Modulation of Myocilin/TIGR Expression in Human Trabecular Meshwork

Abstract

purpose. To study factors that modulate myocilin/trabecular meshwork inducible glucocorticoid response protein (TIGR) mRNA expression in human trabecular meshwork (TM).

methods. mRNA from fresh TM of four human donors, from perfused anterior segment organ cultured TM of three donors, and from four primary TM cell lines of different donors was isolated. The full length cDNA of myocilin/TIGR was cloned from TM mRNA using a polymerase chain reaction approach and used as probe for northern blot analysis hybridization. Trabecular meshwork cell cultures were treated with transforming growth factor (TGF)-β1 (1 ng/ml), dexamethasone (10⁻⁷ M), and mechanical stretch (10%).

results. mRNA for myocilin/TIGR could be readily detected by northern blot analysis hybridization in 2 to 3 μg of total RNA from all fresh and all organ-cultured TM samples. In contrast, no mRNA for myocilin/TIGR could be detected in 20 μg of total RNA isolated from three different primary TM cell lines. Only one TM cell line had a baseline expression of myocilin/TIGR, which was 35- to 55-fold lower than that of fresh or organ-cultured TM samples. Treatment of TM cell cultures with dexamethasone for 1 day markedly increased expression of myocilin/TIGR mRNA, an effect that was even more pronounced after 3 days of treatment. Treatment with TGF-β1 for 24 hours had no effect; however, after 3 and 12 days of treatment a 3.8- and 4-fold increase in myocilin/TIGR mRNA expression was observed. Expression of myocilin/TIGR mRNA was also increased after 10% mechanical stretch; however, in contrast to the effects of TGF-β1, this effect was observed much earlier (8–24 hours) after treatment.

conclusions. Dynamic mechanical stimuli maintain myocilin/TIGR expression in TM in situ and lack of these stimuli in monolayer cell cultures might be involved in downregulation of myocilin/TIGR expression.

Elevated intraocular pressure (IOP) in primary open-angle glaucoma (POAG) results from an increase in outflow resistance in the trabecular meshwork (TM), the major site of aqueous humor outflow. So far, the molecular mechanisms involved in normal and abnormal outflow resistance have not been identified. In some forms of POAG, myocilin, a protein that is also known as TM inducible glucocorticoid response protein (TIGR), might be involved. Recently, Stone et al. ¹ identified three mutations in the gene for myocilin/TIGR, which lies within the interval on chromosome 1 that was originally associated with juvenile open angle-glaucoma (GLC1A). ^{2 3 4} Subsequently, mutations in the same gene of patients with GLC1A-linked juvenile open-angle glaucoma were also reported by other researchers. ^{5 6 7 8 9 10 11 12 13} Juvenile open-angle glaucoma refers to a subset of POAG that has an earlier age of onset, has a highly penetrant mode of inheritance, and is usually associated with high IOP that requires early surgical treatment. ^{14 15 16}  

Myocilin/TIGR was originally isolated from cultured human TM cells that had been treated long term with dexamethasone and, independently, from normal human retina. ^{17 18 19} In addition to TM and retina, mRNA for myocilin/TIGR is expressed in various intraocular and extraocular tissues such as cornea, sclera, ciliary body, iris, heart, skeletal muscle, thymus, small intestine, colon, stomach, thyroid, and trachea. ^{5 19 20 21 22 23} The normal role of myocilin/TIGR and the mechanisms by which mutations in this gene cause glaucoma are unknown. In contrast to findings in juvenile glaucoma, mutations in the myocilin/TIGR gene are present only in a minor percentage (≈4.6%) of patients with randomly screened adult forms of POAG. ²⁴ Still, myocilin/TIGR might well also be involved in the pathogenesis of POAG in patients without mutations in the coding sequences of MYOC/TIGR. A recent immunohistochemical study showed increased staining for myocilin/TIGR in the TM of patients with adult-onset POAG when compared with age-matched control eyes. ²⁵ We wished to clarify what factors other than dexamethasone might alter myocilin/TIGR synthesis in human TM. In the present study, we investigated the effects of various culture conditions, transforming growth factor-β1 (TGF-β1), and mechanical stretch on myocilin/TIGR mRNA expression. 

Materials and Methods

Eyes from 11 human donors were obtained from the National Disease Research Interchange (NDRI, Philadelphia, PA) within 48 hours of death. Trabecular meshwork was dissected and immediately frozen from the eyes of four of the donors (age 1 day and 64, 78, and 84 years). Eyes from three other donors (age 60 years, 69 years, 71 years) were placed for 7 days in an anterior segment perfused organ culture system. ²⁶ For monolayer cell cultures, the dissection and explant preparation was similar to that described previously. ^{27 28 29} Primary cultures from four different donors (ages 19, 66, 72, and 80 years) were used. Cells were maintained at 37°C in Dulbecco’s modified Eagle’s medium (DMEM) that was supplemented with 20% fetal bovine serum (FBS) and gentamicin (all from Life Technologies, Gaithersburg, MD). Cell cultures that were at confluence for at least 7 days were treated daily for 8 hours and 1, 3, and 12 days with 1 ng/ml of human recombinant TGF-β1 and for 8 hours and 1 and 3 days with 10⁻⁷ M dexamethasone (all from Sigma, St. Louis, MO). In some experiments, TGF-β1 was added in the presence of serum-free culture medium and after transferring the cells for 24 hours to this medium. Stretch experiments of monolayer cell cultures were performed as described previously. ^{30 31} Cells were plated onto boats formed from silicone sheets, previously coated with laminin (Sigma), attached to end supports that were anchored into a support frame. Notches in the support frame were available to reset boats to a 10% linear stretch. Cells were plated at 1 × 10⁶ cells/sheet, 48 hours before experiments. In some experiments, the media were changed to serum-free DMEM 24 hours before mechanical stretching. The silicone sheets were stretched and maintained in the stretched position for 1, 2, 4, and 8 hours and 1 day, respectively. Each experiment was repeated at least three times. 

RNA Analysis

Total RNA was isolated from fresh TM samples, from TM isolated from organ-cultured eyes, and from monolayer cell cultures using RNAzol (Tel Test; Friendswood, TX). RNA of TM cell cultures treated with dexamethasone for 3 days was used to isolate the full length cDNA of MYOC/TIGR. Reverse transcription–polymerase chain reaction (RT–PCR) was performed using a one-step RT–PCR kit (PCR-Superscript; Life Technologies) according to the manufacturer’s protocol. The RT–PCR was performed in a total volume of 50 μl for 30 minutes at 44°C (RT step), followed by melting at 94°C for 2 minutes, then 30 cycles of 50-second melting at 94°C, 75-second annealing at 55°C, and 2-minute extension at 72°C. After the last cycle, the polymerization step was extended for 10 minutes so that all strands were completed. The primers were designed according to the published structure of the human MYOC/TIGR gene ¹ and were added at a concentration of 2 pmol. The sequences of the primer pairs were 5′-AGCCTCTGCAATGAGGTTC-3′ and 5′-TTACAGCTTTTGCCCCAA-3′. The RT–PCR amplification product was gel-purified, cloned into pCR-Script (Stratagene, La Jolla, CA), and sequenced with fluorescent dideoxynucleotides on an Applied Biosystems (ABI) model 310 automated sequencer (Perkin–Elmer, Foster City, CA). For northern blot analysis experiments, RNA was separated on a 2.2 M formaldehyde–1.2% agarose gel and blotted onto a Duralon (Stratagene) membrane. After the transfer, the blot was cross-linked using a UV Stratalinker (Stratagene). Blots were hybridized with the full length cDNA for human MYOC/TIGR. The probe was labeled with ³²P-dCTP using a random prime kit (Life Technologies). Prehybridizations were performed at 68°C for 1 hour and hybridizations at 68°C overnight using 6× SSC, 5× Denhardt’s solution, and 0.5% sodium dodecyl sulfate (SDS). Membranes were washed twice for 15 minutes each with 2× SSC–0.1% SDS at 68°C and twice with 0.2× SSC–0.1% SDS at room temperature and autoradiographed using a Kodak XAR5 film at −80°C with an intensifying screen (1–2 days). To monitor the integrity of RNA, the relative amounts of RNA loaded on the gel, and the efficiency of transfer, membranes were hybridized to a cDNA probe for guinea pig 18S rRNA or stained with methylene blue. mRNA size was estimated by reference to the mobility of RNA size markers (Life Technologies) stained with methylene blue. Intensity of hybridization was determined by scanning densitometry using a Lumi-Imager and LumiQuant software (both from Boehringer, Mannheim, Germany). Autoradiograms were normalized to the relative intensity of the 18S band. 

Results

The typical amount of total RNA that could be isolated from fresh TM from one adult eye was 1 to 3 μg. In all these samples, mRNA for myocilin/TIGR could be readily detected by northern blot analysis hybridization after exposure times of less than 1 hour (Fig. 1) . The intensity of the signal after normalization for the amount of RNA on the gel did not vary much between TM samples from different adult donors, with the exception of the TM of the 1-day-old donor. In that case, the band for myocilin/TIGR mRNA was considerably weaker (Fig. 2) . The amount of mRNA for myocilin/TIGR in TM from perfused organ cultures appeared to be similar to that seen in fresh TM (Fig. 3) . In contrast, no mRNA for myocilin/TIGR could be detected in 20 μg of total RNA isolated from three different primary TM cell lines (Figs. 1 3) . Only one TM cell line had a baseline expression of myocilin/TIGR, which was relatively 35- to 55-fold lower than that of fresh TM samples (Fig. 1) . In all primary TM cell lines, treatment with dexamethasone for 1 day markedly increased expression of myocilin/TIGR mRNA, an effect that was even more pronounced after 3 days of treatment (Fig. 4) . Treatment with TGF-β1 for 24 hours had no effect (not shown), whereas after 3 and 12 days of treatment a 3.8- and 4-fold increase in myocilin/TIGR mRNA expression was observed (Figs. 5 6) . The same effect was observed when TGF-β1 was added in the absence of serum (not shown). Expression of myocilin/TIGR mRNA was also increased after 10% mechanical stretch and was observed in the presence of serum (Fig. 7) or under serum-free conditions (Fig. 8) . In contrast to the effects of TGF-β1, this effect was already observed 8 to 24 hours after treatment. 

Discussion

In this first direct comparison by northern blot analysis, TM cells in situ express considerable amounts of myocilin/TIGR mRNA, but this expression is markedly downregulated when the cells are transferred to monolayer cell cultures. No mRNA for myocilin/TIGR could be detected in three primary TM cell lines from different donors, whereas only one cell line showed some baseline expression of myocilin/TIGR that was at least 35- to 55-fold lower than that of fresh or organ-cultured TM samples. These results corroborate findings by other researchers. Adam et al. ⁵ showed the presence of myocilin/TIGR mRNA in fresh TM; however, Ortego et al. ²⁰ found no detectable myocilin/TIGR mRNA in cultured TM by northern blot analysis hybridization. It seems unlikely that this apparent downregulation of myocilin/TIGR mRNA in monolayer cell culture is due to factors that are present or absent in serum-enriched cell culture medium compared with aqueous humor, because TM in organ cultures that were perfused with the same medium demonstrated myocilin/TIGR expression similar to that of fresh TM. 

One obvious difference between TM cells in monolayer culture and TM cells in situ is the fact that the latter are under the influence of mechanical factors. In fresh eyes and perfused anterior segment organ cultures, pressure and flow of aqueous humor or culture medium may have caused dynamic interactions to be set up between TM cells and their associated extracellular matrix components. In support of this hypothesis we observed that TM cells in monolayer culture are induced to synthesize increasing amounts of myocilin/TIGR mRNA on mechanical stretch. This induction occurs faster than that we observed with dexamethasone and that has been reported to require prolonged treatment of TM cells for days. ^{18 32} It is tempting to speculate that response to mechanical stretch indicates that myocilin/TIGR is involved in the stability of the cytoskeleton or in cellular adhesion in the TM. Sequence analysis of myocilin/TIGR indicates that the amino terminus has homology with non–muscle myosin. ^{19 32} The myosin-like domain might be associated with the cytoskeleton or the cell membrane, but experimental data in support of this hypothesis are lacking. Recent studies reported that myocilin/TIGR is localized intracellularly around the nucleus and within the cytoplasm, but thus far an association with specific subcellular structures has not been defined. ^{33 34}  

We recently observed distinct changes in protein and mRNA expression in TM cells after mechanical stretch. ³¹ The expression of the immediate-early gene c-fos was depressed, whereas the level of mRNA for c-jun was unchanged. In addition, actin filaments within TM cells rearranged from a diffuse network to complex geodesic patterns. ³¹ Such a rearrangement of actin filaments has been reported to be highly characteristic for TM cells after long-term treatment with dexamethasone. ³⁵ Because both mechanical stretch and treatment with dexamethasone induce myocilin/TIGR expression, we hypothesized that this induction might be associated with a rearrangement of the actin cytoskeleton. Another agent that causes changes in TM actin is TGF-β1, which inducesα -smooth muscle actin positive stress fibers and an overall contractile phenotype. ²⁸ Our studies on the effects of TGF-β1 on myocilin/TIGR expression in TM indeed found an increase in mRNA after treatment. However, the results showed only a moderate increase and much less than that observed after treatment with dexamethasone. Still, the time course of myocilin/TIGR induction after TGF-β treatment was comparable to that observed previously for the induction of α-smooth muscle actin. ²⁸ Clearly, further studies are required to clarify whether there is a direct association between both events. 

We concluded that dynamic mechanical stimuli maintain myocilin/TIGR expression in TM in situ and that lack of these stimuli in monolayer cell cultures might be involved in the downregulation of myocilin/TIGR expression. Myocilin/TIGR has been found in increasing amounts in the TM of patients with adult-onset POAG but also in patients with glaucoma associated with pseudoexfoliation, ²⁵ a secondary type of open-angle glaucoma. Our results might indicate that myocilin/TIGR increases after a mechanical deformation of the TM that is caused by high IOP and that high amounts of myocilin/TIGR in eyes with glaucoma reflect a symptom rather than the cause of high IOP. 

 Supported by grants from the Deutsche Forschungsgemeinschaft (Ta 115/8-1 and Ta 115/11-1); the American Health Assistance Foundation (ERT); the National Eye Institute (EY01894); and Research to Prevent Blindness (DLE).

 Submitted for publication November 30, 1998; revised April 14, 1999; accepted June 3, 1999.

 Commercial relationships policy: N.

 Presented at the annual meeting of the Association for Research in Vision and Ophthalmology, Fort Lauderdale, Florida, May, 1998.

 

Figure 1.

 

Northern blot analysis of myocilin/TIGR mRNA in fresh TM from both eyes of two human donors (lanes 1 and 2, age 78 years; lanes 3 and 4, age 64 years; OD: right eye, OS: left eye) and two primary TM cell lines (lanes 5 and 6) derived from two different human donors. For fresh TM, all RNA that could be isolated from each sample was loaded (1–2 μg of total RNA per lane); for cell cultures 20 μg of total RNA was loaded per lane. The exposure time of the autoradiograph was 1 hour. Relative amounts and integrity of RNA that were loaded were controlled by staining the membrane with methylene blue (lower panel). Relative densitometric intensities (normalized to 18S RNA) of the myocilin/TIGR bands are as follows: lane 1, 55; lane 2, 17.5; lane 3, 38.5; lane 4, 35.7; lane 5, 0; and lane 6, 1. The size of a molecular marker is given in kilobases.

Figure 2.

 

Northern blot analysis of myocilin/TIGR mRNA in TM from both eyes (OD: right eye; OS: left eye) of a 1-day-old human donor (lane 1) and the left eye (OS, lane 2) from an 84-year-old donor. All RNA that could be isolated from each sample was loaded (1–2 μg of total RNA per lane). Relative amounts and integrity of RNA that were loaded were controlled by reprobing the membrane with a cDNA probe specific for guinea pig 18S ribosomal RNA (lower panel). Relative densitometric intensities (normalized to 18S RNA) of the myocilin/TIGR bands are as follows: lane 1, 1; lane 2, 42. The size of a molecular marker is given in kilobases.

Figure 3.

 

Northern blot analysis of myocilin/TIGR mRNA in TM from a perfused anterior segment organ-cultured eye (lane 1) and two primary TM cell lines (lanes 2 and 3) derived from two human donors different from those shown in Figure 1 . For organ-cultured TM, all RNA that could be isolated was loaded (2–3 μg of total RNA); for cell cultures 20 μg of total RNA was loaded per lane. Exposure times were 1 hour and 1 week as indicated. Relative amounts and integrity of RNA that were loaded were controlled by reprobing the membrane with a cDNA probe specific for guinea pig 18S ribosomal RNA (lower panel). The size of a molecular marker is given in kilobases.

Figure 4.

 

Northern blot analysis of myocilin/TIGR mRNA in TM monolayer cell culture after treatment with dexamethasone (10⁻⁷ M) for 8 hours, 1 day, and 3 days. Twenty micrograms of total RNA was loaded per lane. The exposure time of the autoradiograph was 24 hours. Relative amounts and integrity of RNA that were loaded were controlled by reprobing the membrane with a cDNA probe specific for guinea pig 18S ribosomal RNA (lower panel). Relative densitometric intensities (normalized to 18S RNA) of the myocilin/TIGR bands are as follows: lane 1, 1; lane 2, 3; lane 3, 38; and lane 4, 96. The size of a molecular marker is given in kilobases. Co, control.

Figure 5.

 

Northern blot analysis of myocilin/TIGR mRNA in TM monolayer cell culture after treatment with TGF-β1 (1 ng/ml) for 3 days. Twenty micrograms of total RNA was loaded per lane. The exposure time of the autoradiograph was 24 hours. Relative amounts and integrity of RNA that were loaded were controlled by reprobing the membrane with a cDNA probe specific for guinea pig 18S ribosomal RNA (lower panel). Relative densitometric intensities (normalized to 18S RNA) of the myocilin/TIGR bands are as follows: lane 1, 1; and lane 2, 3.8. The size of a molecular marker is given in kilobases. Co, control.

Figure 6.

 

Northern blot analysis of myocilin/TIGR mRNA in TM monolayer cell culture after treatment with TGF-β1 (1 ng/ml) for 12 days. Twenty micrograms of total RNA was loaded per lane. The exposure time of the autoradiograph was 24 hours. Relative amounts and integrity of RNA that were loaded were controlled by reprobing the membrane with a cDNA probe specific for guinea pig 18S ribosomal RNA (lower panel). Relative densitometric intensities (normalized to 18S RNA) of the myocilin/TIGR bands are as follows: lane 1, 1; and lane 2, 4. The size of a molecular marker is given in kilobases. Co, control.

Figure 7.

 

Northern blot analysis of myocilin/TIGR mRNA in TM monolayer cell culture after mechanical stretch (10%) for 1, 2, 4, and 8 hours in medium supplemented with 20% FBS. Twenty micrograms of total RNA was loaded per lane. The exposure time of the autoradiograph was 24 hours. Relative amounts and integrity of RNA that were loaded were controlled by reprobing the membrane with a cDNA probe specific for guinea pig 18S ribosomal RNA (lower panel). Relative densitometric intensities (normalized to 18S RNA) of the myocilin/TIGR bands are as follows: lane 1, 1; lane 2, 0.72; lane 3, 1.3; lane 4, 3.4; and lane 5, 2.7. The size of a molecular marker is given in kilobases. Co, control.

Figure 8.

 

Northern blot analysis of myocilin/TIGR mRNA in TM monolayer cell culture after mechanical stretch (10%) for 8 and 24 hours in serum-free medium. Twenty micrograms of total RNA was loaded per lane. The exposure time of the autoradiograph was 24 hours. Relative amounts and integrity of RNA that were loaded were controlled by reprobing the membrane with a cDNA probe specific for guinea pig 18S ribosomal RNA (lower panel). Relative densitometric intensities (normalized to 18S RNA) of the myocilin/TIGR bands are as follows: lane 1, 1; lane 2, 3.2; and lane 3, 11. The size of a molecular marker is given in kilobases. Co, control.

The authors thank Terete Borrás (Duke University Eye Center, Durham, NC) for providing the guinea pig 18S ribosomal RNA cDNA probe.