Molecular Vision 2006; 12:774-790 <http://www.molvis.org/molvis/v12/a87/> Received 12 April 2005 | Accepted 22 June 2006 | Published 12 July 2006

Download Reprint

Genome-wide expression profile of human trabecular meshwork cultured cells, nonglaucomatous and primary open angle glaucoma tissue

Paloma B. Liton, Coralia Luna, Pratap Challa, David L. Epstein, Pedro Gonzalez
 
 

Department of Ophthalmology, Duke University, Durham, NC

Correspondence to: Pedro Gonzalez, Duke University Eye Center, Erwin Road, Box 3802, Durham, NC, 27710; Phone: (919) 681-5995; FAX: (919) 684-8983; email: pedro.gonzalez@duke.edu

Abstract

Purpose: To contrast genome-wide gene expression profiles of cultured human trabecular meshwork (HTM) cells to that of control and primary open angle glaucoma (POAG) HTM tissues.

Methods: Cultured HTM cells, HTM tissue dissected from control donors, and HTM tissue from POAG donors receiving medication for glaucoma were fixed in RNA later^TM. Total RNA extracted from these samples was linearly amplified with the Ovation Biotin RNA Amplification and Labeling System and individually hybridized to Affymetrix Human Genome U133 Plus 2.0 high density microarrays. Data analysis was performed using GeneSpring Software 7.0. Selected genes showing significant differential expression were validated by quantitative real-time PCR in nonamplified RNA.

Results: Cultured HTM cells retained the expression of some genes characteristic of HTM tissue, including chitinase 3-like 1 and matrix Gla protein, but demonstrated downregulation of physiologically important genes such as myocilin. POAG HTM tissue showed relatively small changes compared to that of control donors. These changes included the statistically significant upregulation of several genes associated with inflammation and acute-phase response, including selectin-E (ELAM-I), as well as the downregulation of the antioxidants paraoxonase 3 and ceruloplasmin.

Conclusions: Downregulation in cultured HTM cells of genes potentially relevant for outflow pathway function highlights the importance of developing new conditions for the culture of TM cells capable of preserving the characteristics of TM cells in vivo. Comparative analysis between control and POAG tissues suggests that the upregulation of inflammation-associated genes might be involved in the progression of glaucoma.

Introduction

The conventional outflow pathway, consisting of the trabecular meshwork (TM) and Schlemm's canal (SC), is a highly specialized tissue located at the angle formed by the cornea and iris. This tissue is involved in intraocular pressure (IOP) homeostasis by modulating the outflow of aqueous humor (AH) from the anterior chamber to the venous system. Increased IOP resulting from abnormally high outflow resistance is commonly associated with primary open angle glaucoma (POAG) [1-3].

In addition to modulating AH outflow resistance, the conventional outflow pathway is believed to be involved in detoxification of the AH, phagocytosis of cellular debris, and the maintenance of immune privilege in the eye. To accomplish all these functions, the conventional outflow pathway is organized, despite its small size (100-150 μg, containing approximately 200,000-300,000 cells), as a complex structure composed of morphologically and functionally different cell types: the Schwalbe's line (SL) cells, proposed to be the progenitor cells of the TM [4,5]; TM cells, involved in phagocytosis and tissue remodeling; and juxtacanalicular tissue (JCT) cells that, together with the cells of the inner wall of SC, contain the locus for outflow resistance [6,7]. The physiological mechanisms by which the TM/SC outflow pathway regulates the outflow of AH, as well as the cause for the increase in resistance leading to elevated IOP in POAG, remain unknown.

Over the last decade, gene expression profiling has emerged as one of the most powerful approaches for dissecting the regulatory mechanisms and transcriptional networks that underlie biological processes. Although several laboratories, including ours, have extensively worked to define the gene expression profile of the outflow pathway, there is still incomplete information available, mainly due to the small amount of RNA that can be isolated from each sample. To date, two different cDNA libraries from human TM (HTM) tissue have been reported. Using single-pass sequencing, Gonzalez et al. [8] described the 833 genes more highly expressed in a PCR-amplified cDNA library constructed from the TM of a perfused human anterior segment of a single individual. Tomarev et al. [9] identified the 3,459 genes more highly expressed in a cDNA library constructed from a pool of native TM from 28 donors. Although these two libraries have provided important information regarding the genes that are more highly expressed in the TM, we still lack critical information about genes with lower expression that still might be essential for tissue function.

To generate a genome-wide gene expression profile of the TM and identify genes associated with the maintenance of tissue physiology, we performed microarray analysis in both native TM tissue and cultured HTM cells. We used the new Affymetrix Human Genome U133 Plus 2.0 Array in conjunction with a novel linear mRNA amplification method, Ribo-SPIA, recently introduced by NuGEN technologies. In addition, we have analyzed, for the first time, the gene expression profile of TM tissue from POAG donors.

Methods

Human tissue procurement

Human cadaver eyes from donors were obtained from the North Carolina Eye Bank (NCEB). The anterior segments of the five pairs of eyes used for the gene expression profile of TM tissue included three control pairs of eyes without known history of glaucoma and two pairs of eyes from donors with documented history of POAG and glaucoma medication (Table 1). All eyes were enucleated less than 6 h postmortem and immediately fixed in RNA later^TM (Ambion Inc., Austin, TX) to preserve RNA integrity. The anterior segments of the three pairs of eyes used for gene expression profile of cultured TM cells were obtained less than 24 h postmortem and immediately immersed in Optisol. Ocular histories were carefully reviewed by a certified ophthalmologist. Tissues from eye donors were manipulated at all times in accordance to the Tenets of the Declaration of Helsinki.

Human trabeculr meshwork primary cultures

Primary cultures of HTM were prepared from cadaver eyes following previously described guidelines [10] and maintained at 37 °C in a humidified atmosphere of 5% CO₂ in low glucose Dulbecco's Modified Eagle Medium (DMEM) with L-glutamine and 110 mg/l sodium pyruvate, supplemented with 10% fetal bovine serum (FBS), 100 μM nonessential amino acids, 100 units/ml penicillin, 100 μg/ml streptomycin sulfate and 0.25 μg/ml amphotericin B. All reagents were obtained from Invitrogen Corporation (Carlsbad, CA).

RNA extraction and quality analysis

Total RNA, from dissected TM tissues and RNA later-fixed cultured HTM cells at passage three, was isolated using the RNeasy kit (Qiagen Inc., Valencia, CA) following the manufacturer's protocol. After DNase treatment, RNA yields were determined using the RiboGreen® fluorescent dye (Molecular Probes Inc., Eugene, OR). RNA quality was confirmed by assessing the ratio of ribosomal bands 28S and 18S, using the Agilent 2100 Bioanalyzer.

RNA amplification and cDNA biotin labeling

RNA samples were amplified using the Ovation^TM Biotin RNA amplification and Labeling System (NuGEN Technologies Inc., San Carlos, CA) according to the manufacturer's instructions. Briefly, total RNA (20 ng) was reverse transcribed into cDNA using a reverse transcriptase and a DNA/RNA chimeric primer. This strand is copied by a DNA polymerase with reverse transcriptase activity to give double-stranded cDNA with an RNA-DNA heteroduplex at one end. This double-stranded cDNA is amplified using the novel isothermal linear amplification method Ribo-SPIA^TM (NuGEN Technologies Inc.). The cDNA amplification products were fragmented and labeled to generate biotinylated cDNA targets.

Oligonucleotide microarray analysis

Biotinylated cDNA targets were hybridized on Affymetrix Human Genome U133 Plus 2.0 high density microarrays following the manufacturer's instructions. A total of eight cDNA target preparations were performed, and each preparation was analyzed using one microarray. Data analysis was performed using GeneSpring Software 7.0 (Silicon Genetics, Redwood City, CA). Raw data from eight hybridizations were normalized to the fiftieth percentile per chip and to median per gene. Normalized mean values for the three individual experimental groups (cultured TM cells, control, and POAG) were generated for the experimental interpretation. Genes with a differential expression of two fold were selected and then filtered on flags to retain the genes that were presented in all samples of at least one of the experimental groups. Statistical analyses were performed by using the cross-gene-error model in combination with a one-way ANOVA (p<0.05). Since some genes were represented in the arrays in more than one spot, we verified a consistent differential expression in all the spots to eliminate false positives.

Quantitative real-time PCR

First strand cDNA was synthesized from a pool of total RNA (0.5 μg, equal amounts of each individual sample) by reverse transcription using oligodT primer and Superscript II reverse transcriptase (Invitrogen, Carlsbad, CA) according to the manufacturer's instructions. Real-time PCR reactions were performed in a 20 μl mixture containing 1 μl of the cDNA preparation, 1X iQ SYBR Green Supermix (Biorad, Hercules, CA) and 500 nm of each primer, in the Bio-Rad iCycler iQ system (BioRad, Hercules, CA) using the following PCR parameters: 95 °C for 5 min followed by 50 cycles of 95 °C for 15 s, 60 °C for 15 s, and 72 °C for 15 s. The fluorescence threshold value (Ct) was determined using the iCycle iQ system software. Expression levels were represented as the inverse of the normalized mean Ct value (InvCt). Eukaryotic translation initiation factor 4E (EIF4E) served as an internal standard of mRNA expression. The absence of nonspecific products was confirmed by both the analysis of the melt curves and electrophoresis in 3% Super AcrylAgarose gels. The sequences of the primers used for the amplifications are indicated in Table 2.

Results

Gene expression profile of trabecular meshwork

To generate a genome-wide comprehensive gene expression profile of the TM, total RNA (20 ng) from three different cultured TM cell lines and the TM from three pairs of control donors was amplified, independently hybridized to Affymetrix Human Genome U133 Plus 2.0 GeneChip microarrays, and analyzed as detailed in the Methods. The full list of gene expression data, including those for the two additional samples from POAG donors, is published online in the Gene Expression Omnibus (GEO) database (accession number: GSE4316). Figure 1 summarizes the number of transcripts expressed in cultured cells and TM tissue. Table 3 summarizes the 25 genes sharing the highest signal values in both cultured and native TM cells compared with the previously published cDNA libraries from TM tissue [8,9], a cDNA library from infant TM primary culture [11], and the gene expression profile of nonamplified HTM primary cultures using the U95Av2 Affymetrix microarrays (12,626 probe sets) [12]. To simplify the data, we have excluded from the list the generally highly expressed ribosomal proteins and elongation factors. Matrix Gla Protein (MGP), chitinase 3-like 1 (CHI3L1), and the ubiquitously expressed tumor protein translationally controlled 1 (TPT1) were found to be present at high levels in at least three out of the five studies.

Genes differentially expressed in trabecular meshwork tissue and human trabecular meshwork cultured cells

The gene expression profile of cultured TM cells was compared to that of control TM tissue. The full lists of genes differentially expressed at least two fold with a p value lower than 0.05 is included in Appendix 1. In summary, 657 genes were found to be upregulated and 2,646 genes were downregulated in cultured TM cells compared to native TM tissue. Figure 2 represents the functional distribution of the differentially expressed genes. It is worth noting that most of the genes showing the highest upregulated differential fold expression are involved in the maintenance of the extracellular matrix (ECM), including matricellular proteins (microfibril-associated glycoprotein-2 and spondin 2), leptins (fibronectin-1 and vitronectin), several types of collagens, as well as proteins involved in the synthesis of the ECM (cartilage linking protein 1 and lysyl oxidase; Table 4). Among the genes downregulated in cultured TM cells, the highest proportion (43%) corresponded to unknown genes or genes still not associated with cellular function. Table 5 lists genes down regulated in cultured TM cells whose functions may be important or have already been associated with the maintenance of TM tissue functionality, such as cochlin, aquaporin, and adenosine A3 receptor. Of particular interest is the downregulation of the genes highly expressed in native TM tissue: myocilin, angiopoietin-like factor, and apolipoprotein D (Appendix 1; Downregulared genes).

Differential gene expression profile analysis between control and primary open angle glaucoma donors

The gene expression profile of the TM from the two analyzed POAG samples revealed relatively small changes compared to that of control donors. A total of 156 genes showed a significant average fold-change value greater than 2 in POAG tissue; 72 genes were upregulated (Table 6), and 84 genes were downregulated (Table 7). Among all these genes, selectin-E (ELAM-I) demonstrated the highest differential transcriptional levels (31 fold induction), as well as restricted expression to POAG samples. mRNA for this gene was not detected in either control TM tissue or cultured TM cells. Together with selectin-E, the TM from POAG tissue showed an upregulation of several genes coding for proteins involved in inflammation and acute-phase response, such as chemokine (C-X-C motif) ligand 6, chemokine (C-C motif) ligand 5, immune-associated nucleotide, interleukin 1 receptor type II, transthyretin, haptoglobin, and myelin basic protein. Other genes upregulated in POAG samples included ELOVL7 and oxisterol binding protein-like 6, both genes involved in lipid metabolism; several genes related to G-protein signaling regulation, like prokineticin 2, G protein-coupled receptor 146, regulator of G-protein signaling 1, neuropeptide Y receptor Y2, and adenosine A3 receptor; as well as genes involved in ionic transport, such as transient receptor potential cation channel and chloride intracellular channel 2.

Genes downregulated in POAG donors included two members of the vacuolar protein sortin-10 domain receptor family, SORCS3 and SORL1; both the WNT-inhibitory factor 1 and the receptor frizzled homolog 10; several members of the solute carrier family, the mitogen-activated protein kinase 6, ceruloplasmin, as well as paraoxonase 3, an inhibitor of LDL oxidation.

Confirmation of gene microarray analysis by real-time PCR

To further confirm the results of the microarray analysis, we chose a subset of genes differentially expressed among cultured TM cells, control TM tissue, and TM tissue from POAG donors for validation by real-time PCR analysis in nonamplified RNA. With the exception of some small differences at the level of statistical significance, these 24 genes were shown to have consistent trends of change by both microarray analysis and real-time quantitative PCR (Figure 3).

Glaucoma candidate genes

Among all the genes expressed in TM tissue, 929 genes were identified in the 11 loci described to be linked to different forms of glaucoma (Table 8). The complete list of genes is available upon request. To narrow the search for glaucoma candidate genes, we also evaluated the chromosomal location of the genes significantly downregulated in cultured HTM cells compared to TM tissue (Table 9), as well as the genes showing significant differential expression between control and POAG TM tissue (Table 10).

Discussion

In this study, we present for the first time, the genome-wide gene expression profile of both cultured HTM cells and native TM tissue, including that of POAG donors, using Affymetrix gene microarrays in conjunction with the novel linear mRNA amplification method, Ribo-SPIA^TM. This amplification tool, which has been demonstrated to provide an accurate and reliable representation of the transcripts [13,14], allowed us individual hybridizations, thereby preventing potential false positives that might result from pooling RNA from different donors. We performed quantitative real-time PCR analysis of a subset of genes in nonamplified RNA to validate the use of this technique to overcome the limitation of TM size in this type of study.

The comparative gene expression profile analysis of cultured HTM cells showed a high level of similarity to that of native tissue. More than 90% of the genes expressed in the HTM cells in vitro were also present in TM tissue, indicating that cells growing in the culture plates do indeed derive from the TM. To date, no gene strictly specific to the TM has been uncovered. However, the consistent high levels of expression of MGP and CHI3L1 in the TM, together with their restricted pattern of tissue expression, strengthen the potential use of MGP and CHI3L1 as markers for cultured TM cells [12,15].

Despite the high level of similarity, important changes at the gene expression level were found when cultured TM cells were compared to native TM tissue. As recently pointed out by Fautsch et al. [16], TM cells, which rarely replicate in vivo, are "forced" to replicate and grow in bi-dimensional plastic substrates when cultured in vitro. The adaptation to this artificial architecture and environmental cues likely explains the observed increased cellular activity in cultured HTM cells, as well as the upregulation of ECM genes, which are required for the formation of a new biological substrate.

Potentially more relevant to the understanding of outflow pathway physiology are the genes downregulated or even lost in cultured HTM cells. In addition to a three-dimensional structure and the presence of aqueous humor rather than culture rich-media, cells in the outflow pathway are continuously subjected to mechanical forces in situ [17,18]. The absence of such factors in culture conditions could affect the expression of genes essential for outflow pathway functionality in vivo. In addition, since the outflow pathway is composed of different cell types, we cannot rule out the possibility that genes downregulated in cultured conditions are normally expressed in the tissue by cell types that may be underrepresented or absent in primary cultures of HTM cells at passage 3 due to drift in cellular representation. Finally, differences in donor age and post-mortem time could potentially exert some influences in the pattern of gene expression of culture cells.

Although no strictly TM specific genes were identified in this study, many of the genes downregulated in cell culture, such as podocalyxin and angiopoietin-like factor, are known to be involved in specialized functions or present a restricted pattern of tissue distribution indicative of their involvement in specific tissue functions. In contrast, most housekeeping genes did not showed a significant change in the levels of expression in cultured cells. These results suggest some levels of dedifferentiation when subjected to cultured conditions.

Both gene arrays and real-time PCR analysis confirmed the downregulation of three genes highly represented in native TM tissue: MYOC, CDT6, and APOD. Lower expression of MYOC in TM cell primary cultures and perfused TM tissue has already been reported [8,11]. Interestingly, expression of MYOC is rescued when HTM cells are cultured in DMEM supplemented with aqueous humor instead of standard serum [16], suggesting a critical role of aqueous humor composition for the transcription of MYOC. It is more difficult to speculate about the potential causes leading to the downregulation of CDT6 and APOD. CDT6, located within the glaucoma locus GLC3B, encodes a secreted angiopoietin-like factor that is also highly expressed in the human corneal stroma where it is thought to influence the deposition of the ECM [19]. The upregulated expression of CDT6 in cultured HTM cells has been reported after treatment with TGF-β1 and TGF-β2 [20], factors known to be increased in the aqueous humor of pseudoexfoliation glaucoma and POAG donors, respectively. The function of APOD, a carrier protein member of the lipocalins family, and its possible role in the TM remain yet to be determined. Interestingly, the transcription of this gene was significantly induced in the TM of perfused human anterior segments subjected to elevated intraocular pressure [21].

Other genes showing significant downregulated expression in cultured HTM cells and reported to be potentially important in the maintenance of outflow pathway function were aquaporin 1 (AQP1) and adenosine A3 receptor (ADORA3). AQP1 has been previously localized in the endothelium of the TM and SC in addition to the nonpigmented epithelium of the ciliary processes and the iris epithelium [22,23]. The exact role of AQP1 expression in TM and SC cell function has not yet been demonstrated, although it was hypothesized to influence osmotic permeability of the TM plasma membrane as well as the resting intracellular volume and, thus possibly paracellular permeability [24]. AQP1 deletion in mice has been shown, instead, to decrease IOP by reducing the aqueous humor secretion without affecting outflow resistance [25]. A similar role in modulating IOP by altering both aqueous humor production and outflow facility has been proposed for adenosine receptors [26-28]. ADORA3, in particular, has been shown to increase the rate of aqueous humor secretion by activating the chloride channels in the nonpigmented ciliary epithelial cells [29,30]. This gene has recently been reported to be selectively upregulated in the ciliary epithelium of eyes with pseudoexfoliation syndrome and glaucoma [31]. Interestingly, our analysis also demonstrated increased expression of ADORA3 in the TM of POAG donors. Given its protective role in extraocular tissues against oxidative damage [32,33], as well as its anti-inflammatory effects [34], ADORA3 may be particularly important in pathophysiological mechanisms in the outflow pathway.

Consistent with the absence of dramatic morphological changes, a relatively small number of genes showed significant differential expression between the TM from POAG donors and control samples, which resulted mostly from the high levels of individual variability among the samples. Since stringent criteria were applied in the analysis of the data in order to obtain confident results despite the individual variation and sample size, the small number of observed changes is likely to be an underestimate. It also has to be taken in consideration that, due to the difficulty in obtaining samples from untreated donors, a general limitation of this and other studies including POAG donor tissues results from the fact that glaucoma medication may exert effects on the levels of expression of certain genes. Expression profile analysis did not indicate the upregulation of fibrosis- or calcification-associated genes in the POAG phenotype. Likewise, we did not find increased levels of cochlin mRNA expression associated with POAG, suggesting that the reported accumulation of cochlin in the POAG TM [35,36] might result from decreased protein degradation rather than increased synthesis. The TM from POAG donors showed upregulation of several genes involved in inflammatory and acute-phase responses, including the expression of a previously reported molecular marker of the glaucoma disease phenotype, selectin-E (ELAM-1) [37], which, interestingly, was not found to be expressed either in the control TM tissues or in cultured TM cells. A similar inflammatory phenotype accompanies a large number of age-related diseases such as atherosclerosis, Alzheimer's disease, Parkinson's disease, and rheumatoid arthritis [38].

The expression of inflammatory molecules in aged tissues is believed to result from the production of reactive oxygen species (ROS) and free-radical chain reactions generated from lipid peroxidation [39]. The generation of ROS, which may initiate or contribute to the progression of glaucoma, is likely to occur in the TM, a tissue constantly exposed to an oxidative environment [40,41]. Indeed, decreased antioxidant potential [42,43], increased expression of oxidative stress markers [43], as well as increased oxidative DNA damage [44] and peroxized lipids [45] have been described in the TM of glaucoma patients. Moreover, we have recently reported an accumulation of senescent cells in the glaucomatous outflow pathway [46], which may, in turn, increase the generation of ROS. In addition, the TM from POAG donors demonstrated downregulated expression of the antioxidants, paraoxonase 3 and ceruloplasmin. Interestingly, ceruloplasmin, which also showed downregulated expression in cultured HTM cells and whose mutations are related to several diseases [47-49], is located in the glaucoma locus GLC1C.

In summary, we have generated and compared, for the first time, the genome-wide gene expression profile of cultured HTM cells and TM tissue from both control and POAG donors. Although cultured HTM preserved the TM markers, CHI3L1 and MGP, several differences in gene expression compared to the native TM tissue highlight the importance of developing new conditions for the culture of TM cells capable of preserving the characteristics of TM cells in vivo. Likewise, our profile analysis of POAG TM tissue shows that the TM in POAG does not experience extensive changes in gene expression. Our data suggests that upregulation of genes involved in inflammation and acute phase response might initiate or contribute to the progression of glaucoma. However, the observation of great levels of individual variability indicated that the analysis of a relatively high number of samples will be necessary in the future to obtain a reliable glaucoma signature.

Acknowledgements

This work was supported in part by the Research to Prevent Blindness Foundation, and NIH grants EY05722, EY01894, and EY016228.

References

1. Sommer A. Intraocular pressure and glaucoma. Am J Ophthalmol 1989; 107:186-8.

2. Bill A, Phillips CI. Uveoscleral drainage of aqueous humour in human eyes. Exp Eye Res 1971; 12:275-81.

3. Quigley HA. Open-angle glaucoma. N Engl J Med 1993; 328:1097-106.

4. Raviola G. Schwalbe line's cells: a new cell type in the trabecular meshwork of Macaca mulatta. Invest Ophthalmol Vis Sci 1982; 22:45-56.

5. Acott TS, Samples JR, Bradley JM, Bacon DR, Bylsma SS, Van Buskirk EM. Trabecular repopulation by anterior trabecular meshwork cells after laser trabeculoplasty. Am J Ophthalmol 1989; 107:1-6.

6. Tripathi RC, Tripathi BJ. Functional anatomy of the chamber angle. In: Jakobiec FA, editor. Ocular anatomy, embryology and teratology. Philadelphia: Harper & Row; 1982. p. 197-248.

7. Lutjen-Drecoll E. Functional morphology of the trabecular meshwork in primate eyes. Prog Retin Eye Res 1999; 18:91-119.

8. Gonzalez P, Epstein DL, Borras T. Characterization of gene expression in human trabecular meshwork using single-pass sequencing of 1060 clones. Invest Ophthalmol Vis Sci 2000; 41:3678-93.

9. Tomarev SI, Wistow G, Raymond V, Dubois S, Malyukova I. Gene expression profile of the human trabecular meshwork: NEIBank sequence tag analysis. Invest Ophthalmol Vis Sci 2003; 44:2588-96.

10. Stamer WD, Seftor RE, Williams SK, Samaha HA, Snyder RW. Isolation and culture of human trabecular meshwork cells by extracellular matrix digestion. Curr Eye Res 1995; 14:611-7.

11. Wirtz MK, Samples JR, Xu H, Severson T, Acott TS. Expression profile and genome location of cDNA clones from an infant human trabecular meshwork cell library. Invest Ophthalmol Vis Sci 2002; 43:3698-704.

12. Liton PB, Liu X, Stamer WD, Challa P, Epstein DL, Gonzalez P. Specific targeting of gene expression to a subset of human trabecular meshwork cells using the chitinase 3-like 1 promoter. Invest Ophthalmol Vis Sci 2005; 46:183-90.

13. Dafforn A, Chen P, Deng G, Herrler M, Iglehart D, Koritala S, Lato S, Pillarisetty S, Purohit R, Wang M, Wang S, Kurn N. Linear mRNA amplification from as little as 5 ng total RNA for global gene expression analysis. Biotechniques 2004; 37:854-7.

14. Singh R, Maganti RJ, Jabba SV, Wang M, Deng G, Heath JD, Kurn N, Wangemann P. Microarray-based comparison of three amplification methods for nanogram amounts of total RNA. Am J Physiol Cell Physiol 2005; 288:C1179-89.

15. Gonzalez P, Caballero M, Liton PB, Stamer WD, Epstein DL. Expression analysis of the matrix GLA protein and VE-cadherin gene promoters in the outflow pathway. Invest Ophthalmol Vis Sci 2004; 45:1389-95.

16. Fautsch MP, Howell KG, Vrabel AM, Charlesworth MC, Muddiman DC, Johnson DH. Primary trabecular meshwork cells incubated in human aqueous humor differ from cells incubated in serum supplements. Invest Ophthalmol Vis Sci 2005; 46:2848-56.

17. Grierson I, Lee WR. Pressure-induced changes in the ultrastructure of the endothelium lining Schlemm's canal. Am J Ophthalmol 1975; 80:863-84.

18. Johnstone MA, Grant WG. Pressure-dependent changes in structures of the aqueous outflow system of human and monkey eyes. Am J Ophthalmol 1973; 75:365-83.

19. Peek R, Kammerer RA, Frank S, Otte-Holler I, Westphal JR. The angiopoietin-like factor cornea-derived transcript 6 is a putative morphogen for human cornea. J Biol Chem 2002; 277:686-93.

20. Zhao X, Ramsey KE, Stephan DA, Russell P. Gene and protein expression changes in human trabecular meshwork cells treated with transforming growth factor-beta. Invest Ophthalmol Vis Sci 2004; 45:4023-34.

21. Vittitow J, Borras T. Genes expressed in the human trabecular meshwork during pressure-induced homeostatic response. J Cell Physiol 2004; 201:126-37.

22. Hamann S, Zeuthen T, La Cour M, Nagelhus EA, Ottersen OP, Agre P, Nielsen S. Aquaporins in complex tissues: distribution of aquaporins 1-5 in human and rat eye. Am J Physiol 1998; 274:C1332-45.

23. Stamer WD, Snyder RW, Smith BL, Agre P, Regan JW. Localization of aquaporin CHIP in the human eye: implications in the pathogenesis of glaucoma and other disorders of ocular fluid balance. Invest Ophthalmol Vis Sci 1994; 35:3867-72.

24. Stamer WD, Peppel K, O'Donnell ME, Roberts BC, Wu F, Epstein DL. Expression of aquaporin-1 in human trabecular meshwork cells: role in resting cell volume. Invest Ophthalmol Vis Sci 2001; 42:1803-11.

25. Zhang D, Vetrivel L, Verkman AS. Aquaporin deletion in mice reduces intraocular pressure and aqueous fluid production. J Gen Physiol 2002; 119:561-9.

26. Karl MO, Fleischhauer JC, Stamer WD, Peterson-Yantorno K, Mitchell CH, Stone RA, Civan MM. Differential P1-purinergic modulation of human Schlemm's canal inner-wall cells. Am J Physiol Cell Physiol 2005; 288:C784-94.

27. Crosson CE. Adenosine receptor activation modulates intraocular pressure in rabbits. J Pharmacol Exp Ther 1995; 273:320-6.

28. Tian B, Gabelt BT, Crosson CE, Kaufman PL. Effects of adenosine agonists on intraocular pressure and aqueous humor dynamics in cynomolgus monkeys. Exp Eye Res 1997; 64:979-89.

29. Do CW, Civan MM. Basis of chloride transport in ciliary epithelium. J Membr Biol 2004; 200:1-13.

30. Civan MM, Macknight AD. The ins and outs of aqueous humour secretion. Exp Eye Res 2004; 78:625-31.

31. Schlotzer-Schrehardt U, Zenkel M, Decking U, Haubs D, Kruse FE, Junemann A, Coca-Prados M, Naumann GO. Selective upregulation of the A3 adenosine receptor in eyes with pseudoexfoliation syndrome and glaucoma. Invest Ophthalmol Vis Sci 2005; 46:2023-34.

32. Hochhauser E, Kaminski O, Shalom H, Leshem D, Shneyvays V, Shainberg A, Vidne BA. Role of adenosine receptor activation in antioxidant enzyme regulation during ischemia-reperfusion in the isolated rat heart. Antioxid Redox Signal 2004; 6:335-44.

33. Maggirwar SB, Dhanraj DN, Somani SM, Ramkumar V. Adenosine acts as an endogenous activator of the cellular antioxidant defense system. Biochem Biophys Res Commun 1994; 201:508-15.

34. Sands WA, Palmer TM. Adenosine receptors and the control of endothelial cell function in inflammatory disease. Immunol Lett 2005; 101:1-11.

35. Bhattacharya SK, Annangudi SP, Salomon RG, Kuchtey RW, Peachey NS, Crabb JW. Cochlin deposits in the trabecular meshwork of the glaucomatous DBA/2J mouse. Exp Eye Res 2005; 80:741-4.

36. Bhattacharya SK, Rockwood EJ, Smith SD, Bonilha VL, Crabb JS, Kuchtey RW, Robertson NG, Peachey NS, Morton CC, Crabb JW. Proteomics reveal Cochlin deposits associated with glaucomatous trabecular meshwork. J Biol Chem 2005; 280:6080-4.

37. Wang N, Chintala SK, Fini ME, Schuman JS. Activation of a tissue-specific stress response in the aqueous outflow pathway of the eye defines the glaucoma disease phenotype. Nat Med 2001; 7:304-9.

38. Sarkar D, Fisher PB. Molecular mechanisms of aging-associated inflammation. Cancer Lett 2006; 236:13-23.

39. Harman D. The free radical theory of aging. Antioxid Redox Signal 2003; 5:557-61.

40. Spector A, Ma W, Wang RR. The aqueous humor is capable of generating and degrading H2O2. Invest Ophthalmol Vis Sci 1998; 39:1188-97.

41. Garcia-Castineiras S, Velazquez S, Martinez P, Torres N. Aqueous humor hydrogen peroxide analysis with dichlorophenol-indophenol. Exp Eye Res 1992; 55:9-19.

42. Gherghel D, Griffiths HR, Hilton EJ, Cunliffe IA, Hosking SL. Systemic reduction in glutathione levels occurs in patients with primary open-angle glaucoma. Invest Ophthalmol Vis Sci 2005; 46:877-83.

43. Ferreira SM, Lerner SF, Brunzini R, Evelson PA, Llesuy SF. Oxidative stress markers in aqueous humor of glaucoma patients. Am J Ophthalmol 2004; 137:62-9.

44. Sacca SC, Pascotto A, Camicione P, Capris P, Izzotti A. Oxidative DNA damage in the human trabecular meshwork: clinical correlation in patients with primary open-angle glaucoma. Arch Ophthalmol 2005; 123:458-63.

45. Babizhayev MA, Bunin AYa. Lipid peroxidation in open-angle glaucoma. Acta Ophthalmol (Copenh) 1989; 67:371-7.

46. Liton PB, Challa P, Stinnett S, Luna C, Epstein DL, Gonzalez P. Cellular senescence in the glaucomatous outflow pathway. Exp Gerontol 2005; 40:745-8.

47. Horvath R, Freisinger P, Rubio R, Merl T, Bax R, Mayr JA, Shawan, Muller-Hocker J, Pongratz D, Moller LB, Horn N, Jaksch M. Congenital cataract, muscular hypotonia, developmental delay and sensorineural hearing loss associated with a defect in copper metabolism. J Inherit Metab Dis 2005; 28:479-92.

48. Hochstrasser H, Tomiuk J, Walter U, Behnke S, Spiegel J, Kruger R, Becker G, Riess O, Berg D. Functional relevance of ceruloplasmin mutations in Parkinson's disease. FASEB J 2005; 19:1851-3.

49. Panagiotakaki E, Tzetis M, Manolaki N, Loudianos G, Papatheodorou A, Manesis E, Nousia-Arvanitakis S, Syriopoulou V, Kanavakis E. Genotype-phenotype correlations for a wide spectrum of mutations in the Wilson disease gene (ATP7B). Am J Med Genet A 2004; 131:168-73.