Expression Analysis of the Matrix GLA Protein and VE-Cadherin Gene Promoters in the Outflow Pathway

Abstract

purpose. To test the ability of promoter fragments from the matrix Gla protein (MGP) and vascular endothelial-cadherin (VE-cad) genes to target gene expression in a specific manner in the cells of the outflow pathway, by using adenoviral-mediated gene transfer in organ culture.

methods. Perfused anterior segments of human eyes were infected with replication-deficient recombinant adenoviruses expressing the β-galactosidase reporter gene driven by the cytomegalovirus (CMV; control, n = 6), MGP (n = 6), or VE-cad (n = 12) promoters. Forty-eight hours after infection, the anterior segments were fixed and stained for β-galactosidase activity. The distribution of β-galactosidase expression was analyzed in paraffin-embedded sections.

results. The MGP promoter fragment resulted in β-galactosidase expression by the cells of the conventional outflow pathway and did not show any activity in the corneal endothelium or other cells posterior to the scleral spur. Adenovirus containing the VE-cad promoter fragment showed functionality of the promoter in vascular endothelial cells, but failed to produce any detectable expression in the cells of the outflow pathway.

conclusions. Directed expression by the MGP gene promoter specifically to the trabecular meshwork (TM) provides a new tool for specific gene transfer to the outflow pathway. Results with the VE-cad promoter fragment indicate possible differences in the regulation of this gene between vascular and Schlemm’s canal endothelial cells. Taken together, these data demonstrate the feasibility of targeted gene expression to the outflow pathway cells using tissue specific promoters.

The conventional outflow pathway formed by the trabecular meshwork (TM) and Schlemm’s canal (SC) constitutes the major route for aqueous humor outflow from the anterior chamber of the human eye, ¹ and it is believed to contain the site for both normal and pathologic resistance to the outflow of aqueous humor. ^{2 3 4 5} Failure of the outflow pathway to maintain appropriate levels of resistance to aqueous humor outflow results in elevated intraocular pressure (IOP), which is a risk factor for the progression of glaucoma and currently is the primary focus of therapeutic intervention. ⁶  

This central role of the conventional outflow pathway in modulating IOP has brought an increasing interest in developing methods for gene transfer to the cells of the human trabecular meshwork (HTM) and SC. The current approaches include the use of replication-deficient adenoviruses, ^{7 8 9} herpes virus, ¹⁰ and lentiviruses. ¹¹ These methods are useful for experimental purposes to understand the normal and pathophysiology of the tissue ^{12 13 14} and potentially for glaucoma gene therapy treatments. ^{7 8 9 11}  

To date, gene transfer experiments in the TM and SC have been performed using vectors carrying the CMV promoter, which delivers high expression in most mammalian cells. ^{7 8 9 11 12 13 14} However, the use of the CMV promoter results in substantial levels of expression in the corneal endothelium and other areas outside of the outflow pathway. As a result of such nonspecific expression, the biological effects of the transduced genes potentially are not restricted to the outflow pathway, leading to possible undesirable secondary effects in other tissues of the eye. To prevent these undesirable secondary effects, it is important to find a means of targeting the expression of the delivered genes specifically to the cells of interest in the outflow pathway. 

A common approach to accomplish this objective is the use of tissue-specific promoters. ^{15 16 17} Although such genes strictly specific to the outflow pathway have not yet been characterized, the studies of gene expression profile by single-pass sequencing of cDNA clones ^{18 19} have provided some information about those genes that differentiate the trabecular meshwork cells from other cell types. One such gene is the matrix Gla protein (MGP). MGP belongs to the family of vitamin K–dependent, Gla-containing proteins and is expressed at high levels in bone, heart, kidney, and lung. The promoter region for this gene has been characterized by Kirfel et al. ²⁰ Functional analysis of different promoter fragments demonstrated that the proximal promoter region (560 bp) contains the necessary elements to drive expression in in vitro transfection experiments, since no significant functional difference was found with respect to a larger (3600 bp) promoter fragment. ²⁰  

The search for markers of the HTM and SC cells using specific antibodies has also identified genes with promoters potentially useful for specific targeting of gene expression in the outflow pathway. Recently VE-cadherin (VE-cad) has been reported to be a marker of SC cells that is not expressed in TM cells. ²¹ VE-cad mediates cell–cell adhesion between vascular endothelial cells as part of the adherens junction complex. ²² The human VE-cad promoter has not been functionally characterized, but the 5′ region of the gene shares extensive sequence homology with that of the mouse gene, which has been characterized. ^{23 24} This homology extends to the basal transcriptional machinery (nucleotides −1 to −139) and the major specific inhibitory (nucleotides −140 to −289) region that constitute the major functional domains of the mouse promoter. 

In the current study, we analyzed the potential of promoter fragments from the MGP and VE-cad genes, to target gene expression in the cells of the outflow pathway by using recombinant adenoviruses in perfused anterior segments of human cadaveric eyes. 

Methods

Generation of an Expression Vector for β-Galactosidase

To generate a modified pShuttle vector for expression for the β-galactosidase reporter gene under different promoters (pShuttle-LacZ), the LacZ gene with the SV40 polyadenylation signal was released from the pGEM β-Gal control vector (Promega, Madison, WI) by digestion with HindIII and BamHI and cloned between the HindIII and the BglII sites of the pShuttle (Statagene, La Jolla, CA). The simian virus (SV)40 polyadenylation signal was also introduced into the KpnI site to reduce background. The NotI-XhoI sites between the first SV40 polyA and the LacZ gene were used to introduce the promoter fragments to be tested. 

PCR Cloning of Promoter Fragments

A DNA fragment containing the 577 bp of the MGP promoter (550 to +27) was generated by PCR of human genomic DNA using the specific primers: pGLAf, 5′-TTGCGGCCGCATTCAGCCCTACTGGGAAGA-3′; and pGLAr, 5′-TCCTCGAGGTTTCGTCCTGCAGGTCAGT-3′, containing the restriction sites for NotI and XhoI, respectively. 

BLAST analysis (www.ncbi.nlm.nih.gov/blast/ provided in the public domain by the National Center for Biotechnology Information, Bethesda, MD) of the 5′ region from the human VE-cad gene was performed to identify the human genomic sequence homologous to the basal transcriptional machinery and the major specific inhibitory regions of the mouse VE-cad promoter. 

A 445-bp fragment containing this sequence was amplified by PCR with the following specific fragments: pVE-CADf, 5′-TGCGGCCGCTGGGTGGACAAGCACCT-3′; and pVE-CADr 5′-ACTCGAGCTCTGTGGAGCCTGTCA-3′, containing the restriction sites for NotI and XhoI, respectively. PCR reactions were performed in both cases at 94°C, 15 seconds; 68°C, 15 seconds; and 72°C, 60 seconds, for 35 cycles (Advantage-HF PCR kit; BD Biosciences-Clontech, Palo Alto, CA) with 100 ng of human genomic DNA as template. The PCR products were purified in a low-melting-point agarose gel and cloned into the PCR-II TOPO plasmid (Invitrogen, San Diego, CA). Plasmid DNA from selected clones was sequenced using the M13 forward and reverse primers, and sequences were compared with those previously reported to confirm that no mutations were introduced during the PCR amplification. Fragments with the correct sequences were released from the plasmid by digestion with NotI and XhoI, gel purified, and ligated to the same sites as the pShuttle-LacZ plasmid, for generation of recombinant adenoviruses. 

Generation of Recombinant Adenovirus

Replication-deficient recombinant adenoviruses for expression of β-galactosidase under the CMV promoter (AdCMV-LacZ), the MGP promoter fragment (AdMGP-LacZ), or the VE-cad promoter fragment (AdVE-LacZ) were generated with a system (AdEasy; Stratagene) developed by He et al. ²⁵ Briefly, the modified pShuttle vector containing the expression cassette of interest was cotransformed into Escherichia coli (strain BJ5183) with pAdEasy-1 viral DNA plasmid. Transformants were selected with kanamycin and recombinants identified by digestion with PacI. Selected recombinants were used to transform rec(−) Xgold–competent cells (Stratagene, La Jolla, CA) for propagation of the plasmid. The adenoviral plasmids were then linearized with PacI and transfected into HEK 293 cells to be packed into viral particles. High-titer viral stocks were obtained by propagation in 293 cells and purification by CsCl density centrifugation. 

Analysis of Promoter Expression in Primary Cultures of Human TM and SC Cells

Primary cultures of human TM and SC cells were prepared from donor eyes, as previously described. ^{26 27} Cells at passage 3 were infected with 100 plaque-forming units (pfu) of the AdCMV-LacZ, AdMGP-LacZ, or AdVE-LacZ recombinant adenoviruses. Two days after infection, cells were fixed with 1% paraformaldehyde, 0.2% glutaraldehyde, 0.02% NP40, and 0.01% sodium deoxycholate in PBS; washed twice with PBS; and stained for β-galactosidase activity by overnight incubation at 37°C in 1 mg/mL 5-bromo-4-chloro-3 indolyl β-d-galactoside, 5 mM K₃Fe(CN), 5 mM K₄Fe(CN)6–3H₂O, and 2 mM Mg₂Cl in PBS. All protocols involving the use of human tissue were consistent with the tenets of the Declaration of Helsinki. 

Perfusion of Anterior Segments of Human Eyes

Anterior segments of human eyes were cultured as described by Johnson and Tschumper ²⁸ with some modifications. Human cadaveric eyes, aged between 33 and 74 years, were obtained less than 48 hours after death. After bisection at the equator, the lens, iris, and vitreous were removed. The anterior segments were then clamped to a modified Petri dish and perfused with serum-free DMEM at a constant flow of 3 μL per minute, using a microinfusion pump. The medium was supplemented with 110 mg/L sodium pyruvate, 100 U/mL penicillin, 0.1 mg/mL streptomycin, 170 μg/mL gentamicin, and 250 μg/mL amphotericin. Anterior segments were incubated at 37°C in 5% CO₂. IOPs were continuously monitored with a pressure transducer connected to the dish’s second cannula and recorded with an automated computerized system. Only anterior segments with stable outflow facilities between 0.09 and 0.40 μL/min per mm Hg that remained unchanged after viral infection were used. Six eyes from different donors were analyzed for the expression of the CMV and MGP promoters, and 12 eyes, also from different donors, were analyzed for the VE-promoter. 

Transduction of Perfused Anterior Segments

After 48 hours of perfusion, pumps were stopped, and the pressure dropped to less than 5 mm Hg. The segments were inoculated with 10⁷ pfu in 100 μL of perfusion medium at 3 μL/min. Once the total volume was inoculated, the normal perfusion regimen was resumed. 

Histochemical Analysis of β-Galactosidase Expression

For detection of β-galactosidase, anterior segments were fixed by perfusion with 1% paraformaldehyde, 0.2% glutaraldehyde, 0.02% NP40, and 0.01% sodium deoxycholate in PBS at 15 mm Hg. β-Galactosidase activity was detected by overnight incubation at 37°C in 1 mg/mL 5-bromo-4-chloro-3 indolyl β-d-galactoside, 5 mM K₃Fe(CN), 5 mM K₄Fe(CN)6–3H₂O, and 2 mM Mg₂Cl in PBS. After color development, the segments were fixed in 10% neutral buffered formalin, dehydrated in an ethanol and xylene series, and embedded in paraffin for cutting of 5- to 6-μm-thick sections with a microtome. Sections were counterstained with either hematoxylin QS (Vector Laboratories, Burlingame, CA) or 4′,6′-diamino-2-phenylindole (DAPI). 

Immunofluorescence in Histologic Sections

Tissue pieces were fixed in 10% neutral buffered formalin, dehydrated, and processed for paraffin-embedded sectioning or were placed in optimal cutting temperature (OCT) compound, frozen, and processed by cryosectioning. Paraffin-embedded sections were rehydrated and incubated with the appropriate concentration of a β-galactosidase–specific antibody for 30 to 60 minutes, rinsed extensively with Dulbecco’s phosphate-buffered saline (DPBS), and incubated with the fluorescence-labeled secondary antibody (Bio Source International, Camarillo, CA) under the same conditions as for the primary antibody. After three washes with DPBS, the cells are mounted with antifade medium for observation by fluorescence microscope (Axioplan; Carl Zeiss Meditec, Dublin, CA). Cryosections on slides were fixed in 50% methanol/50% acetone and air-dried. Fixed sections of tissue were rehydrated in PBS, blocked with 10% goat serum in PBS containing 0.1% Triton X-100, and incubated overnight with monoclonal IgG (1:1000 dilution, clone 9H7) or anti-β-galactosidase (1:1000, clone GAL-13). Goat serum (10%, Sigma-Aldrich, St. Louis, MO) and Triton X-100 (0.1%, Sigma-Aldrich) were included in incubations to inhibit nonspecific binding of antibodies to tissues. After antibody incubations, tissue sections were washed extensively (four times in 4 mL for 15 minutes each) in phosphate-buffered saline containing 0.1% Triton X-100. Specific binding of antibodies to receptor was detected with CY3-conjugated goat anti-mouse immunoglobulin G (IgG) at a 1:1000 dilution (Jackson ImmunoResearch Laboratories, West Grove, PA). Tissue sections were incubated with secondary antibodies for 2 hours and washed extensively before viewing. Background fluorescence was indicated in tissues processed in the absence of primary antibodies. Labeled tissue sections were visualized and photographed digitally with an inverted fluorescence microscope (model IX70; Olympus, Melville, NY) with a digital camera (Magnifire). 

Results

Expression Analysis of the CMV Promoter

Both TM- and SC-cultured cells infected with the recombinant adenovirus AdCMV-LacZ, in which the LacZ reporter gene is controlled by the CMV promoter, showed strong β-galactosidase activity 1 . Injection of anterior segments with this virus resulted in transgene expression by the cells of the anterior segments similar to that previously reported ^{7 8} 2 . However, some of these eyes showed gaps (discontinuous staining segments) with low or absent β-galactosidase activity segments in the TM that could be indicative of nonfunctional areas of the TM (2 , bottom). No correlation was found between the presence of these gaps and age of the donor, postmortem time, or outflow facility, and no visible histologic differences were observed between these areas and those with intense β-galactosidase staining. In addition, expression in the endothelium from intrascleral veins was visible in two of the six eyes tested for this promoter. 

Expression Analysis of the MGP Promoter Fragment

Infection of human TM and SC primary cell cultures with the AdMGP-LacZ, in which the expression of β-galactosidase was controlled by the MGP promoter fragment, produced intense β-galactosidase staining in TM cells, whereas SC cells did not show any detectable expression of the LacZ gene 1 . In noteworthy contrast to the generalized β-galactosidase staining in eyes transduced with the AdCMV-LacZ virus, all the human anterior segments transduced with the AdMGP-LacZ showed LacZ expression restricted to the TM and SC regions 2 . The specific targeting of the MGP promoter to the outflow pathway cells was confirmed in histologic sections 3 . Analysis of these sections showed a specific and homogeneous distribution of LacZ expression within the cells of the conventional outflow pathway, including the trabecular, juxtacanalicular, and the inner and outer wall SC cells 3 . In contrast to the results obtained with the CMV promoter, β-galactosidase staining did not extend beyond the scleral spur, which delimits the posterior boundary of the TM 3 . No detectable β-galactosidase activity was observed in the cells from the fragments of ciliary body, ciliary muscle, and iris that were still present in the perfused tissue. β-Galactosidase expression did not extend to the anterior nonfiltering portion of the corneoscleral meshwork near the operculum 3 or the corneal endothelium 3 . 

Expression Analysis of the VE-Cad Promoter

The AdVE-LacZ with the VE-cad promoter fragment controlling the LacZ gene failed to show any β-galactosidase activity in either SC or TM cells in both primary cultures 1 and perfused anterior segments 4 . Experiments using anti-β-galactosidase IgG, an alternative method of detecting β-galactosidase protein, also failed to show any level of expression in these cells (data not shown). 

As a control, frozen sections of the perfused eyes infected with the Ad-VE-Cad-β-gal were tested for the presence of the VE-cad protein with specific anti-VE-cad antibodies. Samples from perfused eyes showed similar levels of VE-cad expression (data not shown) in the SC cells as in the paired control and those previously described in nonperfused eyes, indicating that there is not any apparent loss of expression of this protein during perfusion (data not shown). 

However, infection with this virus resulted in expression of the reported gene in the vascular endothelial cells of the intrascleral veins in 3 of the 12 perfused eyes 4 . This result provided evidence that this promoter fragment was functional in the aqueous-humor–draining vascular endothelial cells of the intrascleral veins, and also indicated that, at least in some eyes, the viral particles may be able to reach the blood stream when used at a dose of 10⁷ pfu. To determine whether this result could be related to some possible damage to the TM during the dissection and perfusion, the tissues were subject to careful examination through the stereomicroscope and histologic sections. No evidence of TM damage or alterations was found. The outflow facilities of these three eyes were not higher than those of other eyes where expression in the intrascleral vessels was not detected. 

Discussion

Strategies focused on gene delivery both for experimental purposes and possible gene therapy for glaucoma can benefit from specific targeting of gene expression to cells in the outflow pathway, avoiding expression in other cells of the anterior segment that can complicate experimental interpretation or result in undesirable secondary effects. 

The identification of promoters capable of providing such cell-type–specific expression is therefore an important objective that will help in future efforts to manipulate genetically the outflow pathway cells both for experimental and therapeutic purposes. Because gene transfer with viral vectors results in infection of most cell types in the anterior chamber, the use of these promoters would result in expression of the specific gene in the selected cells of interest, whereas other anterior segment cells, although infected with the vector, would demonstrate little or no biological effects. 

The specific targeting of gene expression obtained with the MGP promoter fragment in our human organ culture makes this promoter a potentially important tool to accomplish specific gene delivery to the HTM cells. In addition, primary cultures of HTM cells retained expression of this promoter, indicating that it can be useful as a marker for this cell type in vitro. 

The presence of gaps (discontinuous staining segments), with low or no expression of the reporter gene in the TM of some eyes observed with the MGP and CMV promoters, could reflect functional differences between different areas of the TM. Segmental differences in the TM including the pigment distribution have been described in both normal and glaucomatous eyes and may correlate with the area of empty space beneath SC. ^{29 30} The pathophysiological relevance of these observations is not totally clear, but may indicate the possibility that areas of the TM may lose functionality over time. 

It is also interesting that the MGP promoter was not functional in the clusters of cells located behind the operculum within the corneoscleral meshwork. This area is considered the nonfiltering portion of the TM because of the comparatively slow movement of aqueous humor tracers through this area. ³¹ However, we believe that the lack of MGP promoter expression did not result from the failure of the adenoviral vector to reach and infect these cells, because adenoviruses with the CMV promoter used at the same multiplicity of infection generated good transgene expression in this area. This anterior portion of the TM is known to contain the Schwalbe line’s cells (SLCs), which have morphologic and ultrastructural features that define them as a distinct subpopulation from other trabecular cells. ^{32 33 34} Although their role is not known, a decline in the number of SLCs and changes in their morphology have been found to be associated with glaucoma in dogs, ³⁵ and some observations indicate that they may function as TM progenitor cells. ³⁶ The difference in MGP expression between the SLCs and other trabecular cells reported herein could be useful as a marker for TM cell differentiation. Such a marker would be particularly helpful to investigate the dynamics of cell death and regeneration in the TM and define the potential role of SLC as progenitor cells. 

High activity of the MGP promoter fragment in the TM but not in other eye tissues is consistent with previous gene expression profiling data of the HTM. ^{18 37} This gene is expressed at high levels in certain tissues like cartilage, kidney, heart, and lung, but its expression in most tissues is very low. Although there are some conflicting reports about the levels of expression of MGP in the corneal endothelium, ^{38 39 40} gene expression profile studies indicate that MGP is not highly expressed in other eye tissues. ^{18 41 42 43 44 45 46 47 48} Based on the results presented herein, if MGP is expressed in the corneal endothelium at high levels, we would hypothesize that its transcription would require other regulatory elements different from those present in the selected promoter fragment. 

It is not known why MGP might be highly expressed in HTM cells. One possible role for MGP in the outflow pathway could be the maintenance of the elastic and contractile properties of the tissue by preventing calcification of the extracellular matrix. The protective role of MGP against calcification in soft tissues is supported by two main observations: first, mutations in this gene are associated with Keutel syndrome, an autosomal recessive disorder characterized by abnormal cartilage calcification, peripheral pulmonary stenosis, and midfacial hypoplasia ⁴⁹ ; and second, transgenic mice with a disrupted MGP allele generated by gene targeting in embryonic stem cells died within 2 months as a result of arterial calcification, which led to blood vessel rupture, inappropriate calcification of various cartilages, osteopenia, and fractures. ⁴⁹ MGP is also known to play a role in cell differentiation. In the arterial wall of the MGP knockout mouse, medial smooth muscle cells are replaced by chondrocyte-like cells undergoing endochondral ossification, and in the growth plate of growing bones, hypertrophic chondrocytes are absent. These effects appear to be mediated, at least in part, by regulation of the bone morphogenetic protein (BMP)-2. ⁵⁰ BMPs, including BMP-2, have been reported in the trabecular meshwork ⁵¹ and may be essential for its normal development, since the ocular phenotypes of the haploinsufficient Bmp4 mice include anterior segment dysgenesis with elevated IOP. ⁵² In addition, in light of the role played by glucocorticoids in glaucoma, it deserves mentioning that MGP has recently been shown to be induced by glucocorticoids in lung cells. ⁵³  

The human promoter fragment from the VE-cad gene showed functionality in vascular endothelial cells but not in SC cells. The lack of expression in cultured cells was expected, because the loss of VE-cad expression in cultured cells has been previously observed. ²¹ However, the lack of expression in SC cells from perfused anterior segments was more surprising. One explanation for these results could be that, given the differences between the physiology of the vascular and the SC endothelia, regulation of VE-cad expression may involve different elements in these cell types. Such alternative regulatory mechanisms in different tissues are not uncommon ^{54 55} and can result, in some cases, in a higher sensitivity of specific promoter fragments to the regulation of their tissue specificity than the endogenous promoter. One relevant example is the regulation of the von Willebrand factor transcription. In vivo analysis of the von Willebrand factor promoter showed that whereas the expression of a proximal promoter fragment was restricted to the endothelial cells of the brain, ⁵⁶ a larger fragment was functional in the brain, heart, and skeletal muscle endothelia. ⁵⁷ These observations indicate possible regional differences in endothelial regulatory factors, and the existence of cell subtype-specific cis-acting DNA domains within the endothelial cells. Such differences would be expected to be more extensive for nonvascular endothelial cells such as those of the SC. 

Another issue raised by the results with both the VE-cad and CMV promoters is the variability with which the adenovirus injected at 10⁷ pfu was able to reach the collector channels and intrascleral veins in one of the eyes. In 3 of the 12 eyes inoculated with the Ad VE-CAD-LacZ and 2 of the 6 eyes inoculated with the AdCMV-LacZ, there was expression in the veins draining the aqueous humor outside of the eye. 

Because perfusion experiments are performed with cadaveric eyes, this level of variability could result from differences in the levels of preservation and integrity of the TM at the time of the experiment. However, it is also important to consider the possibility that naturally occurring individual differences among human eyes might influence the probability of viral particles from the anterior chamber ultimately draining into the blood stream. Regardless, the potential for extraocular spread of viral particles should be remembered in contemplating future gene therapy efforts. 

In summary, the results obtained with the MGP and VE-cad promoter fragments demonstrate the advantages and pitfalls of using tissue-specific promoters to target expression in gene transfer experiments. Although the promoter fragment from VE-cad failed to target expression to the SC endothelium, the results obtained with the MGP promoter fragment indicate that this could be a particularly useful tool to target high levels of gene expression in the TM without affecting other cells in the anterior chamber. Testing of additional promoters and advances in our knowledge about the mechanisms regulating gene expression in the cells of the outflow pathway should provide new tools for targeting expression to specific cells within the outflow pathway that may be useful for new experimental and gene therapy approaches in glaucoma. 

 Supported by National Eye Institute Grants EY01894-25 and EY13861 and Research to Prevent Blindness.

 Submitted for publication May 28, 2003; revised October 22, 2003; accepted November 18, 2003.

 Disclosure: P. Gonzalez, None; M.Caballero, None; P.B. Liton, None; W.D. Stamer, None; D.L. Epstein, 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.

 

Expression of the MGP promoter fragment in human TM and SC cultured cells. Primary cultures from HTM and SC cells infected with the adenovirus AdMGP-LacZ in which the expression of β-galactosidase is controlled by an MGP promoter fragment resulted in intense β-galactosidase staining in all TM cells (left) but no detectable expression in SC cells (right).

Figure 2.

 

β-Galactosidase expression in human anterior segments transduced with recombinant adenoviruses expressing the LacZ gene under the CMV (left) and the MGP (right) promoters. Top: comparison at low magnification between the eyes of the same donor transduced with 10⁷ pfu of the recombinant adenovirus AdCMV-LacZ in which expression of the β-galactosidase reporter gene is driven by the CVM promoter (left) and the adenovirus AdMGP-LacZ in which the expression of β-galactosidase is controlled by an MGP promoter fragment (right). Although the use of the CMV promoter resulted in generalized expression of the reporter gene in the different cell types of the anterior chamber, the MGP promoter fragment targeted the expression more toward the TM. Bottom: higher magnification of a different pair of eyes, transduced in the same way as those in the top panels. In some cases there were gaps (discontinuous staining segments) in the expression of the reporter gene in the TM of the eyes infected either with the AdCMV-LacZ or the AdMGP-LacZ adenovirus that may correspond to areas of the TM with reduced or nonfunctional outflow. Similar results were obtained in six pairs of eyes.

Figure 3.

 

Histochemical analysis of β-galactosidase expression in human anterior segments transduced with recombinant adenoviruses expressing the LacZ gene under the CMV and the MGP promoters. (A) Low-magnification view of paraffin-embedded sections, with the TM region of eyes transduced with the AdCMV-LacZ (left) and the AdMGP-LacZ (right) showing the targeted expression of the MGP promoter fragment to the TM in comparison to the more generalized expression with the CMV promoter. (B) Higher magnifications showing β-galactosidase expression in the TM cells obtained with both CMV (left) and MGP (right) promoters. While expression with the CMV promoter extended beyond the scleral spur, that of the MGP promoter fragment appeared to be restricted to the cells of the outflow pathway. (C) Cells of the anterior corneoscleral meshwork were infected by the adenoviral vectors, as demonstrated by β-galactosidase expression obtained with the CMV promoter (left) and by the fact that the MGP promoter fragment did not appear to be active in these cells (right). (D) Expression of β-galactosidase in the corneal endothelial cells transduced with AdCMV-LacZ and the lack of detectable expression in those transduced with the AdMGP-LacZ (right). Sections were counterstained with DAPI to visualize the nuclei of the endothelial cells. Images show views under fluorescence (top), mixed fluorescence and light transmission (middle), and light transmission alone (bottom) of the same fields. These results are representative of six pairs of eyes.

Figure 4.

 

Histochemical analysis of the β-galactosidase expression in human anterior segments transduced with recombinant adenovirus AdVE-LacZ expressing the LacZ gene under the VE-cad promoter fragment. The low-magnification view of a paraffin-embedded section. Left: TM region showing the lack of β-galactosidase staining in SC and TM cells and positive staining of the endothelium of an intrascleral vein after transduction with the AdVE-LacZ. Right: staining in the endothelium of the same vein at higher magnification. Similar expression of the reporter gene in intrascleral veins was observed in 3 of the 12 eyes transduced with AdVE-LacZ.

 

The authors thank Ronald Heimark for his generous gift of antibodies against vascular endothelial cadherin.