Effect of Hevin Deletion in Mice and Characterization in Trabecular Meshwork

Abstract

Purpose.: Hevin is a matricellular protein and the result of a gene duplication of SPARC. SPARC-null mice have lower intraocular pressure (IOP). The function of hevin in trabecular meshwork (TM) is unknown. The authors hypothesized that hevin is expressed in TM and has a functional consequence on IOP.

Methods.: Reverse transcriptase-polymerase chain reaction (RT-PCR) and immunoblotting were performed to identify transcription and protein expression in TM and cultured TM cells. Toluidine blue stain was performed to compare anterior segments in wild-type (WT) and hevin-null mice. Confocal microscopy localized the structural distribution of hevin in human TM and hevin/SPARC in mouse anterior segments. IOP was measured in WT (C57BL6 × 129SvJ) and hevin-null mice using both rebound tonometry and cannulation tonometry. Central corneal thickness (CCT) was measured by ocular coherence tomography. Cultured TM cells were treated with TGF-β2 because TGF-β2 is associated with primary open-angle glaucoma.

Results.: Hevin mRNA and protein were expressed in TM tissues but not in cultured TM cells. No structural differences were observed in anterior segments of WT and hevin-null mice. IOP between hevin-null (n = 46) and WT (n = 44) mice was equivalent (15.3 ± 1.92 mm Hg and 15.9 ± 2.01 mm Hg, respectively; P = 0.15). CCT was similar between hevin-null and WT mice (107.95 ± 5.06 μm and 106.76 ± 3.46 μm, respectively; P = 0.11). TGF-β2 did not induce hevin, whereas SPARC expression was induced in a dose-dependent manner in human TM cell cultures.

Conclusions.: Hevin does not appear to be critical to regulating IOP. Hevin is expressed in TM but, in contrast to SPARC, does not appear to be regulated by TGF-β2.

Primary open angle glaucoma (POAG) is a major cause of blindness, and elevated intraocular pressure (IOP) is a causative factor for the development and progression of disease. ¹ In POAG, visual impairment occurs from loss of the visual field through the progressive loss of retinal ganglion cell axons. One of the primary mechanisms of glaucomatous damage is barotrauma at the level of the lamina cribrosa in the optic nerve, causing mechanical compression and decreased ocular perfusion. ² Lowering IOP is the proven treatment for POAG. ³ The elevated IOP in POAG is caused by increased resistance to aqueous drainage. ⁴  

Aqueous humor drains through both the conventional pathway, in which the juxtacanalicular portion of the trabecular meshwork (TM) is the anatomic location of the highest resistance to outflow, and the uveoscleral pathway. ⁵ In both pathways, the balance of extracellular matrix (ECM) deposition/turnover has been shown to influence IOP. ⁶ The regulatory mechanisms of ECM turnover remain elusive. Transforming growth factor-beta2 (TGF-β2) is elevated in the aqueous humor of up to 50% of patients with POAG. ^7,8 TGF-β2 increases IOP and alters ECM in the juxtacanalicular TM. ⁹ Proteins known to modulate ECM in other tissues may play a prominent role in ECM deposition/turnover and IOP regulation in the TM. 

Matricellular proteins are nonstructural, secreted glycoproteins that regulate ECM turnover in other human tissues. ¹⁰ The matricellular protein family includes SPARC (secreted protein, acidic and rich in cysteine), thrombospondins 1 and 2, tenascins C and X, hevin, and osteopontin. ¹⁰ SPARC, the prototypical matricellular protein, is located within the previously described GLC1M locus, ¹¹ one of the highest transcribed genes in TM, and is found throughout the TM, especially within the JCT region. ¹⁰ In response to mechanical stretch, a physiologic stress to TM, SPARC is one of the most highly upregulated genes in TM cell cultures. ¹² Furthermore, SPARC-null mice have a 15% to 20% lower IOP and enhanced aqueous drainage than their corresponding wild-type mice. ¹³  

Hevin is an evolutionary product of a gene duplication of SPARC and shares a high degree, approximately 65%, of structural similarity with SPARC. ¹⁴ Hevin is known to regulate decorin and collagen fibrillogenesis and to bind to collagen I and myocilin. ^{15 –17} Myocilin causes an accumulation of hevin, and mutated myocilin further impairs hevin secretion. ¹⁶ It is unknown whether hevin is expressed in TM. 

We sought to investigate the effects of hevin deletion in transgenic mice, characterize the expression of hevin in human and murine TM, and assess the response of SPARC and hevin to TGF-β2. Because of the high degree of similarity to SPARC, ¹⁴ we hypothesized that hevin will participate in the regulation of IOP, the transgenic deletion of hevin will result in a lower IOP compared with corresponding wild-type mice, and hevin will be expressed throughout the TM. 

Materials and Methods

Animal Care and Husbandry

All experiments were performed in compliance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Hevin-null mice and their WT (B6.129SF2/J) strain were provided as a generous gift from Renata Pasqualini (MD Anderson Cancer Center, University of Texas, Houston, TX). B6.129SF2/J was the background strain from which the hevin-null mice were created. They were bred independently and genotyped to confirm homozygosity. All animals used in this experiment were born at the Massachusetts Eye and Ear Infirmary animal facility, fed ad libitum, and housed in clear plastic rodent cages under 12-hour light/12-hour dark cycles (on 07:00, off 19:00) at 21°C. At the time of experimentation, the mice were 5 to 8 weeks old. The minimum age of 5 weeks was selected because the mouse iridocorneal angle and its structures reach maturity by this age. ¹⁸ The maximum age of 8 weeks was selected to control for the potentially confounding effect of age when comparing our results with the effects of SPARC deletion in SPARC-null mice. ¹³  

Tissue Samples, Cell Cultures, and Treatment with TGF-β2

All or parts of human trabecular meshwork (HTM) were dissected from corneoscleral rims discarded from corneal surgeries at the Massachusetts Eye and Ear Infirmary (Boston, MA). We have previously demonstrated the suitability of this tissue for molecular biological experiments. ¹⁹ TMs were isolated from the anterior segment, segmented, and randomly assigned for the development of primary cell cultures or whole tissue homogenate. 

Primary HTM endothelial cells were cultured from cadaveric anterior segments of donors aged 42, 43, 44, 45, 47, 52, 55, 59, 62, and 64 years using a previous published protocol. ²⁰ The cultures were maintained in Dulbecco's modified Eagle's media (DMEM; Invitrogen, Carlsbad, CA) containing 20% fetal bovine serum, 1% l-glutamine (2 mM), and gentamicin (0.1 mg/mL) at 37°C in a 10% CO₂ atmosphere. All the cells used were from confluent passages 4 and 5 cultures that had been allowed to differentiate for 3 days. Immortalized murine TM (MTM) endothelial cells were obtained as a generous gift from Paul Russell (University of California, Davis) and have been characterized elsewhere. ²¹ Only one immortalized murine TM cell line is available. Primary murine TM cell cultures were not attempted because of the technical difficulty in isolating TM tissue. 

TGF-β2 (R&D Systems, Minneapolis, MN) was reconstituted in 4 mM HCl solution containing 0.1% human serum albumin according to the manufacturer's instructions and added to HTM endothelial cell cultures at varying doses (0 ng/mL, 2 ng/mL, and 10 ng/mL) in serum-free media for 24 hours at 37°C. Control cells received 4 mM HCl solution containing 0.1% human serum albumin without TGF-β2 (i.e., vehicle). Conditioned media were harvested, and total protein and RNA were isolated from TGF-β2-treated HTM endothelial cells for immunoblot and reverse transcriptase-polymerase chain reaction (RT-PCR), respectively. 

RNA Isolation and RT-PCR

Total RNA from TM tissues and endothelial cells was isolated using reagent (Trizol; Invitrogen) according to the manufacturer's instructions. Briefly, the tissues or cultured cells were lysed with reagent buffer, and chloroform was added. After centrifugation, total nucleic acid was isolated from the aqueous layer using ethanol precipitation. After digestion of DNA by DNase I, total RNA was isolated and stored at −80°C for no more than 1 month before further experiments. 

RT-PCR was performed to detect hevin mRNA in human TM tissues and cell cultures. Complementary DNA (cDNA) was synthesized using an M-MLV RT kit (Promega, Madison, WI) according to the manufacturer's instructions with minor modifications. Briefly, 250 ng total RNA was used as a template in 50 μL reaction mixture, including 250 ng oligo-dT primer, 1× reaction buffer, 0.2 mM dATP, 0.2 mM dCTP, 0.2 mM dGTP, 0.2 mM dTTP, 20 U RNase inhibitor, and 200 U M-MLV reverse transcriptase. The reaction mixture was incubated at 42°C for 1 hour and used for PCR amplification. PCR amplification with 1 μL cDNA was performed for 25 cycles as follows: 30 seconds at 95°C, 30 seconds at 60°C, and 30 seconds at 72°C. The oligonucleotide primers for hevin, SPARC, and glyceraldehyde 3-phosphate dehydrogenase were designed using the Primer 3 program (http://frodo.wi.mit.edu) spanning at least one intron with an expected RT-PCR product of 200 bp. A negative control reaction containing the RNA sample only and DNA polymerase was performed to exclude DNA contamination as a source for the band. Primers used in this experiment are listed in Table 1. 

Table 1.

 

Primer Sequences

	Forward	Reverse	Intron, spanned
Hevin, human	5′-GACCAACAGGGAAAACCTCA-3′	5′-TGCAGGCTCCAAAATAATCC-3′	7
Hevin, murine	5′-GACTGGCGAGAGTGAGAACC-3′	5′-AGGGGGACAAGTCTCTGGAT-3′	5
SPARC, human	5′-GTGCAGAGGAAACCGAAGAG-3′	5′-AAGTGGCAGGAAGAGTCGAA-3′	4 and 5
SPARC, murine	5′-AATTTGAGGACGGTGCAGAG-3′	5′-AAGTGGCAGGAAGAGTCGAA-3′	3 and 4
GAPDH, human	5′-GAGTCAACGGATTTGGTCGT-3′	5′-TGGAAGATGGTGATGGGATT-3′	1 and 2

RT-PCR products were analyzed using electrophoresis on 1.5% agarose gels. Bands were cut from the gel and sequenced at the sequencing core facility at the Massachusetts Eye and Ear Infirmary. Sequences were compared with the published hevin mRNA using the blastn program (http://blast.ncbi.nlm.nih.gov/Blast.cgi). 

Immunoblot Analysis

TM tissues or endothelial cell cultures were homogenized and lysed in 1× radioimmunoprecipitation assay (RIPA) buffer (150 mM NaCl, 1% Igepal [Sigma, St. Louis, MO], 0.5% sodium deoxycholate, 0.1% SDS, 50 mM Tris, pH 8.0) with protease inhibitors (25 mTIU/mL aprotinin, 0.05 mg/mL phenylmethylsulfonyl fluoride, 1 mM EDTA, 1 mM EGTA, 1 μg/mL leupeptin, and 1 mM sodium orthovanadate to a final volume of 1 mL with RIPA buffer). Immunoblot analysis was performed to determine relative protein levels. Equal amounts of total protein from tissues, cell lysates, or conditioned media were mixed with 2× reducing buffer (125 mM of Tris-HCl, pH 6.8, 4% SDS, 20% glycerol, 0.01% bromophenol blue) at a 1:10 ratio and were boiled at 100°C for 3 minutes with reducing (5 mg/mL dithiothreitol) or nonreducing conditions. The samples were analyzed using 10% PAGE. SDS-PAGE was performed at 100 V in tank buffer (250 mM Tris, 192 mM glycine, and 0.1% SDS) using vertical electrophoresis (XCell SureLock Mini-Cell; Invitrogen). The separated proteins were transferred onto a nitrocellulose membrane with 0.45-μm pore size (Invitrogen) in blotting buffer (250 mM Tris, 192 mM glycine, and 10% methanol). The membrane was incubated for 1 hour in 0.5× blocking buffer (Rockland, Gilbertsville, PA) at room temperature. The membrane was then incubated with a primary antibody in 0.5× blocking buffer overnight at 4°C. The next day, the membrane was washed three times with TBS/T (50 mM Tris-HCl, 150 mM NaCl, 0.05% Tween 20) for 10 minutes and then incubated with contrast agent (IRDye 800; Li-Cor Biosciences, Lincoln, NE)–conjugated donkey anti-goat IgG (1:10,000) or goat anti-rabbit IgG (1:10,000; Rockland) for 1 hour. The membrane was washed with 1× TBS/T three times at room temperature for 10 minutes and was scanned (Odyssey Infrared Imaging System; Li-Cor Biosciences). The band density of proteins was quantified using densitometric software (Odyssey, version 1.2; Li-Cor Biosciences). The primary antibodies used in this experiment are listed in Table 2. 

Table 2.

 

Primary Antibodies

Primary Antibody	Dilution in 0.5 × Blocking Buffer
Goat anti–human SPARC IgG	1:1,000
Goat anti–human hevin IgG	1:1,000
Goat anti–murine SPARC IgG	1:1,000
Rat anti–murine SPARC IgG	1:100 (immunostaining only)
Goat anti–murine hevin IgG	1:1,000
Rabbit anti–human GAPDH IgG	1:10,000

 All antibodies were from R&D Systems, Inc.

Histology

Eight-week-old WT and hevin-null mice were euthanatized with CO₂, and the eyes were enucleated for immediate fixation. An incision was made in the central cornea with a surgical blade (no. 11) to facilitate fixation. The globe was placed in fixative consisting of 2.5% glutaraldehyde and 2% formaldehyde in 0.1 M cacodylate buffer with 0.08 M CaCl₂ at 4°C for 24 hours, washed in 0.1 M cacodylate buffer, and postfixed for 1.5 hours in 2% aqueous OsO₄. Tissue was dehydrated in graded concentrations of ethanol, transitioned in propylene oxide, infiltrated with propylene oxide and Epon mixtures (TAAB 812 resin; Marivac, Quebec, Canada), embedded in Epon, and cured for 48 hours at 60°C. Sections (1-μm thick) were cut on a microtome (Ultracut UCT; Leica, Wetzlar, Germany) and stained with 1% toluidine blue in 1% borate buffer for light microscopy. 

Immunofluorescence Staining

Eight-week-old WT and hevin-null mice were euthanatized with CO₂, and the eyes were enucleated for immediate fixation. After the extraocular muscles were trimmed, the enucleated eyes were bisected at the equator, and the lens was gently removed. Human or mouse anterior segment tissues were fixed with 4% paraformaldehyde in PBS (pH 7.4) at 4°C for 24 hours. To make cryosections, tissues were then replaced and dehydrated with 10% sucrose in PBS for 24 hours at 4°C and with 30% sucrose in PBS for another 24 hours at 4°C. The tissue was mounted in plastic (Tissue-Tek; Sakura Finetek, Torrance, CA) and frozen on dry ice. After keeping overnight at −80°C, 20-μm cryosections were made with a cryostat (CM1850; Leica Microsystems, Bannockburn, IL) and stored at −80°C. Paraffin sections were prepared at the core facility of the Massachusetts Eye and Ear Infirmary. 

The cryosections were gently rinsed with 1× PBS. Paraffin sections were washed with xylene, hydrated with EtOH dilution (100%, 95%, and 70%), and rinsed with 1× PBS. After excess liquid was removed, 10% normal donkey serum in PBS was applied and incubated for 1 hour at room temperature. The tissues were permeabilized with 0.2% Triton-100 in 1× PBS for 5 minutes. Primary antibody was then applied to each section at 4°C overnight. Optimal primary antibody concentrations were empirically determined by serial antibody dilution (1:50–1:200). Slides were washed with PBS/T for 10 minutes three times. Labeled tissues were analyzed with a spectral confocal laser scanning microscope (TCS SP2; Leica Microsystems, Exton, PA). In human eyes, representative sections from two quadrants of each eye were analyzed while representative sections from for quadrants were analyzed. The discrepancy was due to half of each of the human eyes being sent for other studies. 

Optical Coherence Tomography

Eyes of adult mice (aged 6 weeks) were imaged using optical coherence tomography (OCT) (Stratus; Carl Zeiss Meditec Inc., Dublin, CA). Under general anesthesia (xylazine/ketamine), mouse eyes were scanned to acquire basic images that were analyzed using the OCT software (Stratus, version 4.0.7; Carl Zeiss Meditec). Central cornea thickness (CCT) was determined by measuring the distance between 2 peaks, which corresponded to the corneal epithelium and endothelium. Measurements were performed in triplicate for each eye by the same investigator who was masked to the genetic status of the mouse. Values obtained were then averaged and reported as means and standard deviations. We have previously validated the use of OCT in mice to estimate CCT against high-frequency ultrasound and histology. ¹³  

Measurement of IOP

IOP of the mice was measured using our previously described protocol. ¹³ Briefly, the mice were anesthetized by intraperitoneal (IP) injection of a ketamine/xylazine mixture (100 mg/kg and 9 mg/kg, respectively; Phoenix Pharmaceutica, St. Joseph, MO). As indicated by the manufacturer, the rebound tonometer (TonoLab, Colonial Medical Supply, Franconia, NH) was fixed horizontally for all measurements, and the tip of the probe was positioned 2 to 3 mm from the eye. To reduce the variability of measurements, the rebound tonometer was modified to include a pedal that activated the probe, obviating handling of the device. The probe contacted the eye perpendicularly at the central cornea. Verification of targeting was performed under direct visualization with 5.5× magnification. A single measurement was accepted only if the device indicated that there was “no significant variability” (per the protocol manual; Colonial Medical Supply). The average IOP was taken from three sets of six measurements of IOP in each eye. ^22,23 All measurements were taken between 4 and 7 minutes after intraperitoneal injection because previous studies have shown this to be a period of stable IOP. ^24,25 IOP measurement per mouse was taken only once between 11 am and 3 pm at 7 weeks of age—1 week after CCT measurement. Previous studies ²⁶ have shown that weekly administration of this anesthesia mixture (ketamine/xylazine) does not affect IOP. Right and left eye measurements were alternated with the initial eye selected randomly. Once IOP was measured, mice were either further processed for immunofluorescence staining or euthanatized. 

Validating Rebound Tonometer Measurements

Validation experiments were repeated to confirm that our IOP measurement technique was consistent with the commercial calibration. The eye was cannulated through the temporal limbus with a 30-gauge needle attached to a water reservoir and a pressure transducer. IOP measurements using the rebound tonometer were taken at various reservoir heights between 10 and 50 mm Hg in random ascending and descending order with the open stopcock technique. ²⁷  

Statistical Analysis

IOPs of WT and hevin-null mice were analyzed with the use of unpaired Student's t-test and one-way ANOVA, followed by all pairwise multiple comparison procedures (Student-Newman-Keuls method) to determine between-group differences. Statistical analysis was carried out with appropriate software (SigmaStat, version 1.0; (Jandel Corporation, San Rafael, CA). P < 0.05 was considered statistically significant. All data were presented as mean ± SD. Sample size (n) referred to the number of eyes measured for IOP, manometry and CCT or to differently aged TM endothelial cell cultures for TGB-β2 treatment. 

Results

Hevin and SPARC RT-PCR were detected in human TM and murine anterior chamber tissue (n = 6 and n = 6, respectively; Fig. 1A). However, hevin mRNA was not detected in human TM endothelial cell cultures, whereas SPARC was detected (n = 6). Because of the limited availability, only one immortalized murine TM cell line ²¹ was used to investigate hevin and SPARC expression. Similar to results with human TM endothelial cell cultures, this murine TM cell line also showed the absence of hevin mRNA (Fig. 1B). Sequencing of the RT-PCR product demonstrated consistency with the mRNA of hevin with an EXPECT value of 6e⁻⁸⁶, where the EXPECT value is the statistical probability that the match is by chance alone. ¹⁰  

Figure 1.

 

Representative images of RT-PCR products on 1.5% agarose gel of hevin and SPARC. RT-PCR products of hevin and SPARC were amplified from the (A) human cDNA library of TM cell cultures and tissues and the (B) murine cDNA library of immortalized TM cell line and anterior chambers, respectively.

Immunoblot analysis of human TM tissue and murine anterior chamber extracts or conditioned media from TM endothelial cell cultures indicated protein expression of hevin in human TM and murine anterior chamber tissue with a band at approximately 115 kDa (n = 6 and n = 6, respectively). Hevin was not seen in the conditioned media of human TM endothelial cell cultures or immortalized murine TM cell line, but it was detected in the TM tissues of human and mice (Fig. 2; n = 6 and n = 1, respectively). 

Figure 2.

 

Representative immunoblots of hevin and SPARC in (A) human TM cell culture and tissue and (B) murine TM cell culture and anterior chamber. The conditioned media from TM cell culture and the extract of TM tissue or murine anterior chamber were analyzed using 10% nonreducing SDS-PAGE.

A single band was detected in human TM tissues, whereas a doublet was detected in murine anterior chambers (n = 6 and n = 6, respectively). In the murine anterior chamber, the doublet of hevin was previously detected and the upper band was identified as hevin by another group. ²⁸ Although not studied in this report, the doublet is the result of different glycosylation levels. ¹⁷ Calculated molecular weights of hevin are approximately 71 kDa in humans and 69 kDa in mice. We found the estimated molecular weight of hevin on 10% nonreducing SDS-PAGE to be approximately 115 kDa; this was not changed on 10% reducing SDS-PAGE (data not shown). Other studies have identified hevin at 115 kDa on SDS-PAGE. ^{29 –31} This discrepancy was explained as a consequence of the interaction between acidic N-terminal domain and buffers of varying pH ²⁹ or ionic strength. ³¹ Hevin protein was not detected in human TM endothelial cell cultures or the murine TM cell line, which was consistent with our RT-PCR results (Fig. 1). 

Toluidine blue staining showed no structural abnormalities of the anterior chamber segment between WT and hevin-null mice (Fig. 3). By light microscopy, the anterior segments appeared indistinguishable with regard to Schlemm's canal, trabecular meshwork, and ciliary body. 

Figure 3.

 

Representative images of murine anterior segments in WT and hevin-null mice. Murine anterior segment was stained by 1% toluidine blue. C, cornea; CB, ciliary body; I, iris; S, Sclera; SC, Schlemm's canal; TM, trabecular meshwork.

In human and murine anterior segments (n = 5 and n = 8, respectively), immunofluorescence microscopy of the TM showed minimal staining attributable to hevin. Within the TM, staining was seen in the juxtacanalicular region of TM (Fig. 4). Weak hevin signaling was seen in the endothelia of inner and outer walls of Schlemm's canal in both human and murine anterior segments. 

Figure 4.

 

Representative images of (A) hevin expression and distribution in the juxtacanalicular region of a human donor aged 45 years and (B) hevin and SPARC in the murine anterior segment. Hevin was detected in human and WT murine TM tissue. No staining was seen in hevin-null mice. SPARC was localized as a counterpart of hevin in the anterior segment of WT and hevin-null mice. KO, knockout; AC, anterior chamber; C, cornea; CB, ciliary body; I, iris; S, sclera; SC, Schlemm's canal; TM, trabecular meshwork; 1^st, primary antibody; 2nd, secondary antibody. Scale bars: 10 μm (A); 30 μm (B). *Region in which hevin was detected in TM.

Manometric calibration validated our rebound tonometer IOP measurements. These IOP measurements were correlated linearly with manometric pressure. The correlation was calculated as follows: tonometer IOP = 1.01 × IOP_manometric + 0.37 (R ² = 0.98; n = 5) for WT mice and tonometer IOP = 1.01 × IOP_manometric − 0.21 (R ² = 0.99; n = 3) for hevin-null mice. There was good consistency between the reservoir height and pressure transducer recordings (Fig. 5). 

Figure 5.

 

Manometric calibration of rebound tonometer for WT (n = 5) and hevin-null (n = 3) mice. Linear regression formulas are calculated using each value, with a mean of six individual measurements.

WT and hevin-null mice had similar IOPs. The average IOP was 15.9 ± 1.9 mm Hg for WT (n = 46) and 15.3 ± 2.0 mm Hg for hevin-null mice (n = 44; P = 0.15). CCTs were also similar between WT and hevin-null mice: 107.9 ± 5.0 μm for WT (n = 44) and 106.8 ± 3.5 μm for hevin-null mice (n = 42; P = 0.11). 

Increased TGF-β2 in aqueous humor is associated with IOP elevation in POAG. ^7,8 In this study, we found that SPARC was dose dependently induced by TGF-β2 in TM endothelial cell cultures. Because of the structural similarity between SPARC and hevin, we investigated whether hevin was induced by TGF-β2 and found that it was not induced by TGF-β2 in human TM endothelial cells (Fig. 6; n = 6 for 2 ng/mL; n = 3 for 10 ng/mL) compared with levels in control samples. 

Figure 6.

 

Representative expression pattern of hevin and SPARC on TGF-β2 treatment in human TM endothelial cell cultures. Hevin and SPARC expression were analyzed by (A) immunoblot analysis. CL, cell lysates; CM, conditioned media. (B) Fold changes of hevin and SPARC in CM were calculated from densitometric analyses of band sizes (n = 6 for 2 ng/mL; n = 3 for 10 ng/mL). References for comparison were the control samples. (C) RT-PCR products of SPARC and hevin (n = 6 for 2 ng/mL; n = 3 for 10 ng/mL) were analyzed on 1.5% agarose gels. RT-PCR products of hevin and SPARC were amplified from a cDNA library of HTM endothelial cells. Protein and mRNA from human TM tissues were used as a positive control for hevin and SPARC expression. P values were calculated using Student's t-test compared against control samples.

Discussion

Hevin plays an important role in ECM equilibrium in other tissues and organs ^15,17 but does not appear to be relevant to IOP regulation. We previously demonstrated that SPARC-null mice have a lower IOP and enhanced aqueous outflow. ¹³ Hevin and SPARC have a 65% similarity of amino acid sequence. The differences are primarily in the N-terminal sequences of SPARC, ¹⁴ implicating this region as containing either a critical factor or a suppressing element to account for their functional differences in IOP. The low expression in TM and the different response to TGF-β2 might have a contributory role in the differential effect on IOP. In perfused human anterior chambers, TGF-β2 elevates IOP by increasing certain ECM components, such as fibronectin and PAI-1. ⁹ Hevin does not appear to be regulated by TGF-β2. 

We found that hevin was not detected at the mRNA or protein level in human TM endothelial cell cultures or the immortalized murine TM cell line, whereas SPARC was highly expressed. ²⁸ Downregulation of hevin mRNA has been seen with endothelial cells from several different human and murine tissues, implicating possible regulatory factors that are lacking in vitro. ¹⁴ Thus, hevin expression in the TM may be regulated by the microarchitecture or microenvironmental factors secreted by the ciliary body, iris, or cornea not present in vitro. For example, TM endothelial cells express genes differently on different topography and, if grown, aqueous humor. ^32,33 However, hevin was expressed in very low amounts and was barely detectable in the juxtacanalicular region of human and mouse in vivo; conversely, SPARC is highly expressed throughout human and mouse anterior segments. ¹³  

Light microscopy revealed no structural differences in the anterior segments of hevin-null mice compared with WT mice, consistent with previous reports. ³⁴ Regarding differences of expression between human and mouse, hevin was expressed in very low amounts and was barely detectable in the juxtacanalicular region of both human and mouse in vivo; conversely, SPARC is highly expressed throughout human and mouse anterior segments. ¹³ We did not find any differences between nonglaucomatous human and WT murine expression of hevin. 

We demonstrated a linear correlation between manometry and rebound tonometry, indicating that CCT did not have a consequence on IOP measurement in mice. CCTs between WT and hevin-null mice were similar and without physiologic significance. In humans, CCTs outside the average range cause artifactual differences with Goldmann applanation tonometry. ³⁵ Rebound tonometry in mice does not seem to be subject to the same artifact as Goldmann applanation tonometry in humans. ³⁶  

Hevin has been observed to have a compensatory effect for SPARC as a tumor suppressor ³⁷ and a counter adhesive in some cultured cells. ¹⁴ Our study showed hevin and SPARC have different expression patterns in the TM. Hevin-null mice did not have lower IOP, which indicated that SPARC may compensate for hevin. However, SPARC-null mice do have lower IOP, ¹³ indicating that hevin is unable to compensate for SPARC with regard to IOP regulation. Thus, we surmise that hevin and SPARC have distinct functions, especially with regard to IOP regulation in TM. The similarity in IOP between WT and hevin-null mice indicates that hevin is not likely involved in regulating IOP. Low expression in the TM, downregulation in vitro, and lack of response to TGF-β2 may account for the lack of relevance to IOP. Further study into the structural differences between SPARC and hevin may elucidate the regulatory elements of SPARC that are responsible for regulation of IOP. 

Footnotes

 Supported by an American Glaucoma Society Mid-Career Award and by National Eye Institute Grants EY 019654-01 (DJR) and EY14104 (MEEI Vision-Core Grant).

Footnotes

 Disclosure: M.H. Kang, None; D.-J. Oh, None; D.J. Rhee, None

The authors thank Renata Pasqualini (MD Anderson Cancer Center, University of Texas) and Paul Russell (University of California, Davis) for the generous gifts of hevin-null and WT mice and immortalized murine trabecular meshwork endothelial cells, respectively.