Mutated Mouse and Human Myocilins Have Similar Properties and Do Not Block General Secretory Pathway

Abstract

purpose. The present study compared properties of wild-type and mutated mouse and human myocilin (Myoc) proteins as a prerequisite for development of a mouse model of glaucoma.

methods. cDNA encoding full-length mouse Myoc was cloned into the p3XFLAG-CMV-14 vector. Tyr423His and Ile463Ser mutations were introduced into the mouse Myoc protein by in vitro mutagenesis. Intracellular localization and secretion of wild-type and mutated mouse Myoc proteins were studied in immunostaining and Western blotting experiments, respectively, after transfection into COS-7 cells.

results. Similar to human MYOC, wild-type and mutated mouse Myoc demonstrated vesicular staining in transfected cells. However, while wild-type human and mouse Myoc were preferentially located in both the endoplasmic reticulum and Golgi, mutated human and mouse Myoc were located mainly in the endoplasmic reticulum and were excluded from Golgi. Similar to mutations in human MYOC, mutations in mouse Myoc dramatically reduced its secretion from transfected cells. Secretion of mutated Myoc was partially restored by culturing cells at 30°C instead of 37°C. The presence of mutated human MYOC prevented secretion of wild-type mouse Myoc but did not dramatically affect secretion of alkaline phosphatase, thrombospondin, Timp3 or olfactomedin-1.

conclusions. Properties of the mouse Myoc protein are similar to those of the human MYOC. The presence of mutated mouse or human Myoc does not block a general secretory pathway. Expression of mutated Myoc in the eye in mice may mimic human glaucoma and lead to development of a genetic mouse model of glaucoma.

It is now well established that mutations in the myocilin (MYOC) gene are associated with juvenile open-angle glaucoma, often showing high IOP. ^{1 2 3 4} Moreover, between 2.6% and 4.3% of sporadic primary open-angle glaucoma cases are associated with mutations in this gene. ⁵ The MYOC gene encodes a secreted protein containing the olfactomedin domain located in the C-terminal part of the protein. Olfactomedin was first identified as a novel glycoprotein specifically present in the frog olfactory epithelium. ⁶ Subsequently, the olfactomedin domain was identified in several proteins each showing characteristic distribution in different tissues. ^{7 8 9 10 11 12 13 14 15} In particular, the MYOC gene is highly expressed in several eye structures. The highest levels of the MYOC expression are observed in the trabecular meshwork and sclera. ^{2 11 16} It is also expressed in the ciliary body, ^{2 17 18} iris, ¹⁹ retinal pigment epithelium/choroid, ^{11 20} and optic nerve. ²¹ The MYOC gene expression pattern is consistent with its proposed role in aqueous dynamics. However, the molecular mechanisms of normal MYOC action are still not well understood. ²²  

The absence of open-angle glaucoma in an elderly woman homozygous for the Arg46Stop mutation ²³ as well as the absence of glaucoma in people hemizygous for MYOC ²⁴ suggests that loss of functional MYOC is not critical for the development of glaucoma. These observations are supported by data from mice with targeted disruption of the Myoc gene. Mice heterozygous and homozygous for a targeted null mutation in Myoc do not have a detectable eye phenotype. ²⁵ Fifteen-fold increases in the levels of normal Myoc in the eyes of transgenic mice also do not lead to elevation of IOP or glaucoma. ²⁶ These findings indicate that increased levels of MYOC may not lead to steroid-induced glaucoma and that noncoding sequence variants of MYOC are by themselves unlikely to lead to glaucoma, as has been previously suggested. ²⁷ A glaucoma phenotype appears to be dependent on expression of mutated MYOC proteins in the eye tissues. 

More than 70 glaucoma-associated mutations have been identified in the MYOC gene. ^{1 28} Most of these mutations are located in the third exon of the MYOC gene encoding the olfactomedin domain. Mutated MYOC is not secreted and prevents secretion of wild-type (WT) MYOC when both mutated and normal proteins are present in the same cells. ^{29 30 31} The molecular mechanisms by which mutated MYOC causes the elevation of IOP and glaucoma phenotype are not known. Recent data demonstrate that expression of mutated MYOC may induce unfolded protein response ³² and results in abnormal cell morphology and cell death. ^{32 33} It has been suggested that mutated MYOC may interfere not only with secretion of normal MYOC but also affect secretion of other biologically active proteins. ^{11 31} So far, the effects of mutated human MYOC on secretion have been studied for only two proteins, matrix metalloproteinase 3, also known as stromelysin, ³² and optimedin B ¹¹ . Optimedin is an olfactomedin domain–containing protein that may interact with MYOC. ¹¹ It has been reported that matrix metalloproteinase 3 was efficiently secreted from transfected trabecular meshwork cells in the presence of several MYOC mutants tested, while secretion of optimedin B was reduced in the presence of mutated MYOC. ^{11 32}  

Animal models may provide a very useful tool to study the molecular mechanisms involved in IOP elevation after expression of mutated Myoc in eye tissues. However, it is not known whether mutations in the olfactomedin domain of Myoc affect the secretion of these proteins in species other than humans. The purpose of the present study was to examine the effects of mutations in the olfactomedin domain of mouse Myoc on its secretion and intracellular localization. We demonstrated that mutated mouse Myoc has properties similar to those of mutated human MYOC. Although the presence of mutated mouse and human Myoc significantly reduced secretion of WT mouse Myoc, it did not dramatically reduced secretion of other tested proteins. 

Methods

Expression Plasmid Construction and Mutagenesis

Mouse and human cDNA encoding full-length Myoc protein were cloned into the p3XFLAG-CMV-14 (Sigma, St. Louis, MO), pcDNA3.1/Myc-His(+)B (Invitrogen, Carlsbad, CA), or pCS2-Flag vector. Full-length mouse tissue inhibitor of metalloproteinase 3 (Timp3) was cloned into the p3XFLAG-CMV-14 vector. Mouse optimedin-A, rat optimedin-B, and rat olfactomedin-1 cDNA were cloned into the pcDNA3.1/Myc-His(+)B vector ¹¹ . Human thrombospondin-1-myc–containing plasmid was a kind gift from David D. Roberts (Laboratory of Pathology, Center for Cancer Research, National Cancer Institute, National Institutes of Health, Bethesda, MD). A plasmid encoding secreted alkaline phosphatase (SEAP) cloned in pcDNA3 ³⁴ was obtained from Sven-Ulrik Gorr (University of Louisville School of Dentistry, Louisville, KY). Point mutations were introduced into the indicated cDNA using a site-directed mutagenesis kit (Quickchange; Stratagene, La Jolla, CA). Identity of the constructs was confirmed by sequencing. 

Intracellular Localization of Mouse Myoc

COS-7 cells were grown on chamber slides (Permanox Laboratory-Tek; Nulge Nunc International, Naperville, IL) and transfected as described elsewhere. ³⁵ Cells were fixed in 4% paraformaldehyde prepared in phosphate buffered saline (PBS, pH 7.4) for 10 minutes 48 hours after transfection, washed several times in PBS, and permeabilized with 0.1% Triton X-100 prepared in PBS for 5 minutes. For immunofluorescence, cells were first incubated in PBS with 2% bovine serum albumin (BSA) for 30 minutes, then incubated with the monoclonal mouse anti-Flag antibody (dilution 1:2000, Sigma) and with the polyclonal antibody against calnexin (dilution 1:100, Stressgen, Victoria, BC, Canada) or with anti-Myoc polyclonal antibody (dilution 1:100) ²⁵ and anti-golgin-97 monoclonal antibody (dilution 1:100; Molecular Probes, Eugene, OR) in PBS with 2% BSA. After repeated washing in PBS, the signals were visualized using rhodamine anti-mouse antibody (TRITC; dilution 1:100; Jackson ImmunoResearch, West Grove, PA) and Alexa 488 anti-rabbit antibody (dilution 1:100; Molecular Probes) or Alexa 488 anti-mouse antibody (dilution 1:100; Molecular Probes) and Cy-3 anti-rabbit antibody (dilution 1:100; Jackson ImmunoResearch) in PBS with 2% BSA. Images were collected using a confocal microscope (Leica SP2; Leica Microsystems, Exton, PA). Multiple fields (at least 10) containing several transfected cells were examined for each sample, and typical images were collected. 

Secretion Experiments

To study protein secretion, 8 × 10⁵ COS-7 cells were plated per 60-mm dish. Cells were transfected 24 hours after plating, using a transfection reagent (FuGENE 6; Roche, Indianapolis, IN) according to the protocol recommended by the manufacturer. Cells were incubated for 48 hours in DMEM containing 10% fetal bovine serum. To study the effects of reduced temperature on the secretion of WT and mutated Myoc, cells were incubated at 37°C overnight after transfection, then transferred to 30°C and incubated for an additional 24 to 26 hours. Cells were kept in DMEM without serum for the last 16 to 17 hours. Incubation media were concentrated 10× using concentration units (Centricon YM-10 or YM-30; Millipore, Bedford, MA). Cells were lysed in 0.5 mL extraction buffer (50 mM Tris-HCl [pH 7.5], 1 mM EDTA, 20 mM dithiothreitol [DTT], 1% NP-40, 0.2% SDS, and complete EDTA-free protease inhibitor cocktail; Roche). Equal volumes of concentrated incubation medium from different samples and ∼20 μg protein from cell lysates were used per lane in SDS-PAGE. Proteins were detected by Western blotting, using monoclonal antibodies against Flag (1:4000 dilution) or Myc (1:2000 dilution) epitopes (Sigma). All experiments were repeated at least twice. Secretion of SEAP was determined by measuring enzymatic activity using a commercial kit (Phospho Light; Applied Biosystems, Foster City, CA). All samples were prepared in triplicate, and experiments were repeated three times. 

Immunoprecipitation

COS-7 cells were grown and transfected as described in the previous section. Cells were washed twice with ice-cold PBS and lysed in 0.6 mL RIPA buffer (Santa Cruz Biotechnology; Santa Cruz, CA) supplemented with 20 mM DTT and complete EDTA-free protease inhibitor cocktail (Roche). The cell debris was removed by centrifugation at 12,000 rpm for 15 minutes, and 50 μL supernatant was saved to analyze the inputs by Western blotting. To reduce nonspecific binding, supernatants were incubated with 40 μL Protein A-agarose and 0.5 μg rabbit IgG for 2 hours at 4°C. After removing agarose beads by centrifugation, precleared lysates were incubated with 40 μL anti-Flag agarose beads at 4°C overnight. Agarose beads were washed four times with the lysis buffer, and bound proteins were eluted from the beads by boiling for 5 minutes in 2× SDS-PAGE sample buffer containing 100 mM DTT. The eluted proteins were analyzed by Western blotting using monoclonal antibodies against the Myc epitope (Sigma). 

Results

Intracellular Localization of Mouse Myoc and Effects of Mutations in the Olfactomedin Domain on Secretion and Localization of Mouse Myoc

Human and mouse Myoc proteins are 85% identical. ^{36 37 38} Moreover, amino acids in the positions corresponding to the mutations causing severe glaucoma are identical in the mouse and human sequences. Only about half of these amino acids are conserved in optimedin (Fig. 1) . Tyr437His, Ile477Asn, and Ile477Ser mutations in the human MYOC gene lead to the severe glaucoma phenotype. ^{39 40} Two of these mutations were introduced into mouse Myoc. In the mouse Myoc, these human mutations correspond to Tyr423His and Ile463Ser, respectively. 

Properties of the WT and mutated Ile463Ser mouse Myoc were tested after transfection into COS-7 cells. WT and mutated Ile477Asn human MYOC were used for comparison. In most transfected cells, Myoc proteins demonstrated a vesicular pattern of staining. WT human and mouse Myoc were preferentially located in the endoplasmic reticulum (ER) and Golgi, as was demonstrated by immunostaining of transfected cell with antibodies against ER marker calnexin (not shown) and a Golgi marker, golgin-97 (Figs. 2A 2B 2C and 2G 2H 2I) . At the same time, mutated human and mouse Myoc were located mainly in the ER and were excluded from Golgi (Figs. 2D 2E 2F and 2J 2K 2L) . We concluded that, similarly to human mutated MYOC, ³³ mutated mouse Myoc is not properly transported from the ER to Golgi. 

To examine the effects of mutations on mouse Myoc secretion, COS-7 cells were transfected with WT and mutated Myoc constructs. Similarly to human MYOC, ^{11 29 31 32 33} WT mouse Myoc was secreted from transfected COS-7 cells, while secretion of mutated Tyr423His and Ile463Ser mouse Myoc was significantly reduced (Fig. 3A) . It has been reported that culturing cells at 30°C, a condition that facilitates protein folding, promotes secretion of mutant human MYOC. ³³ Similarly, secretion of mutated mouse Myoc did indeed increase when transfected cells were incubated at 30°C instead of 37°C (Fig. 3B) . Secretion of WT mouse Myoc was not significantly affected by decreasing the temperature (Fig. 3B) . To analyze whether the presence of mutated mouse Myoc would affect the secretion of WT Myoc, COS-7 cells were cotransfected with mutated and WT Myoc constructs. As previously observed for mutated human MYOC, ^{29 30} the secretion of WT mouse Myoc was strongly inhibited in the presence of mutated mouse Myoc (Fig. 3C) . To test whether this inhibition of secretion of WT Myoc was due to formation of heteromeric complexes between WT and mutated Myoc, as observed for human MYOC, ³⁰ COS-7 cells were cotransfected with WT and mutated Myoc tagged with the Flag or Myc epitopes. Protein complexes were immunoprecipitated from cell lysates using anti-Flag-agarose beads, as described in the Methods section. The presence of Myc-tagged Myoc in the immunoprecipitated complexes indicated that WT and mutated mouse Myoc were able to form at least heterodimers (Fig. 4A) . Similarly to human MYOC, ^{30 31 41} mouse Myoc was often detected as a doublet of closely migrating bands. This doublet probably represents glycosylated and unglycosylated forms of mouse Myoc. 

To check how mutations in the olfactomedin domain may affect secretion of other olfactomedin domain–related proteins, mutations corresponding to the Tyr437His and Ile477Ser mutations in the human MYOC were introduced in rat optimedin A and B. Rat optimedin is identical with mouse optimedin. These two mutations corresponded to Tyr385His and Leu424Ser, respectively, in optimedin A; they corresponded to Tyr405His and Leu444Ser, respectively, in optimedin B. It has been previously shown that although both optimedin A and B are secreted from transfected COS-7 cells, optimedin A is secreted more efficiently than optimedin B ¹¹ . The Leu424Ser mutation in optimedin A significantly reduced its secretion, while the Leu444Ser mutation almost completely blocked secretion of optimedin B (Fig. 5) . At the same time, the Tyr405His mutation dramatically reduced secretion of optimedin B but did not significantly affect secretion of optimedin A (Fig. 5) . 

Effect of Myoc Mutations on Secretion of Different Proteins

The inhibition of WT MYOC secretion in the presence of the mutated MYOC may be a consequence of the inhibition of general secretion by mutated MYOC or, alternatively, a result of the specific interaction between WT and mutated MYOC. ³² To test the effects of mutated MYOC on general secretion, we selected several proteins which were unrelated to MYOC. 

SEAP, a constitutive secretory protein, ³⁴ was efficiently secreted from COS-7 cells 24 or 48 hours after transfection (Fig. 6) . The presence of WT or mutated human (Fig. 6A)and mouse (Fig. 6B)Myoc did not significantly affect the secretion of SEAP. 

Timp3, a tissue inhibitor of metalloproteinases, is a secreted protein that associates with the extracellular matrix and is expressed in different tissues. In the eye, Timp3 is highly expressed by the retinal pigmented epithelium ^{42 43} and localized in Bruch’s membrane. ⁴⁴ Myoc is also expressed in the retinal pigmented epithelium. ^{11 20} Timp3 was secreted after transfection into COS-7 cells, and the presence of mutated MYOC did not significantly reduce its secretion (data not shown). Thrombospondin-1 is a multimodular secreted protein associated with the extracellular matrix. It possesses multiple biological functions, including platelet aggregation, inflammatory response, and the regulation of angiogenesis during wound repair and tumor growth. It is expressed in many tissues, including the trabecular meshwork of the eye. ¹⁶ Thrombospondin was secreted after transfection into COS-7 cells, and its secretion was not affected by the presence of WT or mutated human MYOC (data not shown). On the basis of these results, we concluded that the general secretory pathway was not affected by the presence of mutated MYOC. 

Since the general secretory pathway was apparently not affected in cells expressing mutated MYOC, inhibition of the secretion of other proteins may require the formation of complexes of these proteins with mutated MYOC. To test whether mouse and human Myoc proteins are able to form heterodimers, COS-7 cells were cotransfected with mutated human MYOC and WT mouse Myoc tagged with the Myc and Flag epitopes, respectively. Thrombospondin-1-Myc was used as a control in these experiments. Transfected cells were lysed and protein complexes were immunoprecipitated using anti–Flag-agarose beads. The presence of Myc-tagged human MYOC and the absence of Myc-tagged thrombospondin-1 in the immunoprecipitated material indicated that mouse Myoc formed complexes with human MYOC but not with thrombospondin-1 (Fig. 4B) . The formation of heterodimers between human and mouse Myoc proteins may prevent secretion of WT mouse Myoc in the presence of mutated human MYOC and vice versa. Indeed, mouse Myoc was efficiently secreted in the presence of WT human MYOC (Fig. 7) . However, its secretion was blocked in the presence of mutated human MYOC (Fig. 7) . It is interesting to note that a dramatically reduced secretion of WT human MYOC in the presence of mutated Ile477Asn human MYOC was observed in some experiments (Fig. 7 , lower panel; compare lanes 1 and 3). 

To study whether mutated human MYOC may affect secretion of more distantly related proteins, COS-7 cells were cotransfected with olfactomedin and WT or mutated human MYOC. Olfactomedin is a secreted protein that may play an important role in neural crest production in chickens ⁹ and in neurogenesis in Xenopus. ⁴⁵ In the eye, the olfactomedin gene is expressed in the retina. ^{11 16} Olfactomedin protein was also detected in the epithelial cells of the iris, ciliary body, and cornea. ⁴⁶  

Similarly to MYOC, olfactomedin contains the olfactomedin domain in the C-terminal part of the protein molecule. The olfactomedin domains of MYOC and olfactomedin are 40% identical. Olfactomedin was efficiently secreted from COS-7 cells in the presence of WT and mutated Ile477Asn human MYOC (Fig. 8) . We concluded that MYOC and olfactomedin did not form complexes that may be retained in the ER. 

Discussion

The functions of WT MYOC in the eye or in other tissues where it is expressed are still not clear. MYOC has been implicated in steroid-induced ocular hypertension with glaucoma, because MYOC is upregulated in cultured trabecular meshwork treated with glucocorticoids, and glucocorticoid treatment may induce elevated IOP in ∼40% of people. ⁴⁷ In vitro data suggest that overexpression of WT MYOC in cultured trabecular meshwork cells induces a dramatic loss of actin stress fibers and focal adhesion and reduces cell adhesion to fibronectin. ⁴⁸ However, there is no direct in vivo evidence that elevated MYOC levels lead to IOP elevation and glaucoma. Moreover, no pathologic changes in the eye were detected in mice either in the absence of Myoc ²⁵ or in the presence of significantly higher levels of Myoc. ^{26 49}  

It is now well established that expression of mutated MYOC may lead to glaucoma associated with elevated IOP. In the eye, the highest levels of MYOC expression were detected in the trabecular meshwork, an essential component of the aqueous humor outflow system, and the sclera. ¹⁶ Expression of high levels of misfolded nonsecreted mutated MYOC may lead to trabecular meshwork cell death, trabecular meshwork dysfunction, and, subsequently, a glaucoma phenotype. ³³ It appears that expression of low levels of mutated MYOC in the skeletal muscles or brain does not lead to detectable pathologic changes. These observations indicate that cells are able to tolerate low levels of mutated MYOC or that the cells of the eye irido-corneal angle (and probably sclera) are more sensitive to the presence of mutated MYOC. 

Animal models, including mouse models, represent a reasonable approach to studying the mechanisms of WT and mutated MYOC action in the eye and other tissues. Recent data indicate that elevated levels of Tyr437His human MYOC in the lens of transgenic mice lead to structural alterations in nuclear lens fibers, resulting in nuclear cataract and eventual lens rupture at the posterior pole. ⁴⁹ Elevated levels of WT mouse Myoc were produced in the eyes of transgenic animals with BAC DNA containing the full-length mouse Myoc gene and long 5′- and 3′-flanking sequences. This BAC DNA reproduced the expression pattern of the endogenous Myoc gene. The same BAC DNA with a mutation in the Myoc gene could be used to express mutated mouse Myoc in the eyes of transgenic animals. Because it is not known how mutations in the mouse Myoc gene will affect the mouse Myoc protein, as a prerequisite for the creation of transgenic mice expressing mutated mouse Myoc, it was important to demonstrate that the properties of mutated mouse Myoc are similar to those of mutated human MYOC. In this work, the properties of mutated mouse Myoc were tested using COS-7 cells transiently transfected with WT and mutated constructs. These cells are often used in cell biology for studies on protein trafficking and secretion. ⁵⁰  

Two types of trials were used to compare the properties of mutated human and mouse Myoc: intracellular localization and secretion. The properties of mutated human and mouse Myoc were very similar by both of these tests. WT human and mouse Myoc were detected in both the ER and Golgi. However, mutated human Ile477Asn MYOC and mutated mouse Ile463Ser Myoc were retained in the ER and did not transit to the Golgi. As a consequence of ER retention, two mutated mouse Myoc tested, Tyr423His and Ile463Ser, were not secreted after transfection into COS-7 cells, similar to mutated human MYOC. Therefore, expression of mutated mouse Myoc in the tissues of the irido-corneal angle of transgenic mice may result in the same effects as expression of mutated human Myoc in the human eye. Experiments are under way in our laboratory to characterize transgenic mice with BAC DNA containing point mutations in the mouse Myoc gene. 

Another approach would be to express mutated human MYOC in the tissues of the mouse irido-corneal angle. In this case, it was important to establish how mutated human MYOC would affect the secretion of normal mouse Myoc. Results obtained in this work demonstrated that the presence of mutated human MYOC prevented the secretion of WT mouse Myoc by transfected COS-7 cells. At the same time, the secretion of several other proteins tested (TIMP3, thrombospondin-1, alkaline phosphatase, olfactomedin) was not dramatically affected. Although these data do not exclude the possibility that the presence of mutated Myoc suppresses the secretion of other proteins, these observations, together with previously published data, suggest that the presence of mutated Myoc does not block the general secretory pathway. It has been shown that MYOC is able to form dimers and higher-molecular-weight aggregates and that the N-terminal part of MYOC is critical for dimer formation. ^{11 51} Mouse Myoc is also able to form heterodimers with mutated mouse and human Myoc proteins. These heterodimers may form higher-molecular-weight aggregates that accumulate in the ER and are not secreted. Several lines of transgenic mice expressing mutated human MYOC in the tissues of the irido-corneal angle were produced in our laboratory. These mice are currently being analyzed. 

Olfactomedin is also able to form dimers and high-molecular-weight aggregates. ⁶ Secretion of olfactomedin in the presence of mutated MYOC indicates that olfactomedin and myocilin do not form heterodimers and do not form intracellular aggregates containing both proteins. Another possibility is that the interaction between olfactomedin and myocilin is too weak to retain olfactomedin in the ER in the presence of mutated MYOC. 

When different olfactomedin domain–containing proteins are compared, the olfactomedin domain is more conserved compared with other parts of the protein molecules. Most glaucoma-causing mutations are located in the olfactomedin domain of myocilin. Only a few of the amino acids affected by mutations in MYOC are conserved in other olfactomedin domain–containing proteins. Data obtained in this work indicate that conservation of the residues does not necessary correlate with the properties of the proteins. For example, the Tyr405His mutation in optimedin A corresponding to the Tyr437His mutation in human MYOC does not prevent the secretion of mutated optimedin A from COS-7 cells. It is possible that the three-dimensional organization of the olfactomedin domain is different in different olfactomedin domain–containing proteins. 

In summary, the properties of mutated mouse Myoc are similar to those of mutated human MYOC. Our data indicate that mouse and human MYOC are able to form heterodimers and probably higher-molecular-weight aggregates. Expression of mutated mouse or human Myoc does not block a general secretory pathway. Because the properties of the mutated mouse and human Myoc are similar, the expression of mutated mouse or human Myoc in the eye in transgenic animals or the generation of Myoc knock-in mutant lines may lead to the development of a genetic mouse model of Myoc-induced glaucoma. 

 Supported by the National Eye Institute Intramural Program.

 Submitted for publication February 18, 2005; revised July 6, 2005; accepted November 16, 2005.

 Disclosure: I. Malyukova, None; H.-S. Lee, None; R.N. Fariss, None; S.I. Tomarev, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

Sequence comparison of the olfactomedin domains of the human and mouse myocilins with that of the mouse optimedin. The human MYOC sequence is shown in full; only differing amino acids are shown for other sequences. Positions of known glaucoma-causing mutations are highlighted (green and rose); rose: positions that were mutated in the mouse Myoc and optimedin sequences in the present work. Numbering for optimedin corresponds to the optimedin B form. (*) Insertion introduced in the mouse optimedin sequence to maximize identity.

Figure 2.

 

Subcellular localization of recombinant WT and mutated human and mouse Myoc in transfected COS-7 cells. COS-7 cells were transfected with WT human (A–C), Ile477Asn human (D–F), WT mouse (G–I), and Ile463Ser mouse Myoc (J–L). COS-7 cells were stained with polyclonal antibodies against mouse Myoc (A, D, G, J) and monoclonal antibodies against golgin-97 (B, E, H, K) 48 hours after transfection. Merged images are included (C, F, I, L). Typical images are shown.

Figure 3.

 

Secretion of WT and mutated mouse Myoc after transfection in COS-7 cells. Cells were transiently transfected with mouse Myoc tagged with the Flag epitope. Proteins in cell lysates and in incubation media were separated by SDS-PAGE, transferred to nitrocellulose, and stained with Flag monoclonal antibodies. (A) Flag immunostaining of incubation media and cell extracts after transfection of COS-7 cells with WT or mutated Myoc. (B) As in (A), but cells were incubated at 30°C or at 37°C during the last 24 hours before collection of the incubation media and cell lysis. (C) Flag immunostaining of cell extracts and incubation media after cotransfection of COS-7 cells with WT and mutated Myoc.

Figure 4.

 

Heterodimer formation between mutated and WT Myoc proteins. (A) COS-7 cells were transiently transfected with WT and mutated Tyr423His mouse Myoc tagged with the Flag and Myc epitopes, respectively. Total cell lysates (20 μg per lane) were separated by SDS-PAGE, transferred to nitrocellulose, and stained with Myc (upper panel) or Flag antibodies (middle panel). The rest of the cell extracts were immunoprecipitated with anti-Flag antibodies. The formation of heterodimers between WT and mutated mouse Myoc was detected by staining of immunoprecipitates with Myc antibodies (lower panel). (B) As in (A), but mutated Ile477Asn human MYOC and WT mouse Myoc were used in these experiments. Thromobospondin-1 tagged with the Myc epitope was used as a negative control. Note the absence of the thrombospondin-1 band in the immunoprecipitates. CE, cell extracts; IP, immunoprecipitates.

Figure 5.

 

Secretion of mutated optimedin A and optimedin B after transfection in COS-7 cells. Cells were transiently transfected with mutated optimedin A or optimedin B tagged with the Myc epitope. Proteins in cell lysates and in incubation media were separated by SDS-PAGE, transferred to nitrocellulose, and stained with Myc monoclonal antibodies.

Figure 6.

 

Secretion of SEAP from COS-7 cells in the presence of WT and mutated Ile477Asn human (A) and Tyr423His and Ile463Ser mouse Myoc (B). Cells were incubated for 24 or 48 hours, incubation media was diluted 200×, and activity of SEAP was measured.

Figure 7.

 

Secretion of mouse Myoc from COS-7 cells in the presence of WT and mutated Ile477Asn human MYOC. Cells were transiently cotransfected with WT or mutated human MYOC together with WT mouse Myoc. Mouse and human MYOC were tagged with Flag (upper) and myc epitopes (lower), respectively. Proteins in cell lysates and in incubation media were separated by SDS-PAGE, transferred to nitrocellulose, and stained with Flag or Myc antibodies. IM, incubation media; CE, cell extracts.

Figure 8.

 

Secretion of olfactomedin from COS-7 cells in the presence of WT and mutated Ile477Asn human MYOC. Cells were transiently cotransfected with WT or mutated MYOC together with olfactomedin. MYOC and olfactomedin were tagged with Flag (upper) and Myc epitopes (lower), respectively. Cells were also transfected with olfactomedin alone for control. Proteins in cell lysates and in incubation media were separated by SDS-PAGE, transferred to nitrocellulose, and stained with Flag or Myc antibodies.

The authors thank Ritu Trivedi for the production of mutated optimedin constructs and David Rawnsley for the professional editing of this manuscript.