treated control cells, as determined by urea/glycerol-polyacrylamide gel electrophoresis followed by (A) Western blot analysis of equal amounts of protein using polyclonal antibody directed against Thr18/Ser19 di-phospho-MLC and (B) Western blot analysis of equal volumes of lysates (prepared from equal numbers of cells) using polyclonal antibody directed against MLC.
Figure 3.
 
Increased MLC phosphorylation in LPA- and S1P-treated human TM cells. Serum-starved human TM cells treated with either LPA (20 μM) or S1P (1 μM) for 1 hour exhibited increased MLC phosphorylation (di-phosphoform) relative to BSA-treated control cells, as determined by urea/glycerol-polyacrylamide gel electrophoresis followed by (A) Western blot analysis of equal amounts of protein using polyclonal antibody directed against Thr18/Ser19 di-phospho-MLC and (B) Western blot analysis of equal volumes of lysates (prepared from equal numbers of cells) using polyclonal antibody directed against MLC.
Figure 4.
 
LPA- and S1P-induced activation of Rho GTPase in human TM cells. Cell lysates containing equal amounts of total protein (300 μg/sample) prepared from serum-starved human TM cells treated with either LPA (20 μM) or S1P (1 μM) demonstrated an increased amount of GTP-Rho in agarose-rhotekin “pull-down” assays compared with BSA control-treated cells, indicating activation of Rho GTPase (GTP-loading). Western blot analysis of nonincubated cell lysates (25 μg/sample) demonstrated no change in total Rho as a result of lysophospholipid treatment.
Figure 4.
 
LPA- and S1P-induced activation of Rho GTPase in human TM cells. Cell lysates containing equal amounts of total protein (300 μg/sample) prepared from serum-starved human TM cells treated with either LPA (20 μM) or S1P (1 μM) demonstrated an increased amount of GTP-Rho in agarose-rhotekin “pull-down” assays compared with BSA control-treated cells, indicating activation of Rho GTPase (GTP-loading). Western blot analysis of nonincubated cell lysates (25 μg/sample) demonstrated no change in total Rho as a result of lysophospholipid treatment.
Figure 5.
 
Mobilization of intracellular calcium in LPA- and S1P-treated human TM cells. Serum-starved human TM cells preloaded with the fluorescent Ca2+ indicator fura-2/AM were subsequently treated with LPA (20 μM) or S1P (10 μM), and intracellular calcium release was determined by spectrofluorometric digital-imaging microscopy. (A) Representative time-dependent intracellular calcium mobilization curves for individual TM cells treated with LPA (20 μM), S1P (10 μM), or BSA control. Arrow: time at which factors were added. (B) The mean peak intracellular calcium concentration of TM cells (clusters of 9–15 cells) treated with LPA (20 μM), S1P (10 μM), or BSA control. HTM cells treated with LPA possessed nearly a twofold higher intracellular calcium concentration than HTM cells treated with S1P.
Figure 5.
 
Mobilization of intracellular calcium in LPA- and S1P-treated human TM cells. Serum-starved human TM cells preloaded with the fluorescent Ca2+ indicator fura-2/AM were subsequently treated with LPA (20 μM) or S1P (10 μM), and intracellular calcium release was determined by spectrofluorometric digital-imaging microscopy. (A) Representative time-dependent intracellular calcium mobilization curves for individual TM cells treated with LPA (20 μM), S1P (10 μM), or BSA control. Arrow: time at which factors were added. (B) The mean peak intracellular calcium concentration of TM cells (clusters of 9–15 cells) treated with LPA (20 μM), S1P (10 μM), or BSA control. HTM cells treated with LPA possessed nearly a twofold higher intracellular calcium concentration than HTM cells treated with S1P.
Figure 6.
 
Pharmacological inhibition of LPA- and S1P-induced changes in MLC phosphorylation in human TM cells. To determine the involvement of Rho kinase, PKC, and MLC kinase in LPA- and S1P-induced changes in MLC phosphorylation, human TM cells were pretreated with specific inhibitors of Rho kinase (5 μM Y-27632), PKC (10 μM GF-109203X), or MLC kinase (25 μM ML-7) and with DMSO control for 20 minutes before treatment with either LPA (20 μM) or S1P (1 μM) for 1 hour. Cell lysates containing equal amounts of protein from treated cells were subjected to urea/glycerol-polyacrylamide gel electrophoresis and Western blot analysis using polyclonal antibody directed against Thr18/Ser19 di-phospho-MLC. Control-pretreated HTM cells exhibited markedly increased levels of di-phospho-MLC when treated with either LPA or S1P. Whereas LPA-mediated increases in di-phospho-MLC were only marginally sensitive to pretreatment with either GF-109203X or ML-7, S1P-mediated increases in di-phospho-MLC were partially inhibited by pretreatment with GF-109203X and were marginally sensitive to ML-7 pretreatment. However, pretreatment of HTM cells with Y-27632 markedly attenuated both LPA- and S1P-induced effects on MLC phosphorylation.
Figure 6.
 
Pharmacological inhibition of LPA- and S1P-induced changes in MLC phosphorylation in human TM cells. To determine the involvement of Rho kinase, PKC, and MLC kinase in LPA- and S1P-induced changes in MLC phosphorylation, human TM cells were pretreated with specific inhibitors of Rho kinase (5 μM Y-27632), PKC (10 μM GF-109203X), or MLC kinase (25 μM ML-7) and with DMSO control for 20 minutes before treatment with either LPA (20 μM) or S1P (1 μM) for 1 hour. Cell lysates containing equal amounts of protein from treated cells were subjected to urea/glycerol-polyacrylamide gel electrophoresis and Western blot analysis using polyclonal antibody directed against Thr18/Ser19 di-phospho-MLC. Control-pretreated HTM cells exhibited markedly increased levels of di-phospho-MLC when treated with either LPA or S1P. Whereas LPA-mediated increases in di-phospho-MLC were only marginally sensitive to pretreatment with either GF-109203X or ML-7, S1P-mediated increases in di-phospho-MLC were partially inhibited by pretreatment with GF-109203X and were marginally sensitive to ML-7 pretreatment. However, pretreatment of HTM cells with Y-27632 markedly attenuated both LPA- and S1P-induced effects on MLC phosphorylation.
Figure 7.
 
Effects of perfusion of LPA and S1P on aqueous outflow facility in enucleated porcine eyes. Freshly enucleated porcine eyes were perfused with either (A) 50 μM LPA or (B) 5 μM S1P at a constant pressure of 15 mm Hg, after establishing the baseline outflow facility with perfusion media containing d-glucose at 33°C. Outflow facility was observed to decrease significantly after 1 hour of either LPA (P < 0.04) or S1P (P < 0.03) treatment, respectively, from the baseline outflow facility. Outflow facility continued to decrease significantly at each of the four remaining time points over the perfusion time course with maximum decreases in outflow facility from baseline of 37% and 31%, for LPA and S1P, respectively. Fellow-paired control eyes perfused with media containing 0.2% fatty acid-free BSA in the LPA series and S1P series showed 14% and 34% increases, respectively, over the corresponding initial baseline outflow facility, after 3 hours.
Figure 7.
 
Effects of perfusion of LPA and S1P on aqueous outflow facility in enucleated porcine eyes. Freshly enucleated porcine eyes were perfused with either (A) 50 μM LPA or (B) 5 μM S1P at a constant pressure of 15 mm Hg, after establishing the baseline outflow facility with perfusion media containing d-glucose at 33°C. Outflow facility was observed to decrease significantly after 1 hour of either LPA (P < 0.04) or S1P (P < 0.03) treatment, respectively, from the baseline outflow facility. Outflow facility continued to decrease significantly at each of the four remaining time points over the perfusion time course with maximum decreases in outflow facility from baseline of 37% and 31%, for LPA and S1P, respectively. Fellow-paired control eyes perfused with media containing 0.2% fatty acid-free BSA in the LPA series and S1P series showed 14% and 34% increases, respectively, over the corresponding initial baseline outflow facility, after 3 hours.
Figure 8.
 
Histologic changes in enucleated porcine eyes perfused with S1P. Enucleated porcine eyes perfused with either 5 μM S1P or 0.2% BSA sham control for 5 hours at 33°C were fixed for histologic examination by both light (A, B) and electron (C, D) microscopy. Both S1P- (B) and control-treated (A) specimens exhibited compact trabecular beams and both had similar-appearing juxtacanalicular regions (C, D, arrows), with no observable difference in filtration space. In addition, filtration space was noted to be free of any accumulated extracellular material. Magnification: (A, B) ×1000; (C) ×2500; and (D) ×2000 of the same specimen, respectively.
Figure 8.
 
Histologic changes in enucleated porcine eyes perfused with S1P. Enucleated porcine eyes perfused with either 5 μM S1P or 0.2% BSA sham control for 5 hours at 33°C were fixed for histologic examination by both light (A, B) and electron (C, D) microscopy. Both S1P- (B) and control-treated (A) specimens exhibited compact trabecular beams and both had similar-appearing juxtacanalicular regions (C, D, arrows), with no observable difference in filtration space. In addition, filtration space was noted to be free of any accumulated extracellular material. Magnification: (A, B) ×1000; (C) ×2500; and (D) ×2000 of the same specimen, respectively.
The authors thank Joe G. Garcia for a kind gift of rabbit polyclonal myosin light-chain antibody, Pedro Gonzalez for providing plasmid-derived cDNA libraries, and Wenxiu Zhang for assistance with electron microscopy. 
Lütjen-Drecoll E. Functional morphology of the trabecular meshwork in primate eyes. Prog Retinal Eye Res. 1998;18:91–119.
Johnson M, Erickson K. Mechanism and routes of aqueous humor drainage. Albert DM Lakobiec FA eds. Principles and Practice of Ophthalmology. 2000 Glaucoma. 2000;4:2577–2595. WB Saunders Co. Philadelphia.
Tian B, Geiger B, Epstein DL, Kaufman PL. Cytoskeletal involvement in the regulation of aqueous humor outflow. Invest Ophthalmol Vis Sci. 2000;41:619–623. [PubMed]
Wiederholt M, Thieme H, Stumff F. The regulation of trabecular meshwork and ciliary muscle contractility. Prog Retinal Eye Res. 2000;19:271–295. [CrossRef]
Rao PV, Deng PF, Kumar J, Epstein DL. Modulation of aqueous humor outflow facility by the Rho kinase-specific inhibitor Y-27632. Invest Ophthalmol Vis Sci. 2001;42:1029–1037. [PubMed]
Khurana RN, Deng PF, Epstein DL, Rao PV. The role of protein kinase C in modulation of aqueous humor outflow facility. Exp Eye Res. 2003;76:39–47. [CrossRef] [PubMed]
Epstein DL, Freddo TF, Bassett-Chu S, Chung M, Karageuzian I. Influence of ethacrynic acid on outflow facility in the monkey and calf eye. Invest Ophthalmol Vis Sci. 1987;28:2067–2075. [PubMed]
Erickson-Lamy K, Schroeder A, Epstein DL. Ethacrynic acid induces reversible shape and cytoskeletal changes in cultured cells. Invest Ophthalmol Vis Sci. 1992;33:2631–2640. [PubMed]
Tian B, Kaufman PL, Volberg T, Gabelt BT, Geiger B. H-7 disrupts the actin cytoskeleton and increases outflow facility. Arch Ophthalmol. 1998;116:633–643. [CrossRef] [PubMed]
Pang IH, Hellberg PE, Fleenor DL, Jacobson N, Clark AF. Expression of matrix metalloproteinases and their inhibitors in human trabecular meshwork cells. Invest Ophthalmol Vis Sci. 2003;44:3485–3493. [CrossRef] [PubMed]
Welge-Lussen U, May CA, Lütjen-Drecoll E. Induction of tissue transglutaminase in the trabecular meshwork by TGF-β1 and TGF-β2. Invest Ophthalmol Vis Sci. 2000;41:2229–2238. [PubMed]
Bradley JM, Kelley MJ, Zhu X, Anderssohn AM, Alexander JP, Acott TS. Effects of mechanical stretching on trabecular matrix metalloproteinases. Invest Ophthalmol Vis Sci. 2001;42:1505–1513. [PubMed]
Gong H, Ruberti J, Overby D, Johnson M, Freddo TF. A new view of the human trabecular meshwork using quick-freeze, deep-etch electron microscopy. Exp Eye Res. 2002;75:347–358. [CrossRef] [PubMed]
Epstein DL, Rowlette LL, Roberts BC. Acto-myosin drug effects and aqueous outflow function. Invest Ophthalmol Vis Sci. 1999;40:74–81. [PubMed]
Kaufman PL, Gabelt B, Tian B, Liu X. Advances in glaucoma diagnosis and therapy for the next millennium: new drugs for trabecular and uveoscleral outflow. Semin Ophthalmol. 1999;14:130–143. [CrossRef] [PubMed]
Brubaker RF. Targeting outflow facility in glaucoma management. Surv Ophthalmol. 2003;48:S17–S20. [CrossRef] [PubMed]
Tripathi RC, Li J, Chan WF, Tripathi BJ. Aqueous humor in glaucomatous eyes contains an increased level of TGF-β2. Exp Eye Res. 1994;59:723–727. [CrossRef] [PubMed]
Tezel G, Kass MA, Kolker AE, Becker B, Wax MB. Plasma and aqueous humor endothelin levels in primary open-angle glaucoma. J Glaucoma. 1997;6:83–89. [PubMed]
Moolenaar WH. Bioactive lysophospholipids and their G protein-coupled receptors. Exp Cell Res. 1999;253:230–238. [CrossRef] [PubMed]
Hla T, Lee MJ, Ancellin N, Paik JH, Kluk MJ. Lysophospholipid-receptor revelations. Science. 2001;294:1875–1878. [CrossRef] [PubMed]
Lee MJ, Van Brocklyn JR, Thangada S, et al. Sphingosine-1-phosphate as a ligand for the G protein-coupled receptor Edg-1. Science. 1998;279:1552–1555. [CrossRef] [PubMed]
English D, Brindley DN, Spiegel S, Garcia JGN. Lipid mediators of angiogenesis and the signalling pathways they initiate. Biochim Biophys Acta. 2002;1582:228–239. [CrossRef] [PubMed]
Siess W. Athero- and thrombogenic actions of lysophosphatidic acid and sphingosine-1-phosphate. Biochim Biophys Acta. 2002;1582:204–215. [CrossRef] [PubMed]
Pagès C, Simon MF, Valet P, Saulnier-Blache JS. Lysophosphatidic acid synthesis and release. Prostaglandins Other Lipid Mediat. 2001;64:1–10. [CrossRef] [PubMed]
Yatomi Y, Ozaki Y, Ohmori T, Igarashi Y. Sphingosine 1-phosphate: synthesis and release. Prostaglandins Other Lipid Mediat. 2001;64:107–122. [CrossRef] [PubMed]
Ishizaki T, Uehata M, Tamechika I, et al. Pharmacological properties of Y27632, a specific inhibitor of Rho-associated kinases. Mol Pharmacol. 2000;57:976–983. [PubMed]
Toullec D, Pianetti P, Coste H, et al. The bisindolylmaleimide GF 109203X is a potent and selective inhibitor of protein kinase C. J Biol Chem. 1991;266:15771–15781. [PubMed]
Saitoh M, Ishikawa T, Matsushima S, Naka M, Hidaka H. Selective inhibition of catalytic activity of smooth muscle myosin light chain kinase. J Biol Chem. 1987;262:7796–7801. [PubMed]
Garcia JGN, Davi HW, Patterson CE. Regulation of endothelial cell gap formation and barrier dysfunction: role of myosin light chain phosphorylation. J Cell Physiol. 1995;163:510–522. [CrossRef] [PubMed]
Bradford M. A rapid and sensitive method for the quantification of microgram quantities of protein utilizing the principle of protein dye-binding. Anal Biochem. 1976;72:248–254. [CrossRef] [PubMed]
Maddala R, Reddy VN, Epstein DL, Rao PV. Growth factor induced activation of Rho and Rac GTPases and actin cytoskeletal reorganization in human lens epithelial cells. Mol Vision. 2003;9:329–336.
Odom AR, Misra UK, Pizzo SV. Nickel inhibits biding of α2-macroglobulin-methylamine to the low density lipoprotein receptor related protein/α2-macroglobulin receptor but not the α2-macroglobulin signaling receptor: evidence for a nickel effect on a region of α2-macroglobulin upstream of the receptor binding domain. Biochemistry. 1997;36:12395–12399. [CrossRef] [PubMed]
Misra UK, Chu TCC, Gawdi G, Pizzo SV. The relationship between low density lipoprotein-related protein/α2-macroglobulin (α2M) receptors and the newly described α2M signaling receptor. J Biol Chem. 1994;269:18803–18806.
Goetzl EJ, An S. Diversity of cellular receptors and functions for the lysophospholipid growth factors lysophosphatidic acid and sphingosine 1-phosphate. FASEB J. 1998;12:1589–1598. [PubMed]
Kluk MJ, Hla T. Signaling of sphingosine-1-phosphate via the S1P/EDG-family of G-protein-coupled receptors. Biochim Biophys Acta. 2002;1582:72–80. [CrossRef] [PubMed]
Meacci E, Cencetti F, Donati C, et al. Down-regulation of EDG5/S1P2 during myogenic differentiation results in the specific uncoupling of sphingosine 1-phosphate signalling to phospholipase D. Biochim Biophys Acta. 2003;1633:133–142. [CrossRef] [PubMed]
Tham CS, Lin FF, Rao TS, Yu N, Webb M. Microglial activation state and lysophospholipid acid receptor expression. Int J Dev Neurosci. 2003;21:431–443. [CrossRef] [PubMed]
Dolezalova H, Shankar G, Huang MC, Bikle DD, Goetzl EJ. Biochemical regulation of breast cancer cell expression of S1P2 (Edg-5) and S1P3 (Edg-3) G protein-coupled receptors for sphingosine 1-phosphate. J Cell Biochem. 2003;88:732–743. [CrossRef] [PubMed]
Igarashi J, Erwin PA, Dantas AP, Chen H, Michel T. VEGF induces S1P1 receptors in endothelial cells: implications for cross-talk between sphingolipid and growth factor receptors. Proc Natl Acad Sci USA. 2003;100:10664–10669. [CrossRef] [PubMed]
Sah VP, Seasholtz TM, Sagi SA, Brown JH. The role of Rho in G protein-coupled receptor signal transduction. Annu Rev Pharmacol Toxicol. 2000;40:459–489. [CrossRef] [PubMed]
Salomone S, Yoshimura S, Reuter U, et al. S1P3 receptors mediate the potent constriction of cerebral arteries by sphingosine-1-phosphate. Eur J Pharmacol. 2003;469:125–134. [CrossRef] [PubMed]
Yun JK, Kester M. Regulatory role of sphingomyelin metabolites in hypoxia-induced vascular smooth muscle cell proliferation. Arch Biochem Biophys. 2002;408:78–86. [CrossRef] [PubMed]
Sauzeau V, Rolli-Derkinderen M, Lehoux S, Loirand G, Pacaud P. Sildenafil prevents change in RhoA expression induced by chronic hypoxia in rat pulmonary artery. Circ Res. 2003;93:630–637. [CrossRef] [PubMed]
McMurtry IF, Bauer NR, Fagan KA, Nagaoka T, Gebb SA, Oka M. Hypoxia and Rho/Rho-kinase signaling: lung development versus hypoxic pulmonary hypertension. Adv Exp Med Biol. 2003;543:127–137. [PubMed]
Takada Y, Kato C, Kondo S, Korenaga R, Ando J. Cloning of cDNAs encoding G protein-coupled receptor expressed in human endothelial cells exposed to fluid shear stress. Biochem Biophys Res Commun. 1997;240:737–741. [CrossRef] [PubMed]
Wojciak-Stothard B, Ridley AJ. Shear stress-induced endothelial cell polarization is mediated by Rho and Rac but not Cdc42 or PI 3-kinases. J Cell Biol. 2003;161:429–439. [CrossRef] [PubMed]
Figure 1.
 
Expression of Edg receptors in human TM cells and TM tissue was determined by RT-PCR of total RNA from two primary cultures of cells from different donors, age 26 (lane 1) and age 54 (lane 2) and by PCR amplification of two HTM cell plasmid-derived cDNA libraries and one HTM tissue cDNA library, using sequence-specific oligonucleotide primers. Expression of Edg receptors 1 to 4 isoforms was readily detectable in both cell RT-PCR libraries and in one of the cell cDNA libraries, whereas Edg2 was the only isoform amplified from one of the HTM cell-derived cDNA libraries and from the HTM tissue-derived cDNA library.
Figure 1.
 
Expression of Edg receptors in human TM cells and TM tissue was determined by RT-PCR of total RNA from two primary cultures of cells from different donors, age 26 (lane 1) and age 54 (lane 2) and by PCR amplification of two HTM cell plasmid-derived cDNA libraries and one HTM tissue cDNA library, using sequence-specific oligonucleotide primers. Expression of Edg receptors 1 to 4 isoforms was readily detectable in both cell RT-PCR libraries and in one of the cell cDNA libraries, whereas Edg2 was the only isoform amplified from one of the HTM cell-derived cDNA libraries and from the HTM tissue-derived cDNA library.
Figure 2.
 
LPA- and S1P-induced actin cytoskeletal reorganization and formation of focal adhesions in human TM cells. Treatment of serum-starved human TM cells with either LPA (20 μM) or S1P (1 μM) for 1 hour induced formation of actin stress fibers (phalloidin staining) and focal adhesions (vinculin staining) compared with BSA-treated control cells. S1P-treated cells exhibited extensive cortical actin stress fibers with obvious cortical rigidity compared with LPA-treated cells.
Figure 2.
 
LPA- and S1P-induced actin cytoskeletal reorganization and formation of focal adhesions in human TM cells. Treatment of serum-starved human TM cells with either LPA (20 μM) or S1P (1 μM) for 1 hour induced formation of actin stress fibers (phalloidin staining) and focal adhesions (vinculin staining) compared with BSA-treated control cells. S1P-treated cells exhibited extensive cortical actin stress fibers with obvious cortical rigidity compared with LPA-treated cells.
Figure 3.
 
Increased MLC phosphorylation in LPA- and S1P-treated human TM cells. Serum-starved human TM cells treated with either LPA (20 μM) or S1P (1 μM) for 1 hour exhibited increased MLC phosphorylation (di-phosphoform) relative to BSA-treated control cells, as determined by urea/glycerol-polyacrylamide gel electrophoresis followed by (A) Western blot analysis of equal amounts of protein using polyclonal antibody directed against Thr18/Ser19 di-phospho-MLC and (B) Western blot analysis of equal volumes of lysates (prepared from equal numbers of cells) using polyclonal antibody directed against MLC.
Figure 3.
 
Increased MLC phosphorylation in LPA- and S1P-treated human TM cells. Serum-starved human TM cells treated with either LPA (20 μM) or S1P (1 μM) for 1 hour exhibited increased MLC phosphorylation (di-phosphoform) relative to BSA-treated control cells, as determined by urea/glycerol-polyacrylamide gel electrophoresis followed by (A) Western blot analysis of equal amounts of protein using polyclonal antibody directed against Thr18/Ser19 di-phospho-MLC and (B) Western blot analysis of equal volumes of lysates (prepared from equal numbers of cells) using polyclonal antibody directed against MLC.
Figure 4.
 
LPA- and S1P-induced activation of Rho GTPase in human TM cells. Cell lysates containing equal amounts of total protein (300 μg/sample) prepared from serum-starved human TM cells treated with either LPA (20 μM) or S1P (1 μM) demonstrated an increased amount of GTP-Rho in agarose-rhotekin “pull-down” assays compared with BSA control-treated cells, indicating activation of Rho GTPase (GTP-loading). Western blot analysis of nonincubated cell lysates (25 μg/sample) demonstrated no change in total Rho as a result of lysophospholipid treatment.
Figure 4.
 
LPA- and S1P-induced activation of Rho GTPase in human TM cells. Cell lysates containing equal amounts of total protein (300 μg/sample) prepared from serum-starved human TM cells treated with either LPA (20 μM) or S1P (1 μM) demonstrated an increased amount of GTP-Rho in agarose-rhotekin “pull-down” assays compared with BSA control-treated cells, indicating activation of Rho GTPase (GTP-loading). Western blot analysis of nonincubated cell lysates (25 μg/sample) demonstrated no change in total Rho as a result of lysophospholipid treatment.
Figure 5.
 
Mobilization of intracellular calcium in LPA- and S1P-treated human TM cells. Serum-starved human TM cells preloaded with the fluorescent Ca2+ indicator fura-2/AM were subsequently treated with LPA (20 μM) or S1P (10 μM), and intracellular calcium release was determined by spectrofluorometric digital-imaging microscopy. (A) Representative time-dependent intracellular calcium mobilization curves for individual TM cells treated with LPA (20 μM), S1P (10 μM), or BSA control. Arrow: time at which factors were added. (B) The mean peak intracellular calcium concentration of TM cells (clusters of 9–15 cells) treated with LPA (20 μM), S1P (10 μM), or BSA control. HTM cells treated with LPA possessed nearly a twofold higher intracellular calcium concentration than HTM cells treated with S1P.
Figure 5.
 
Mobilization of intracellular calcium in LPA- and S1P-treated human TM cells. Serum-starved human TM cells preloaded with the fluorescent Ca2+ indicator fura-2/AM were subsequently treated with LPA (20 μM) or S1P (10 μM), and intracellular calcium release was determined by spectrofluorometric digital-imaging microscopy. (A) Representative time-dependent intracellular calcium mobilization curves for individual TM cells treated with LPA (20 μM), S1P (10 μM), or BSA control. Arrow: time at which factors were added. (B) The mean peak intracellular calcium concentration of TM cells (clusters of 9–15 cells) treated with LPA (20 μM), S1P (10 μM), or BSA control. HTM cells treated with LPA possessed nearly a twofold higher intracellular calcium concentration than HTM cells treated with S1P.
Figure 6.
 
Pharmacological inhibition of LPA- and S1P-induced changes in MLC phosphorylation in human TM cells. To determine the involvement of Rho kinase, PKC, and MLC kinase in LPA- and S1P-induced changes in MLC phosphorylation, human TM cells were pretreated with specific inhibitors of Rho kinase (5 μM Y-27632), PKC (10 μM GF-109203X), or MLC kinase (25 μM ML-7) and with DMSO control for 20 minutes before treatment with either LPA (20 μM) or S1P (1 μM) for 1 hour. Cell lysates containing equal amounts of protein from treated cells were subjected to urea/glycerol-polyacrylamide gel electrophoresis and Western blot analysis using polyclonal antibody directed against Thr18/Ser19 di-phospho-MLC. Control-pretreated HTM cells exhibited markedly increased levels of di-phospho-MLC when treated with either LPA or S1P. Whereas LPA-mediated increases in di-phospho-MLC were only marginally sensitive to pretreatment with either GF-109203X or ML-7, S1P-mediated increases in di-phospho-MLC were partially inhibited by pretreatment with GF-109203X and were marginally sensitive to ML-7 pretreatment. However, pretreatment of HTM cells with Y-27632 markedly attenuated both LPA- and S1P-induced effects on MLC phosphorylation.
Figure 6.
 
Pharmacological inhibition of LPA- and S1P-induced changes in MLC phosphorylation in human TM cells. To determine the involvement of Rho kinase, PKC, and MLC kinase in LPA- and S1P-induced changes in MLC phosphorylation, human TM cells were pretreated with specific inhibitors of Rho kinase (5 μM Y-27632), PKC (10 μM GF-109203X), or MLC kinase (25 μM ML-7) and with DMSO control for 20 minutes before treatment with either LPA (20 μM) or S1P (1 μM) for 1 hour. Cell lysates containing equal amounts of protein from treated cells were subjected to urea/glycerol-polyacrylamide gel electrophoresis and Western blot analysis using polyclonal antibody directed against Thr18/Ser19 di-phospho-MLC. Control-pretreated HTM cells exhibited markedly increased levels of di-phospho-MLC when treated with either LPA or S1P. Whereas LPA-mediated increases in di-phospho-MLC were only marginally sensitive to pretreatment with either GF-109203X or ML-7, S1P-mediated increases in di-phospho-MLC were partially inhibited by pretreatment with GF-109203X and were marginally sensitive to ML-7 pretreatment. However, pretreatment of HTM cells with Y-27632 markedly attenuated both LPA- and S1P-induced effects on MLC phosphorylation.
Figure 7.
 
Effects of perfusion of LPA and S1P on aqueous outflow facility in enucleated porcine eyes. Freshly enucleated porcine eyes were perfused with either (A) 50 μM LPA or (B) 5 μM S1P at a constant pressure of 15 mm Hg, after establishing the baseline outflow facility with perfusion media containing d-glucose at 33°C. Outflow facility was observed to decrease significantly after 1 hour of either LPA (P < 0.04) or S1P (P < 0.03) treatment, respectively, from the baseline outflow facility. Outflow facility continued to decrease significantly at each of the four remaining time points over the perfusion time course with maximum decreases in outflow facility from baseline of 37% and 31%, for LPA and S1P, respectively. Fellow-paired control eyes perfused with media containing 0.2% fatty acid-free BSA in the LPA series and S1P series showed 14% and 34% increases, respectively, over the corresponding initial baseline outflow facility, after 3 hours.
Figure 7.
 
Effects of perfusion of LPA and S1P on aqueous outflow facility in enucleated porcine eyes. Freshly enucleated porcine eyes were perfused with either (A) 50 μM LPA or (B) 5 μM S1P at a constant pressure of 15 mm Hg, after establishing the baseline outflow facility with perfusion media containing d-glucose at 33°C. Outflow facility was observed to decrease significantly after 1 hour of either LPA (P < 0.04) or S1P (P < 0.03) treatment, respectively, from the baseline outflow facility. Outflow facility continued to decrease significantly at each of the four remaining time points over the perfusion time course with maximum decreases in outflow facility from baseline of 37% and 31%, for LPA and S1P, respectively. Fellow-paired control eyes perfused with media containing 0.2% fatty acid-free BSA in the LPA series and S1P series showed 14% and 34% increases, respectively, over the corresponding initial baseline outflow facility, after 3 hours.
Figure 8.
 
Histologic changes in enucleated porcine eyes perfused with S1P. Enucleated porcine eyes perfused with either 5 μM S1P or 0.2% BSA sham control for 5 hours at 33°C were fixed for histologic examination by both light (A, B) and electron (C, D) microscopy. Both S1P- (B) and control-treated (A) specimens exhibited compact trabecular beams and both had similar-appearing juxtacanalicular regions (C, D, arrows), with no observable difference in filtration space. In addition, filtration space was noted to be free of any accumulated extracellular material. Magnification: (A, B) ×1000; (C) ×2500; and (D) ×2000 of the same specimen, respectively.
Figure 8.
 
Histologic changes in enucleated porcine eyes perfused with S1P. Enucleated porcine eyes perfused with either 5 μM S1P or 0.2% BSA sham control for 5 hours at 33°C were fixed for histologic examination by both light (A, B) and electron (C, D) microscopy. Both S1P- (B) and control-treated (A) specimens exhibited compact trabecular beams and both had similar-appearing juxtacanalicular regions (C, D, arrows), with no observable difference in filtration space. In addition, filtration space was noted to be free of any accumulated extracellular material. Magnification: (A, B) ×1000; (C) ×2500; and (D) ×2000 of the same specimen, respectively.
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×