Molecular Vision 2006; 12:832-840 <http://www.molvis.org/molvis/v12/a94/> Received 24 May 2006 | Accepted 25 July 2006 | Published 31 July 2006

Download Reprint

Optimized bacterial expression of myocilin proteins and functional comparison of bacterial and eukaryotic myocilins

Bum-Chan Park,¹ Xiang Shen,¹ Michael P. Fautsch,² Martin Tibudan,¹ Douglas H. Johnson,² Beatrice Y. J. T. Yue¹
 
 

¹Department of Ophthalmology and Visual Sciences, University of Illinois at Chicago College of Medicine, Chicago, IL; ²Department of Ophthalmology, Mayo Clinic College of Medicine, Rochester, MN

Correspondence to: Dr. Beatrice Yue, PhD, Department of Ophthalmology and Visual Sciences, University of Illinois at Chicago College of Medicine, 1855 West Taylor Street, Chicago, IL, 60612; Phone: (312) 996-6125; FAX: (312) 996-7773; email: beatyue@uic.edu

Abstract

Purpose: To maximize the expression level of myocilin and its truncated proteins in Escherichia coli (E. coli) and to examine the biological effects of bacterially expressed myocilin as compared to eukaryotic myocilin on cultured human trabecular meshwork (TM) cells.

Methods: Myocilin full length (1-504 amino acids) and two truncated proteins, myocilin 1-270 and 271-504, were expressed and purified from an E. coli strain, Rosetta2(DE3)pLysS. The eukaryotic myocilin was purified from cultured medium of a transformed TM cell line (TM5) transduced with feline immunodeficiency virus that contains an internal cassette expressing full length myocilin. The morphology and adhesion of human TM cells plated either on fibronectin alone or on fibronectin/purified myocilin mixtures were assessed by phase contrast microscopy. Actin cytoskeleton was examined using Oregon Green phalloidin. Immunofluorescence staining for paxillin was also performed.

Results: The expression of full length and truncated myocilin proteins in Rosetta2(DE3)pLysS was markedly increased especially when the bacteria were grown in media supplemented with 1.0% glucose. Cell adhesion was impaired and microspikes were formed when TM cells were plated onto fibronectin/bacterial full length myocilin mixtures. Loss of actin stress fibers and focal adhesions was also observed. This myocilin phenotype was also seen with myocilin 1-270, but not with myocilin 271-504. The eukaryotic full length myocilin produced nearly identical de-adhesive effects as those of the bacterially expressed myocilin.

Conclusions: The condition for a high level expression of full length and truncated myocilins in E. coli was optimized. The bacterial and eukaryotic recombinant full length myocilin produced similar biological consequence on TM cells. The myocilin phenotype appears to be largely due to the NH₂-terminal half of the protein.

Introduction

The trabecular meshwork (TM) at the chamber angle is the major regulation site of normal bulk flow of the aqueous humor [1]. The tissue is composed of TM cells that largely cover layers of trabecular beams made up of extracellular matrix (ECM) components, including fibronectin and collagens [2]. Cells in the TM system and their ECM are believed to contribute to the outflow resistance [2]. High resistance may lead to elevated intraocular pressure (IOP) and ultimately glaucoma, a heterogeneous disease characterized by progressive neural loss and visual impairment.

Myocilin is the product of the GLC1A gene linked directly to both juvenile- and adult-onset primary open angle glaucoma (POAG) [3,4]. Multiple mutations of the myocilin gene have been identified in a number of families [3-5]. Myocilin was initially identified as a 55-57 kDa protein secreted into the media of TM cultures after induction with glucocorticoids such as dexamethasone [6,7]. Analyses of the genomic sequence of myocilin have identified an NH₂-terminal myosin-like domain and a COOH-terminal olfactomedin-like domain [7,8]. The NH₂-terminus of myocilin has a cleavable signal peptide to target the protein for secretion [9] and, within the myosin-like domain, a leucine zipper motif necessary for interactions with itself or other proteins [10,11].

In both TM cells and tissues, myocilin has been localized to both intracellular and extracellular sites [12-14]. Extracellular myocilin interacts with fibronectin in TM tissue [15] and cell cultures [16]. As a component in the substrata, myocilin can block the adhesion of cultured TM cells onto fibronectin and induce loss of actin stress fibers and focal adhesions [17,18].

Myocilin is also present in the aqueous humor [19,20]. Perfusion of recombinant myocilin, especially when supplemented with the aqueous humor, increased outflow resistance in human anterior segments [21,22]. The COOH-terminal olfactomedin domain of myocilin had little effect [23], whereas the entire NH₂-terminus plus 98 amino acids (a.a.) of the olfactomedin domain in perfused human anterior segment cultures caused an increase in the outflow facility [24].

Although much progress has been made in characterization of myocilin during the past several years, its function is yet to be defined. Having an expression system for production of high levels of recombinant myocilin would greatly facilitate functional or mechanistic studies. In the field, eukaryotic recombinant full length or truncated myocilin has been obtained using yeast [25], insect [18], and human cell lines [21,23,24]. The yield, however, is in general fairly low and myocilin expression in eukaryotic systems remains challenging and time-consuming. A bacterial expression system would provide an easier alternative for rapid protein production. To date, investigators, including our group [17,22], have reported the production of bacterially expressed myocilin using standard protocols. The expression level nevertheless is still less than optimal [17,22], rendering it difficult to chart for structure and function type of investigations.

In the present study, we introduced new strategies to maximize the yield of bacterially expressed recombinant myocilin and its truncated proteins. We assessed the extracellular effects of bacterial myocilins on cultured TM cells and compared their biological activities with eukaryotically expressed myocilin.

Methods

Human TM cell cultures

Normal human eyes from donors 29, 33, 41, 44, and 55 years of age were obtained from the Illinois Eye Bank (Chicago, IL). TM tissues excised from these eyes were cultured on Falcon Primaria flasks with complete medium containing Eagle's minimum essential medium (MEM, Sigma, St. Louis, MO), 10% fetal bovine serum (FBS), 5% calf serum, essential and nonessential amino acids, and antibiotics. When the cells reached confluence, they were trypsinized and subcultured.

Production of recombinant protein in Escherichia coli

Full length human myocilin (1-504 amino acids) was subcloned into pRSET as described previously [17]. The construct pRSET/myocilin was made to produce a fusion protein with an Xpress tag and an NH₂-terminal six amino acid histidine (His) tag to facilitate its purification. For the two truncated myocilins, myocilin 1-270 and 271-504, DNA fragments were amplified by PCR using pRSET/myocilin as template and primers as follows: 5'-GGC GGA TCC ATG AGG TTC TTC TGT GCA CG-3' and 5'-GGC GAA TTC CCT ACC ACA CAC CAT ACT TGC CAG-3' for myocilin 1-270; 5'-GGC GGA TCC ATG CGA GAC CCC AAG CCC-3' and 5'-GGC GAA TTC CTC ACA TCT TGG AGA GCT TGA TGT C-3' for myocilin 271-504. The resulting PCR products were digested with BamHI and EcoRI and cloned in frame into pRSET (Invitrogen, Carlsbad, CA) at the same restriction sites to yield pRSET/myocilin 1-270 and pRSET/myocilin 271-504. An E. coli strain, Rosetta2(DE3)pLysS (Novagen, Madison, WI), was transformed with pRSET/myocilin, pRSET/myocilin 1-270, or 271-504 and grown at 30 °C in 1 l of Luria-Bertani (LB) medium with 50 μg/ml carbenicillin and 34 μg/ml chloramphenicol in the absence or presence of 1.0% glucose. When the OD₆₀₀ reached 0.4-0.6, cells were collected by centrifugation at 5,000x g for 15 min. They were then grown at 25 °C for an additional 4 h in 1 l LB with 200 μg/ml ampicillin and 0.2 mM isopropyl β-D-1-thiogalactopyranoside (IPTG). The bacteria collected by centrifugation were frozen at -80 °C until use.

Purification of inclusion body and refolding of solubilized bacterial recombinant protein

The bacterial pellet was resuspended in protein storage buffer (50 mM NaH₂PO₄, pH 8.0, 300 mM NaCl) containing 1 mg/ml lysozyme (Roche, Indianapolis, IN) and incubated on ice for 30 min. Cells were disrupted by sonication on ice with six 20 s bursts and 20 s pauses. Cell debris and inclusion bodies (IB) were pelleted at 15,000x g for 15 min. To purify only IB, the pellet was washed 3-5 times in a buffer containing 50 mM Tris, 10 mM EDTA, pH 7.5, 5 mM dithiothreitol (DTT), 2% Triton X-100, and 500 mM NaCl. The pellet was resuspended in IB solubilization buffer (6 M guanidine-HCl, 50 mM Tris, pH 8.0, and 20 mM DTT) and nonsolubilized material was removed by ultracentrifugation at 100,000x g for 30 min. The supernatant was desalted through a PD-10 column (Amersham, Piscataway, NJ) to remove DTT. Proteins were further purified with nickel-nitriloacetic acid (Ni-NTA) agarose beads (Qiagen, Valencia, CA) per the manufacturer's instructions.

Refolding of purified proteins was carried out using the Pro-Matrix^TM Protein Refolding kit (Pierce, Rockford, IL) according to the manufacturer's instructions. In brief, optimal concentrations of guanidine, L-arginine, polyethyleneglycol (PEG), divalent cations, and ratios of reduced and oxidized glutathione (GSH:GSSG), were screened by various combinations. All refolding experiments were conducted for 18-24 h at 4 °C. Refolded protein samples were dialyzed overnight at 4 °C in protein storage buffer.

Analysis of expressed proteins

The purified bacterial myocilin samples were resolved on 12% SDS-PAGE gels. The gels were either stained with Coomassie brilliant blue (Bio-Rad, Hercules, CA) or electroblotted for 1 h onto Protran nitrocellulose membrane (Midwest scientific, St. Louis, MO). For western blotting, the amount of proteins loaded was 1/100 of that for Coomassie staining. The blot was blocked for 1 h with 5% nonfat dry milk in TBST buffer (20 mM Tris-HCl, pH 7.4, 150 mM NaCl, 0.1% Tween 20) and incubated for 1 h with rabbit polyclonal anti-myocilin (1:500; Santa Cruz Biotechnology, Santa Cruz, CA) specific to either NH₂- or COOH-terminus of myocilin, or mouse monoclonal anti-polyHistidine (1:3,000; Sigma). It was then incubated for 1 h with horseradish peroxidase-conjugated secondary antibody (1:10,000; Jackson ImmunoResearch Laboratories, West Grove, PA). Protein bands were detected using SuperSignal Substrate (Pierce).

Expression and purification of eukaryotic recombinant myocilin

Eukaryotic full length myocilin was expressed and purified as described previously [21] from conditioned media of a transformed TM cell line, TM5, transduced with feline immunodeficiency virus that contains wild type/full length myocilin fused to COOH-terminal His and V5 tags.

Cell morphology

Lab-Tek 8 well chamber slides (Nalge Nunc, Rochester, NY) were coated overnight at 4 °C with fibronectin (5 μg/ml, Sigma) alone, or with fibronectin (5 μg/ml) mixed with bovine serum albumin (BSA, Sigma, 5 μg/ml), bacterial recombinant full length myocilin (5 μg/ml), myocilin 1-270 (2.5 μg/ml), myocilin 271-504 (2.5 μg/ml), or eukaryotic recombinant full length myocilin (5 μg/ml) in MEM. After coating, the slides were washed twice with MEM and once with MEM containing trypsin inhibitor (1 mg/ml, Sigma) and BSA (10 mg/ml). Human TM cells were plated (20,000 cells/well) onto the coated wells and incubated at 37 °C for 1, 4, 6, and 24 h. Cell morphology was examined under a phase contrast microscope (Zeiss Axioscope; Carl Zeiss MicroImaging, Thornwood, NY). The protein coating, as evidenced by immunostaining, was uniform throughout the wells (data not shown).

Immunofluorescence and actin staining

Human TM cells were plated onto 8 well glass chamber slides coated as described above. Cells were fixed in 4% paraformaldehyde followed by permeabilization in 0.2% Triton X-100. After blocking, cells were incubated for 1 h with mouse monoclonal anti-paxillin (1:100; Upstate, Lake Placid, NY), washed, and incubated for 1 h with Cy3-conjugated goat anti-mouse IgG (1:200; Jackson ImmunoResearch Laboratories). The actin structure was visualized using Oregon Green phalloidin 488 (1:40; Invitrogen/Molecular Probes, Carlsbad, CA) for 20 min. Slides were mounted in Vectashield with DAPI (Vector Laboratories, Burlingame, CA) and examined under Zeiss Axioscope with the aid of Metamorph software (Molecular Devices, Downingtown, PA).

Results & Discussion

Expression of recombinant full length and truncated myocilin in E. coli

Full length human myocilin has previously been expressed in BL21(DE3)pLysS, a bacterial strain widely used for protein expression [17,22]. The overall yield however was low (approximately 20-100 μg/l). Furthermore, using BL21(DE3)pLysS, expression of a truncated myocilin, myocilin 1-270, was proved to be unfeasible (data not shown). Efforts were thus made to optimize conditions for bacterial expression of both full length and truncated myocilins.

In many cases, an excess of rare codons, particularly the extremely rare arginine AGG and AGA codons [26], could create translational problems in E. coli [27]. Sequence analyses (Table 1) revealed that the myocilin gene contains 47 rare codons (9.3% out of a total of 504 codons). The rare codon frequency is higher in the cDNA encoding myocilin 1-270 (72.3%, 34 out of a total of 47 rare codons) than that encoding myocilin 271-504 (27.7%, 13 out of 47 rare codons). Moreover, the myocilin mRNA is highly enriched (58.3%, 21 out of 36 total arginine codons) in AGG and AGA, which are the most rare in E. coli. Out of them, myocilin 1-270 cDNA contains 17 AGG and AGA codons, representing 81% of the 21 such codons (Table 1).

Expression of heterologous mRNA species such as those of myocilins that contain a large number of the rare codons has been shown to result in translation errors [28-30]. Coexpression of minor tRNAs could be utilized in vivo to overcome such difficulties [31,32]. In light of these assessments, we replaced the BL21(DE3)pLysS strain with another E. coli strain, Rosetta2(DE3)pLysS. This Rosetta2 strain carries a plasmid containing tRNA genes that decode seven rare E. coli codons (AGG, AGA, CGG, AUA, CUA, GGA, and CCC) designed to enhance the expression of eukaryotic proteins.

In addition, we grew the bacteria in a medium supplemented with glucose. The low expression levels can be caused by toxicity of the target protein. Myocilin might be toxic and the toxicity could be manifested due to incomplete repression of protein expression before IPTG induction. The pRSET vector system used for myocilin expression is under the control of the lac promoter, which is leaky. To repress the induction of the lac promoter by lactose, a standard ingredient of culture media such as LB, and to completely block the basal level expression of myocilin before IPTG induction, we added glucose to the media. As shown in Figure 1, the expression level of full length myocilin and myocilin1-270 was markedly increased when the bacteria were grown in the presence of 1.0% glucose (Figure 1A,B; lanes 1-5 versus lanes 7-11). The increase was noticeable after a 30 min incubation and reached the maximal level at 3 h for both myocilin constructs. The effect of glucose concentration on myocilin expression was tested from a range of 0.5-2.0% and the optimal concentration was concluded to be 1.0-1.2% (data not shown). The expression of myocilin 271-504 was robust irrespective of whether extra glucose was present in the culture medium (Figure 1C, lanes 1-5 versus lanes 7-11).

The yield of purified full length myocilin, myocilin 1-270, and myocilin 271-504 using our optimized Rosetta2(DE3)pLysS expression system was approximately 3.2, 0.2, and 6.8 mg/l, respectively. Compared to the previous BL21(DE3)pLysS method, the yield for the full length myocilin was improved by at least 30 fold (3.2 mg/l versus 20-100 μg/l). Furthermore, expression of bacterial myocilin 1-270 was never successful until now. The Rosetta2 bacterial system and the glucose supplementation therefore prove to be successful, augmenting the expression levels of both full length and truncated myocilin proteins in bacteria.

Purification of recombinant full length and truncated myocilins

Our initial attempts to purify full length and truncated myocilins under native conditions resulted in little target proteins amid vast amounts of impurities. We noted that the myocilin proteins could be purified only under denaturing conditions with strong denaturants such as 6 M guanidine-HCl and 8 M urea, indicating that they were sequestered into insoluble inclusion bodies. Such inclusion bodies were isolated, solubilized in a 6 M guanidine-HCl-containing buffer, and purified by Ni²⁺-affinity chromatography. Coomassie blue-stained gels presented in Figure 2A showed multiple protein bands prior to, and one major band after, Ni purification. Furthermore, western blotting with anti-polyHistidine or antibody specific to NH₂-or COOH-terminal myocilin detected a single major band following affinity purification at approximately 61, 34, or 32 kDa, corresponding to the mass of His and Xpress tags (4 kDa) from pRSET vector plus myocilin (57 kDa), myocilin 1-270 (30 kDa), or myocilin 271-504 (28 kDa), respectively (Figure 2B). The optimized refolding condition for denatured full length and truncated myocilins was determined after intensive screening to be 50 mg of proteins in 1 ml buffer containing 50 mM Tris-HCl, 300 mM NaCl, 0.8 mM KCl, pH 8.2, 0.4 M guanidine, 0.4 M arginine, and 2 mM:0.4 mM GSH:GSSG at 4 °C.

Biological effects of recombinant truncated myocilins expressed in a bacterial system

Human TM cells were plated onto fibronectin/bacterial full length myocilin, fibronectin/myocilin 1-270, or fibronectin/myocilin 271-504 mixtures. As was reported previously [17], the bacterial full length myocilin caused an impairment of cell adhesion. Within 4 h of plating, the percentage of cells attached in fibronectin/myocilin mixture-coated wells (approximately 30-40%) was lower than that in fibronectin controls (about 90%).

The morphology of TM cells plated onto fibronectin mixed with BSA (fibronectin/BSA) was similar to that seen with fibronectin alone. By contrast, mixing bacterial full length myocilin with fibronectin induced the cells to assume a stellate morphology with broad cell bodies and microspikes (Figure 3). This phenotype was readily observed within 1 and 4 h of plating in a majority of the cells. It was still detected in a few cells at 6 h, but not 24 h after plating. At these later times, most or all of the cells remaining in the fibronectin/full length myocilin wells attached to the substrata, displaying a flat, normal spread morphology. It is of note that the total density of the cells remained on these wells was lower than that in controls, possibly reflecting the fact that the cell adhesion during the initial plating phase was impeded by the presence of myocilin in the substrata.

Normal human TM cells plated and spread on fibronectin alone or on fibronectin/BSA exhibited robust actin stress fibers and prominent paxillin-positive focal contacts or focal adhesions (Figure 4). By contrast, cells plated on fibronectin/bacterial myocilin showed a substantial loss of actin stress fibers and a decrease in focal contacts at both 1 and 4 h (Figure 4). These data confirmed previous findings [17] that bacterial full length myocilin, when mixed with fibronectin, blocked the attachment of TM cells and induced the formation of microspikes, probably as a consequence of the loss of actin stress fibers and focal adhesions.

Similarly, adherence of TM cells on fibronectin/myocilin 1-270 mixtures appeared to be compromised. Microspikes were formed (Figure 3), the long and parallel actin stress fibers were reduced, and staining for paxillin was diminished (Figure 4). On the other hand, myocilin 271-504 did not seem to elicit any effect on the TM cell morphology, actin architecture, or focal contact formation (Figure 3, Figure 4). The full length myocilin phenotype thus was also seen with myocilin 1-270, but not with myocilin 271-504.

The NH₂-terminal half of myocilin contains a signal peptide for secretion [9] and, within the myosin-like motif, a leucine zipper motif for interaction with itself to form multimers [10]. Our data is consistent with a recent report by Gobeil et al. [11] that deletion of the leucine zipper motif resulted in a loss of adhesive ability of myocilin to the ECM. Myocilin has been shown by in vitro assays to be able to interact with ECM molecules such as fibronectin and type I collagen [14,16]. The heparin II (Hep II) domain of fibronectin in particular has been found to be the interacting site for myocilin [16,18]. Therefore, the NH₂-terminal half of myocilin is probably the region for interaction with the Hep II domain that is known for its biological role in cell adhesion, organization of the cytoskeleton, signal transduction, and phagocytosis [33,34]. In accordance, the COOH-terminal half of myocilin, unlike the NH₂-terminal half, confers little, if any, phenotypes on TM cells. Of note however is that the relevance of these notions has to be addressed in in vivo systems. In the literature, myocilin level has been found to be increased in the TM of a population of POAG patients [35]. There is no indication yet whether myocilin and fibronectin form a complex and accumulate in the glaucomatous TM.

Effects of recombinant myocilin expressed in eukaryotic system on TM cells

Eukaryotic recombinant myocilin was expressed and purified as described previously [21]. Human TM cells were plated onto fibronectin/eukaryotic myocilin mixtures. The effects produced by the eukaryotic myocilin were nearly identical to those described above by the bacterially expressed full length myocilin. Microspikes (Figure 5), and loss of actin stress fibers and focal adhesions (Figure 6) were observed within 1 and 4 h of plating. This suggests that posttranslational modification such as glycosylation typically lacking in proteins produced in bacterial systems may not be critical in determination of the intracellular myocilin effects. The glycosylation status of myocilin is somewhat controversial. Several groups [9,36,37] have shown that myocilin is N-glycosylated and an upper 57 kDa band of the 55-57 kDa doublet accounts for the glycosylated form of myocilin. Evidence however is also provided that myocilin is not glycosylated [38,39]. Nguyen et al. [7] claimed that the appearance of the 55-57 kDa doublet is resulted from a proteolytic cleavage of a 32 amino acid signal peptide. In any case, it appears that glycosylation per se does not alter biological functions of the intracellular myocilin.

Myocilin is the product of the GLC1A gene linked to POAG [3]. Heterozygous mutations in this gene have been found to account for approximately 3 to 4% of the POAG cases. So far, more than 70 mutations have been identified [4,40]. Upregulation of myocilin has also been implicated in corticosteroid glaucoma [41]. The exact mechanisms how mutated or overexpressed myocilin would lead to pathology however still remain to be fully depicted. Availability of sufficient quantities of functionally active recombinant myocilin would undoubtedly facilitate further investigations on functional significance of myocilin. In the current study, we developed a new method for a high level expression of full length and truncated myocilins in E. coli. With this tool, we demonstrated that the bacterially and eukaryotically expressed myocilin behave similarly and that the NH₂-terminal half of myocilin is the major determinant of the de-adhesive consequence produced by the protein. In vivo, TM cells are continually subjected to stress from the aqueous flow and IOP fluctuations. Cell adhesion to the ECM substratum is an important aspect of the TM biology. It is conceivable that the altered actin cytoskeleton and compromised adhesion induced by overabundant extracellular myocilin may disturb the adhesion homeostasis, rendering TM cells in situ in a weakened or vulnerable state. The vulnerability, along with additional stress or factors, may be the underlying basis for development of pathologic conditions.

Acknowledgements

This work was supported by grants EY05628 (BYJTY), EY03890 (BYJTY), EY015736 (MPF), EY007065 (DHJ), and core grant EY01792 from the National Institutes of Health, Bethesda, MD.

References

1. Bill A. Editorial: The drainage of aqueous humor. Invest Ophthalmol 1975; 14:1-3.

2. Yue BY. The extracellular matrix and its modulation in the trabecular meshwork. Surv Ophthalmol 1996; 40:379-90.

3. Stone EM, Fingert JH, Alward WL, Nguyen TD, Polansky JR, Sunden SL, Nishimura D, Clark AF, Nystuen A, Nichols BE, Mackey DA, Ritch R, Kalenak JW, Craven ER, Sheffield VC. Identification of a gene that causes primary open angle glaucoma. Science 1997; 275:668-70.

4. Alward WL, Fingert JH, Coote MA, Johnson AT, Lerner SF, Junqua D, Durcan FJ, McCartney PJ, Mackey DA, Sheffield VC, Stone EM. Clinical features associated with mutations in the chromosome 1 open-angle glaucoma gene (GLC1A). N Engl J Med 1998; 338:1022-7.

5. Shimizu S, Lichter PR, Johnson AT, Zhou Z, Higashi M, Gottfredsdottir M, Othman M, Moroi SE, Rozsa FW, Schertzer RM, Clarke MS, Schwartz AL, Downs CA, Vollrath D, Richards JE. Age-dependent prevalence of mutations at the GLC1A locus in primary open-angle glaucoma. Am J Ophthalmol 2000; 130:165-77.

6. Polansky JR, Fauss DJ, Chen P, Chen H, Lutjen-Drecoll E, Johnson D, Kurtz RM, Ma ZD, Bloom E, Nguyen TD. Cellular pharmacology and molecular biology of the trabecular meshwork inducible glucocorticoid response gene product. Ophthalmologica 1997; 211:126-39.

7. Nguyen TD, Chen P, Huang WD, Chen H, Johnson D, Polansky JR. Gene structure and properties of TIGR, an olfactomedin-related glycoprotein cloned from glucocorticoid-induced trabecular meshwork cells. J Biol Chem 1998; 273:6341-50.

8. Kubota R, Noda S, Wang Y, Minoshima S, Asakawa S, Kudoh J, Mashima Y, Oguchi Y, Shimizu N. A novel myosin-like protein (myocilin) expressed in the connecting cilium of the photoreceptor: molecular cloning, tissue expression, and chromosomal mapping. Genomics 1997; 41:360-9.

9. Shepard AR, Jacobson N, Sui R, Steely HT, Lotery AJ, Stone EM, Clark AF. Characterization of rabbit myocilin: Implications for human myocilin glycosylation and signal peptide usage. BMC Genet 2003; 4:5.

10. Fautsch MP, Johnson DH. Characterization of myocilin-myocilin interactions. Invest Ophthalmol Vis Sci 2001; 42:2324-31.

11. Gobeil S, Letartre L, Raymond V. Functional analysis of the glaucoma-causing TIGR/myocilin protein: integrity of amino-terminal coiled-coil regions and olfactomedin homology domain is essential for extracellular adhesion and secretion. Exp Eye Res 2006; 82:1017-29.

12. Wentz-Hunter K, Shen X, Yue BY. Distribution of myocilin, a glaucoma gene product, in human corneal fibroblasts. Mol Vis 2003; 9:308-14 <http://www.molvis.org/molvis/v9/a43/>.

13. Wentz-Hunter K, Ueda J, Shimizu N, Yue BY. Myocilin is associated with mitochondria in human trabecular meshwork cells. J Cell Physiol 2002; 190:46-53.

14. Ueda J, Wentz-Hunter K, Yue BY. Distribution of myocilin and extracellular matrix components in the juxtacanalicular tissue of human eyes. Invest Ophthalmol Vis Sci 2002; 43:1068-76.

15. Ueda J, Yue BY. Distribution of myocilin and extracellular matrix components in the corneoscleral meshwork of human eyes. Invest Ophthalmol Vis Sci 2003; 44:4772-9.

16. Filla MS, Liu X, Nguyen TD, Polansky JR, Brandt CR, Kaufman PL, Peters DM. In vitro localization of TIGR/MYOC in trabecular meshwork extracellular matrix and binding to fibronectin. Invest Ophthalmol Vis Sci 2002; 43:151-61.

17. Wentz-Hunter K, Kubota R, Shen X, Yue BY. Extracellular myocilin affects activity of human trabecular meshwork cells. J Cell Physiol 2004; 200:45-52.

18. Peters DM, Herbert K, Biddick B, Peterson JA. Myocilin binding to Hep II domain of fibronectin inhibits cell spreading and incorporation of paxillin into focal adhesions. Exp Cell Res 2005; 303:218-28.

19. Jacobson N, Andrews M, Shepard AR, Nishimura D, Searby C, Fingert JH, Hageman G, Mullins R, Davidson BL, Kwon YH, Alward WL, Stone EM, Clark AF, Sheffield VC. Non-secretion of mutant proteins of the glaucoma gene myocilin in cultured trabecular meshwork cells and in aqueous humor. Hum Mol Genet 2001; 10:117-25.

20. Russell P, Tamm ER, Grehn FJ, Picht G, Johnson M. The presence and properties of myocilin in the aqueous humor. Invest Ophthalmol Vis Sci 2001; 42:983-6.

21. Fautsch MP, Bahler CK, Vrabel AM, Howell KG, Loewen N, Teo WL, Poeschla EM, Johnson DH. Perfusion of his-tagged eukaryotic myocilin increases outflow resistance in human anterior segments in the presence of aqueous humor. Invest Ophthalmol Vis Sci 2006; 47:213-21.

22. Fautsch MP, Bahler CK, Jewison DJ, Johnson DH. Recombinant TIGR/MYOC increases outflow resistance in the human anterior segment. Invest Ophthalmol Vis Sci 2000; 41:4163-8.

23. Goldwich A, Ethier CR, Chan DW, Tamm ER. Perfusion with the olfactomedin domain of myocilin does not affect outflow facility. Invest Ophthalmol Vis Sci 2003; 44:1953-61.

24. Caballero M, Rowlette LL, Borras T. Altered secretion of a TIGR/MYOC mutant lacking the olfactomedin domain. Biochim Biophys Acta 2000; 1502:447-60.

25. Nagy I, Trexler M, Patthy L. Expression and characterization of the olfactomedin domain of human myocilin. Biochem Biophys Res Commun 2003; 302:554-61.

26. Zhang SP, Zubay G, Goldman E. Low-usage codons in Escherichia coli, yeast, fruit fly and primates. Gene 1991; 105:61-72.

27. Kane JF. Effects of rare codon clusters on high-level expression of heterologous proteins in Escherichia coli. Curr Opin Biotechnol 1995; 6:494-500.

28. Spanjaard RA, Chen K, Walker JR, van Duin J. Frameshift suppression at tandem AGA and AGG codons by cloned tRNA genes: assigning a codon to argU tRNA and T4 tRNA(Arg). Nucleic Acids Res 1990; 18:5031-6.

29. Brinkmann U, Mattes RE, Buckel P. High-level expression of recombinant genes in Escherichia coli is dependent on the availability of the dnaY gene product. Gene 1989; 85:109-14.

30. Calderone TL, Stevens RD, Oas TG. High-level misincorporation of lysine for arginine at AGA codons in a fusion protein expressed in Escherichia coli. J Mol Biol 1996; 262:407-12.

31. Jiang X, Nakano H, Kigawa T, Yabuki T, Yokoyama S, Clark DS, Yamane T. Dosage effect of minor arginyl- and isoleucyl-tRNAs on protein synthesis in an Escherichia coli in vitro coupled transcription/translation system. J Biosci Bioeng 2001; 91:53-7.

32. Sorensen HP, Sperling-Petersen HU, Mortensen KK. Production of recombinant thermostable proteins expressed in Escherichia coli: completion of protein synthesis is the bottleneck. J Chromatogr B Analyt Technol Biomed Life Sci 2003; 786:207-14.

33. Bultmann H, Santas AJ, Peters DM. Fibronectin fibrillogenesis involves the heparin II binding domain of fibronectin. J Biol Chem 1998; 273:2601-9.

34. Santas AJ, Peterson JA, Halbleib JL, Craig SE, Humphries MJ, Peters DM. Alternative splicing of the IIICS domain in fibronectin governs the role of the heparin II domain in fibrillogenesis and cell spreading. J Biol Chem 2002; 277:13650-8.

35. Lutjen-Drecoll E, May CA, Polansky JR, Johnson DH, Bloemendal H, Nguyen TD. Localization of the stress proteins alpha B-crystallin and trabecular meshwork inducible glucocorticoid response protein in normal and glaucomatous trabecular meshwork. Invest Ophthalmol Vis Sci 1998; 39:517-25.

36. Caballero M, Borras T. Inefficient processing of an olfactomedin-deficient myocilin mutant: potential physiological relevance to glaucoma. Biochem Biophys Res Commun 2001; 282:662-70.

37. Gobeil S, Rodrigue MA, Moisan S, Nguyen TD, Polansky JR, Morissette J, Raymond V. Intracellular sequestration of hetero-oligomers formed by wild-type and glaucoma-causing myocilin mutants. Invest Ophthalmol Vis Sci 2004; 45:3560-7.

38. Hardy KM, Hoffman EA, Gonzalez P, McKay BS, Stamer WD. Extracellular trafficking of myocilin in human trabecular meshwork cells. J Biol Chem 2005; 280:28917-26.

39. Huang W, Jaroszewski J, Ortego J, Escribano J, Coca-Prados M. Expression of the TIGR gene in the iris, ciliary body, and trabecular meshwork of the human eye. Ophthalmic Genet 2000; 21:155-69.

40. Gong G, Kosoko-Lasaki O, Haynatzki GR, Wilson MR. Genetic dissection of myocilin glaucoma. Hum Mol Genet 2004; 13 Spec No 1:R91-102. Erratum in: Hum Mol Genet 2004; 13:991.

41. Clark AF, Steely HT, Dickerson JE Jr, English-Wright S, Stropki K, McCartney MD, Jacobson N, Shepard AR, Clark JI, Matsushima H, Peskind ER, Leverenz JB, Wilkinson CW, Swiderski RE, Fingert JH, Sheffield VC, Stone EM. Glucocorticoid induction of the glaucoma gene MYOC in human and monkey trabecular meshwork cells and tissues. Invest Ophthalmol Vis Sci 2001; 42:1769-80.