Molecular Vision 1999; 5:19 <http://www.molvis.org/molvis/v5/p19/> Received 30 June 1999 | Accepted 13 August 1999 | Published 23 August 1999

Download Reprint

A trabecular meshwork glucocorticoid response (TIGR) gene mutation affects translocational processing

Carin C. Zimmerman,¹ Vishwanath R. Lingappa,² Julia E. Richards,^3,4 Frank W. Rozsa,³ Paul R. Lichter,³ Jon R. Polansky¹
 
 

¹Cellular Pharmacology Laboratory, Department of Ophthalmology and ²Department of Physiology, University of California San Francisco, San Francisco, CA; ³Department of Ophthalmology and Genetics and ⁴Department of Epidemiology, University of Michigan, Ann Arbor, MI

Correspondence to: Jon Polansky or Carin Zimmerman, Department of Ophthalmology, Room K301, University of California San Francisco, San Francisco, CA, 94143-0730; Phone: (415) 476-7993; email: drcarin@itsa.ucsf.edu

Abstract

Purpose: To examine possible effects of the E323K mutation in the trabecular meshwork glucocorticoid response (TIGR) gene (also known as myocilin [MYOC]), using assays of translocational processing through the endoplasmic reticulum (ER). The E323K mutation was of particular interest, since the mutation shows a strong association with early onset open-angle glaucoma, but has a minimal predicted effect on protein structure.

Methods: Normal and mutant TIGR cDNA constructs were used to generate protein products in the presence of endoplasmic reticulum (ER) membranes, using an assay previously developed to detect alterations in the ER translocation function. "Paused" regions for potential protein modifications were defined by proteinase K (PK) sensitivity in the presence of ER membranes, with the ability to restart translocation when treated with EDTA. The effects of the E323K mutation were evaluated, as well as mutations located on either side of E323K (G246R, G364V, P370L) as the other mutations had substantial predicted structural changes in addition to clear disease associations.

Results: The native TIGR molecule was observed to have a paused region that corresponds to the region of highest olfactomedin (OLF) homology. The E323K mutation, located near the beginning of this region, dramatically altered the normal pattern of nascent proteins observed in the translocational pausing assay. A prominent band appeared with the E323K mutation, which could represent a new product or a marked enhancement of a faint band normally seen, approximately 3 kDa higher than the major paused band. The other TIGR mutants examined did not show this effect.

Conclusions: The major translocational pause that starts near the beginning of the region of high OLF homology may help to explain the high frequency of glaucoma-associated mutations in this area. The observed effect of the E323K mutation on the products of translocational processing suggests a delay in the normal pausing process of TIGR biogenesis. This delay points to a potentially distinct pathogenic mechanism for E323K as compared with the other TIGR mutations so far evaluated.

Introduction

Mutations in the trabecular meshwork glucocorticoid response (TIGR) gene, also known as myocilin (MYOC), are associated with adult and juvenile glaucomas, including glaucoma previously linked to the GLC1A locus [1-5]. The 55 kDa TIGR protein is induced by dexamethasone and oxidative stress [4,5]. Increased staining for TIGR was detected by immunohistochemistry in the trabecular meshwork of patients with exfoliation glaucoma as well as those with adult primary open-angle glaucoma (POAG) [6]. TIGR exists in multiple forms cellularly and extracellularly, and cleavage of its predicted signal sequence has been demonstrated [4,5]. TIGR expression has also been reported in a number of other tissues including muscle [5] and photoreceptor cells [1]. Most of the known TIGR mutations cluster in exon 3, which includes the olfactomedin (OLF) domain [7], but no function has yet been identified for OLF domains in any of the several contexts in which they have been found so far [7-13]. Thus, in spite of a variety of ongoing studies, little is known about normal TIGR functions, about the role of modified forms of the protein, or about the mechanisms by which TIGR sequence variants lead to disease [4,5].

A translocational processing assay, which examines translocation of a nascent protein chain across the membrane of the endoplasmic (ER) into the lumen, has previously provided a means for investigating the regulation of secretory molecules and effects of altered protein sequence in other disease processes [14-17]. This assay, involving in vitro translation of message in the presence of microsomal membranes, allows detection of novel transient intermediates that may be related to the mechanisms by which the protein causes disease. Our evidence shows that TIGR has a cleaved signal sequence, is secreted extracellularly, and shows N-glycosylation [4,5], a modification normally carried out in the ER. Therefore, it appeared reasonable to use assays of the ER processing pathway to examine TIGR.

A recent paper by Rozsa and colleagues that examined predicted structural consequences of TIGR sequence variants and localized some clusters of mutations in the vicinity of motifs at which modifications such as phosphorylation are predicted to occur [7]. Interestingly, the study suggested a mechanism by which certain TIGR mutations occurring in clusters could produce similar effects by preventing phosphorylation through alteration of the residues of conserved casein kinase II or cAMP-dependent kinase motifs. The study also predicted possible misfolding of TIGR produced by some of the sequence variants, which raises further questions regarding possible effects on correct trafficking of the protein through cellular pathways in which protein modifications occur. Although these analyses suggested related mechanisms for multiple mutations and provide a framework for subsequent experimental design, they could not account for how some mutations associated with disease phenotype, especially E323K, might produce their effects. Although the E323K sequence variant causes a net +2 change in charge, it is not predicted to cause structural alterations and is not located in the immediate vicinity of one of the conserved phosphorylation motifs at which some of the mutations cluster.

Here we report differences attributable to the E323K mutation that are detected by an assay of translocational processing of secretory molecules into the ER [14-17]. Biochemical species observed for E323K differed markedly from those produced when normal TIGR was tested in the same system, while mutations that alter residues 246, 364, and 370 showed little or no effects in the assay. A model by which the E323K mutation produces this effect is discussed.

Methods

Evaluations of TIGR protein processing used recently described methods [14,15] to investigate alterations during ER translocation. Chuck and Lingappa in their studies of apolipoprotein B, [14] were able to identify regions of the nascent protein where translocation (but not translation) was apparently stopped for a period of time, and then restarted later during chain growth. Most importantly, although translocation was stopped during the pause window, translation continued, with the growing chain exposed to the cytosol via an altered ribosome-membrane junction. Termed translocational "pausing", this process is thought to be important for complex events such as protein folding and modification that could occur during the "pause."

As pausing is a transient process, it was necessary to design a system where paused intermediates could be detected. This was accomplished using truncated constructs [14,15]. For the full-length constructs, the presence of a stop codon results in the dissociation of the ribosome and the subsequent release of the polypeptide chain into the ER lumen. In a truncated construct, because there is no stop codon, the ribosome remains attached (at least for a prolonged period of time). Hence the chain will be "frozen", and a transient paused intermediate can be detected. Since pausing is characterized as stopping and restarting, truly paused chains should have the ability to restart. The addition of EDTA, which substantially disassembles the ribosome, releases the polypeptide chain and allows it to complete translocation into the ER lumen. This results in a full-length band upon subsequent proteinase K (PK) digestion because the protein is completely protected from PK digestion when it is within the ER lumen [14,15].

Constructs

The full-length TIGR cDNA clone has been previously described by Nguyen et al [5]. The TIGR construct for subsequent transcription/translation, p1ATG-TIGR, was made by creating a novel BamHI site at the 5' end of the TIGR coding sequence by PCR, cutting with BamHI and SalI to excise the TIGR coding sequence and insert it into a pSP64-derived expression vector (Promega, Madison WI) containing the 5' UT of Xenopus globin [14,15]. The p1ATG-TIGR plasmid is shown in Figure 1.

The Quik-Change Site-directed Mutagenesis Kit (Stratagene, La Jolla CA) was used on the p1ATG-TIGR construct to create constructs containing one of the four mutations to be studied: G246R, E323K, G364V, or P370L. All were verified by sequencing. TIGR was transcribed/translated from p1ATG-TIGR (Figure 1) to produce native TIGR. In some of the translocational pausing experiments, p1ATG-TIGR or the mutant cDNA constructs were digested with restriction enzymes whose sites are unique within the cDNA sequence, Avr II or Hinc II (Figure 2), prior to their use in the transcription step. SP6 polymerase was used as previously described to produce truncated transcripts from the normal or mutant constructs that had been cut with AvrII or HincII, or to produce full-length transcript by transcription of non-truncated versions of the constructs [8-10]. Proteinase K (PK) digestion was used to detect points on the protein that are sensitive to PK digestion when the nascent protein chain is "frozen" in a paused state that results from the failure of the ribosome to release the nascent peptide chain [14-17].

Translocation assays

Translation, incorporation of ³⁵S-methionine, and proteolysis were as described previously [14-17]. The translation step was carried out in vitro through the use of rabbit reticulocyte lysates and microsomal membranes from dog pancreas prepared as described previously [14,18]. The translation reactions were incubated for 50 min at 25 °C, after which aliquots were taken and treated with EDTA to a final concentration of 20 mM and incubated at 25 °C for 10 min and then transferred to ice. The addition of EDTA before PK treatment is used as an essential control to "restart" translocation by dissociating the ribosome and releasing the nascent chain into the ER lumen. An additional control, treatment with PK and detergent, was performed to test whether the observed protection from PK digestion was due to an association with the ER membranes in the preparation. Aliquots with and without EDTA treatment were adjusted to 0.2 mg/ml proteinase K and incubated for 60 min at 0 °C before proteolysis was terminated by PMSF to 10 mM and heating in 10 volumes of 2% SDS for 5 min. This method of terminating the proteolysis reactions was found to completely inactivate the PK, leaving proteins in the lumen of the ER undigested [14-17]. Volumes equivalent to 1 µl of initial translation reaction were applied for each sample.

Samples were analyzed by SDS-PAGE on 15% gels and the proteins were visualized by autoradiography. For the TIGR truncated constructs where pausing is observed (as defined by Chuck and Lingappa [14]), this region was estimated by subtracting the observed molecular weight of the new band that appeared after PK digestion from that of the undigested translation product.

Results

Assays of Translocational Pausing of Native TIGR

Figure 2 presents the important motifs and functional domains of TIGR, as well as the mutations considered in this study. The position of restriction enzyme sites used to produce truncated constructs (Figure 2) are indicated by the blue arrows. The location of the translocational paused region, TPR, approximates the area of highest olfactomedin homology, as determined experimentally by the data in Figure 3.

Figure 3 shows the results of the translocational pausing assay for the native (non-mutated) TIGR molecule produced from the full-length construct, along with results for truncated TIGR produced from constructs cut at the Avr II and Hinc II sites. As shown by similarity of band sizes in Lanes 1, 2, and 3 of Figure 3A, use of a full-length transcript results in production of only full-length TIGR species; as expected, lower molecular weight products that are indicative of pausing for TIGR are not readily visualized in this assay (Figure 3A). Lane 1 shows the translation products in the absence of proteinase K (PK), with the expected doublet bands for TIGR (due to N-glycosylation as previously documented) at approximately 55 kDa [4]. Lane 2 shows the same product after PK digestion and lane 3 shows treatment with EDTA, followed by PK digestion. The addition of EDTA before PK treatment restarted translocation, resulting in a fully protected (full-length) band (Lane 3). No lower molecular weight bands with any treatment are observed. A control, treatment with PK plus detergent, results in the complete digestion of the TIGR bands. This demonstrates that the protection from PK digestion in the previous lanes was indeed due to association with the ER membrane (lane 4).

Figure 3B shows the results for the protein produced from the Avr II truncated construct. Lane 1 shows the translation product without PK digestion, approximately 32 kDa, as expected with the Avr II truncation. Treatment with PK, and EDTA followed by PK (lanes 2 and 3), show no shift in band size, suggesting that there are no paused domains in the region just prior to the Avr II truncation site.

Use of the Hinc II truncated construct allowed the detection of a translocational pause as shown in Figure 3C. Lane 1 shows the translation products without PK treatment. A protein band is visualized on the gel, approximately 47 kDa, as expected with the HincII truncation. Lane 2 shows translation products digested with PK. Here we see a lower molecular weight band, approximately 33 kDa. The assay predicts that the region of the nascent chain that is paused is susceptible to PK digestion which results in this lower molecular weight band. The approximate location of the paused region can be estimated based on the molecular weight of this band.

Assays of Translocational Pausing on TIGR Mutations

Figure 4 shows the results of testing HincII truncated mutant cDNA expression constructs containing missense mutations known to cause glaucoma in humans. Figure 4A shows G246R, Figure 4B shows E323K, Figure 4C shows G364V, and Figure 4D shows P370L. The products without PK digestion (Lane 1), with PK digestion (Lane 2) and EDTA treatment followed by PK digestion (Lane 3) are shown for each mutant construct. The same lower molecular weight band at approximately 33 kDa is present in all of the mutant constructs after PK digestion (Lane 2); indicating that the same pause was occurring for both normal (Figure 3) and mutant (Figure 4). In the E323K construct (Figure 4B), however, there was a strong additional band, approximately 36 kDa, and indicated by the arrow. The presence of a higher molecular weight band indicates that more of the chain was protected from protease and therefore this pause occurred downstream of the original pause. One possibility is that the E323K mutation may allow a subset of the nascent proteins to continue beyond the normal pause point until they arrive at a new pause point farther along the protein chain. Alternatively, the mutation may cause more of the chains to pause at this second location for a longer time (Figure 4). The 36 kDa band was detected reproducibly in three additional experiments, two of which contained the N-glycosylation peptide inhibitor leupeptin (data not shown). None of the mutations tested had a major effect on the observed paused intermediates (compared to native TIGR in Figure 2C) except E323K.

Discussion

The finding that TIGR protein biogenesis is characterized by a major translocational pause provides potentially useful leads concerning the mechanism(s) for different disease-associated mutations. The observation that the region of TIGR high olfactomedin homology corresponds with the paused region digested by PK is particularly important, and suggests a role for modifications of the molecular interactions that take place as part of the translocational pathway involving the translocon (see recent review by Lingappa [19]). Since this pathway is part of the ER protein secretory process, effects of crucial ER sorting functions and chaperone proteins need to be considered.

Recent studies have shown that certain "simple" secretory proteins, such as prolactin, have no paused regions, while others like apolipoprotein B, and now TIGR, show a translocational profile that is discontinuous. The evidence suggests that translocation of the nascent protein chain appears to transiently stop before completing transport into the ER lumen. This translocational pausing may actually represent a crucial step for cellular modifications, folding, subunit assembly, etc. as described by Chuck and Lingappa [14], and Hegde and Lingappa [17].

Like other early events of protein biogenesis (such as targeting to the ER membrane, signal sequence cleavage, and N-linked glycosylation), translocational pausing occurs cotranslationally. The study of protein assembly in the ER of living cells is often impaired by the occurrence of rapid and regulated degradation of a subpopulation of nascent protein assembly intermediates which could be quite important in understanding normal and aberrant secretory processes [20,21]. Use of the cell-free translation system with microsomal membranes provides a means to "biochemically dissect" important processes within nascent chain translocation [22], with the usefulness of this approach for defining events of in vivo relevance supported by studies of other secretory proteins [23-25].

Detection of translocational pausing requires detecting transient intermediates. Prior studies by Chuck and Lingappa [14] demonstrated that this could be achieved using truncations of the full-length cDNA construct, which generate a relatively homogeneous nascent chain population engaged in translocational pausing [14-17]. The results obtained here with the Hinc II construct matched the criteria for defining a paused region and distinguishing them from non-paused translocation intermediates: (i) the translocationally paused chain is accessible to PK digestion from the cytosolic side under conditions where simple secretory proteins are fully protected by the ribosome-membrane junction from protease digestion, (ii) the proteolysis of a paused nascent chain from the cytosolic side of the membrane not only diminishes the amount of full length chains, but also generates a discrete lower molecular weight product that represents the domain that is in the ER lumen (protection of that fragment from proteases is, of course, dependent on membrane integrity), (iii) the translocationally paused nascent chain intermediates, but not integrated transmembrane chains, are shown to restart translocation after EDTA treatment [14-17]. By assessing the protease accessibility of the nascent chain after EDTA treatment, truly paused nascent chains are distinguished (they become fully protected from proteases) from integrated nascent chains or artifacts of protease overdigestion (their proteolytic accessibility is not changed by EDTA treatment).

Having established these criteria for the pause in TIGR and determining its estimated location, it appeared reasonable to consider that some mutations near the beginning of the paused region could produce their effects by disrupting the "signals" that initiate or regulate the pausing process itself. The findings observed with the E323K mutation in the translocational pausing assay supports this idea since the mutation itself has little or no predicted structural effect, yet is associated with juvenile glaucoma. The 323 position is located just prior to the region of higher OLF homology [4], as shown in Figure 2. As a result of a defect in pausing, insufficient or abnormal modification of the nascent chain may occur, although further studies will be needed to evaluate these potential implications of a pause at this location. This could represent a mechanism by which one would observe a similar phenotype as compared with a mutation causing structural changes.

The discovery of a major pause in TIGR protein processing corresponding to the region defined by others as the OLF homology domain by itself provides a reason to investigate the potential role(s) of the pausing process and protein intermediates generated with regards to new TIGR gene mutations. Structural changes such as phosphorylation, as suggested by Rozsa et al. [7], and subunit interactions, as suggested by Morissette et al [26], involve sites within the OLF homology domain, the paused area of TIGR. Other potential alterations in the paused region, including effects on protein folding, subunit associations, etc. are also possible and could be relevant to the native TIGR functions and pathogenic mechanisms. The sorting process at the ER level for protein alterations could be involved in producing pathogenic effects, including roles for chaperone proteins. Effects on ER processing, as well as direct structural alterations, could influence the ability of TIGR to reach or influence its appropriate cellular/extracellular sites. Appropriate analyses of the complex interactions of TIGR mutations may require a consideration for cell-specific and hormonally induced factors. The evaluation of aberrant intermediates may also be important. Such functional intermediates can be studied and manipulated, much as has been done for various translocation-associated events [14-17,27].

Clearly, there is a long-term need to relate biochemical and cellular findings such as those presented here, back to the clinical situation. The investigation of families, rather than just individuals, is beginning to reveal patterns of clinical characteristics associated with some of the different mutations. For example, P370L may be the most severe of the TIGR coding sequence mutations so far identified, with consistently earlier age at onset and accompanying high pressures and tendency to be unresponsive to medications [3,7]. By contrast, the E323K mutation, which may involve different mechanisms, does not appear to be associated with such an extremely high IOP and seems more responsive to medications [7,28]. Future investigations of how alterations associated with TIGR protein processing, including effects on translocational pausing, influence disease mechanisms may help to explain differences in age at diagnosis, such as that observed when comparing data on the E323K family with data for the P370L family [7]. We know that differences, including apparent anticipation in the E323K family, cannot be accounted for by differences in medical monitoring of these two families. Hopefully, additional studies of translocational processing and other aspects of mutant TIGR protein trafficking and processing may offer further insights into these fascinating questions of how the clinical observations relate to the underlying pathophysiology of the disease.

Overall, our observation of a major pause in TIGR protein biogenesis, together with the shift in products seen with the E323K mutation, indicate a need for further and more detailed evaluations of TIGR as a complex secretory protein. Perhaps these efforts combined with organ culture and in vivo models that allow a measurement of physiological changes to compare with biochemical assessments will allow an improved understanding of how specific TIGR gene mutations result in glaucoma.

Acknowledgements

This study was supported by National Institutes of Health grants EY02477 (JRP), EY09580 (JR), EY02162 (UCSF Ophthalmology Core Grant), The Glaucoma Research Foundation, The Van Arnam Glaucoma Research Fund, and That Man May See, Inc.

References

1. Kubota R, Kudoh J, Mashima Y, Asakawa S, Minoshima S, Hejtmancik JF, Oguchi Y, Shimizu N. Genomic organization of the human myocilin gene (MYOC) responsible for primary open angle glaucoma (GLC1A). Biochem Biophys Res Commun 1998; 242:396-400.

2. Stone EM, Fingert JH, Alward WLM, Nguyen TD, Polansky JR, Sunden SLF, 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.

3. Adam MF, Belmouden A, Binisti P, Brezin AP, Valtot F, Bechetoille A, Dascotte JC, Copin B, Gomez L, Chaventre A, Bach JF, Garchon HJ. Recurrent mutations in a single exon encoding the evolutionarily conserved olfactomedin-homology domain of TIGR in familial open-angle glaucoma. Hum Mol Genet 1977; 6:2091-7.

4. Polansky JR, Fauss DJ, Chen P, Chen H, Lufjen-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.

5. 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.

6. 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.

7. Rozsa FW, Shimizu S, Lichter PR, Johnson AT, Othman MI, Scott K, Downs CA, Nguyen TD, Polansky J, Richards JE. GLC1A mutations point to regions of potential functional importance on the TIGR/MYOC protein. Mol Vis 1998; 4:20 <http://www.molvis.org/molvis/v4/p20/>.

8. Yokoe H, Anholt RR. Molecular cloning of olfactomedin, an extracellular matrix protein specific to olfactory neuroepithelium. Proc Natl Acad Sci U S A 1993; 90:4655-9.

9. Danielson PE, Forss-Petter S, Battenberg EL, deLecea L, Bloom FE, Sutcliffe JG. Four structurally distinct neuron-specific olfactomedin-related glycoproteins produced by differential promoter utilization and alternative mRNA splicing from a single gene. J Neurosci Res 1994; 38:468-78.

10. Davletov BA, Shamotienko OG, Lelianova VG, Grishin EV, Ushkaryov YA. Isolation and biochemical characterization of a Ca2+-independent alpha-latrotoxin-binding protein. J Biol Chem 1996; 271:23239-45.

11. Lelianova VG, Davletov BA, Sterling A, Rahman MA, Grishin EV, Totty NF, Ushkaryov YA. Alpha-latrotoxin receptor, latrophilin, is a novel member of the secretin family of G protein-coupled receptors. J Biol Chem 1997; 272:21504-8.

12. Abderrahim H, Jaramillo-Babb VL, Zhou Z, Vollrath D. Characterization of the murine TIGR/myocilin gene. Mamm Genome 1998; 9:673-5.

13. Karavanich C, Anholt RR. Evolution of olfactomedin. Structural constraints and conservation of primary sequence motifs. Ann N Y Acad Sci 1998; 855:294-300.

14. Chuck SL, Lingappa VR. Pause transfer: a topogenic sequence in apolipoprotein B mediates stopping and restarting of translocation. Cell 1992; 68:9-21.

15. Kivlen MH, Dorsey CA, Lingappa VR, Hegde RS. Asymmetric distribution of pause transfer sequences in apolipoprotein B-100. J Lipid Res 1997; 38:1149-62.

16. Perara E, Lingappa VR. A former amino terminal signal sequence engineered to an internal location directs translocation of both flanking protein domains. J Cell Biol 1985; 101:2292-301.

17. Hegde RS, Lingappa VR. Sequence-specific alteration of the ribosome-membrane junction exposes nascent secretory proteins to the cytosol. Cell 1996; 85:217-28.

18. Walter P, Blobel G. Preparation of microsomal membranes for cotranslational protein translocation. Methods Enzymol 1983; 96:84-93.

19. Hegde RS, Lingappa VR. Regulation of protein biogenesis at the endoplasmic reticulum membrane. Trends Cell Biol 1999; 9:132-7.

20. Borchardt RA, and Davis RA. Intrahepatic assembly of very low density lipoproteins. Rate of transport out of the endoplasmic reticulum determines rate of secretion. J Biol Chem 1987; 262:16394-402.

21. Furukawa S, Sakata N, Ginsberg HN, Dixon JL. Studies of the sites of intracellular degradation of apolipoprotein B in Hep G2 cells. J Biol Chem 1992; 267:22630-8.

22. Blobel G, Dobberstein B. Transfer of proteins across membranes. I. Presence of proteolytically processed and unprocessed nascent immunoglobulin light chains on membrane-bound ribosomes of murine myeloma. J Cell Bio 1975; 67:835-51.

23. Simon K, Perara E, Lingappa VR. Translocation of globin fusion proteins across the endoplasmic reticulum membrane in Xenopus laevis oocytes. J Cell Biol 1987; 104:1165-72.

24. Hann BC, Walter P. The signal recognition particle in S. cerevisiae. Cell 1991; 67:131-44.

25. Deshaies RJ, Sanders SL, Feldheim DA, Schekman R. Assembly of yeast Sec proteins involved in translocation into the endoplasmic reticulum into a membrane-bound multisubunit complex. Nature 1991; 349:806-8.

26. Morissette J, Clepet C, Moisan S, Dubois S, Winstall E, Vermeeren D, Nguyen TD, Polansky JR, Cote G, Anctil JL, Amyot M, Plante M, Falardeau P, Raymond V. Homozygotes carrying an autosomal dominant TIGR mutation do not manifest glaucoma. Nat Genet 1998; 19:319-21.

27. Hegde RS, Voigt S, Rapoport TA, Lingappa VR. TRAM regulates the exposure of nascent secretory proteins to the cytosol during translocation into the endoplasmic reticulum. Cell 1998; 92:621-31.

28. Lichter PR, Richards JE, Boehnke M, Othman M, Cameron BD, Stringham HM, Downs CA, Lewis SB, Boyd BF. Juvenile glaucoma linked to the GLC1A gene on chromosome 1q in a Panamanian family. Am J Ophthalmol 1997; 123:413-6.