Thrombospondins (TSPs) are a family of extracellular matrix glycoproteins (five members) whose expression is temporally and spatially regulated during embryonic development. TSP1 and TSP2 are very similar in their domain structure, form trimers, and inhibit angiogenesis (Sheibani and Frazier,1999; Lawler,2000). TSP3-5 lack the angioinhibitory domains and form pentamers. Mice deficient in TSP1 and/or TSP2 are viable and breed, but they exhibit increased vascular density in many tissues with defects in wound healing, bleeding time, and abnormalities in lung and connective tissues (Lawler et al.,1998; Kyriakides et al.,1998; Agah et al.,2002). Studies of mice that lack TSP1 and TSP2 demonstrate that these proteins have nonoverlapping functions during wound healing and, perhaps, other angiogenesis-dependent processes (Agah et al.,2002).
TSP1 is expressed by a variety of cell types including endothelial cells (EC), smooth muscle cells/pericytes, and astrocytes. TSP1 inhibits EC proliferation and migration in vitro and inhibits angiogenesis in vivo (Sheibani and Frazier,1999). TSP1 induces apoptosis of EC through its interaction with its angioinhibitory receptor (CD36), and by Jun-N-terminal kinase (JNK) and caspase activation (Jimenez et al.,2000,2001; Nor et al.,2000). However, the physiological role of TSP1 during vascular development and angiogenesis requires further delineation.
The developing mouse retinal vasculature provides a unique opportunity to study all aspects of vascular development postnatally, and it is readily amenable to biochemical and histological manipulations. The retinal vasculature normally does not extend into the vitreous or beyond the inner nuclear layer. This highly restricted pattern of retinal vascularization has interested many investigators and led to the hypothesis that there are factors present at these sites that limit the growth of blood vessels. We have shown that TSP1 is present in vitreous and aqueous humor, and lack of expression at these sites correlates with early vasculopathies associated with diabetes (Sheibani et al.,2000). Therefore, TSP1 expression may play an important role in retinal vascular homeostasis and, potentially, in the establishment and maintenance of its restricted organization pattern.
Our recent studies with TSP1-deficient (TSP1−/−) mice indicate that TSP1 expression is essential for appropriate vascular pruning and remodeling. This is an apoptosis-dependent process, which occurs during later stages of retinal vascular development (Wang et al.,2003). These mice are also less sensitive to oxygen-induced vessel obliteration during oxygen-induced ischemic retinopathy (OIR). In addition, the regression of hyaloid (ocular embryonic) vessels, which occurs through apoptosis, is also delayed in the TSP1−/− mice. Furthermore, retinal EC prepared from TSP1−/− mice are more proliferative and migratory compared to wild type cells (Su et al.,2003; Wang et al.,2006). Thus, TSP1 expression appears to be essential during later stages of vascular development for promoting the quiescent, differentiated state of endothelium as well as eliminating excessive blood vessels by inducing EC apoptosis.
In the studies presented here, we show that over-expression of TSP1 in the eye affects normal retinal vascularization and limits retinal neovascularization during OIR without a dramatic effect on ocular development. Therefore, appropriate expression of TSP1 may be essential for normal development of retinal vasculature, and modulation of TSP1 expression may provide an effective mechanism to regulate ocular angiogenesis.
Generation of Transgenic Mice That Express TSP1 Specifically in the Lens
TSP1 is normally present in the vitreous and aqueous humor of human, mouse, and rat eyes (Sheibani et al.,2000). In the studies described here we used lens-specific expression of TSP1 to produce TSP1 in the vitreous and/or aqueous humor. Transgenic mice were generated with an αA-crystalline promoter driving the expression of TSP1 (Fig. 1A). This allowed us to determine if early expression of TSP1 (a secreted protein) in the eye affects retinal vascular development. Genes linked to the αA-crystalline promoter are expressed prior to or concurrent with the onset of terminal differentiation of lens fiber cells, which is at or before 12.5 days of embryonic development (Stolen and Griep,2000; Ash and Overbeek,2000). This is well before the initiation of retinal vascularization, which occurs postnatally. The αA-crystalline promoter has been successfully used previously to drive expression of other secreted proteins, such as FGF-3, yielding proteins in the aqueous and vitreous humors (Robinson et al.,1998).
Transgenic mice with TSP1 expression driven by an αA-crystalline promoter were identified by PCR. The positive founders were then confirmed by Southern blot analysis, using transgene-specific primers and probes (see Experimental Procedures section). Six independent lines of transgenic animals were produced (OVE 704, 711, 717, 719, 723, and 729). The OVE 704 and 719 lines had high copy numbers of the TSP1 transgene, while the 711 and 729 lines had moderate copy numbers of the transgene. The OVE 717 and 723 lines had low copy numbers of the transgene (Fig. 1B). Transgenic lines with moderate and low copy numbers of the transgene (711, 717, 729, and 723) had very low TSP1 protein expression (Fig. 1C) within the eye and did not show a phenotype (see below). Since only the progeny of founders with high copy numbers (OVE 704 and 719) produced significant amounts of TSP1 protein within the eye (Fig. 1C) and exhibited a retinal vascular phenotype, these transgenic lines were chosen for further analysis.
Retinal Vascularization in TSP1 Transgenic Mice
In mice, development of the retinal vasculature initiates right after birth (postnatal day zero, P0) and is completed by three weeks of age (P21). The superficial vascular bed develops during the first week of life, and the intermediate and deep capillary plexuses are formed during the second and third weeks of life (Michaelson et al.,1954; Fruttiger,2002; Dorrel et al.,2002). The retinal vasculature, however, continues to undergo pruning and remodeling, which is completed by six weeks of age (Wang et al.,2003). To determine the consequences of TSP1 expression on postnatal retinal vascularization, we initially evaluated the vascular densities by trypsin digest of wholemount preparations of retinas from wild type and TSP1 transgenic mice at P21. We observed a decrease in retinal vascular density, which was inversely proportional to the level of transgene expression (Fig. 2). The largest decrease was observed in transgenic mice expressing the highest levels of TSP1 within their eyes (Fig. 2B). The quantitative assessments of endothelial cell/pericyte ratios and endothelial and pericyte densities demonstrated a significant decrease in the number of EC and pericytes without affecting their ratio (Table 1).
Table 1. Endothelial and Pericytes Ratios and Their Numbers Per 100 μm2 in P21 Wild Type and TSP1-Transgenic (OVE 704) Mice (Mean ± SD)a
The number of retinas (mice) counted is given in parentheses.
PECAM-1 staining of wholemount retinas and collagen IV staining of frozen eye sections were then used to visualize retinal vasculature in TSP1 transgenic and wild type mice at P4, P10, and P21. In P4 wild type mice, the retinal vasculature was organized uniformly and was evenly spaced over the superficial retina (Fig. 3A,C). In contrast, retinas from P4 TSP1 transgenic mice demonstrated a dramatic decrease in the density of superficial retinal vessels (Fig. 3B,D). Furthermore, the retinal vasculature appeared nonuniform, abnormal in size and distribution, and immature. Similar abnormalities in retinal vasculature were observed in frozen sections prepared from P4 transgenic mice stained with antibody to collagen IV (not shown). The quantitative assessment of vascular density was performed by comparing pixel intensities in z-series of wholemount retinas as described previously (Wang et al.,2003). We observed a significant decrease in vascular density of retinas from TSP1 transgenic mice compared to wild type mice (Fig. 3E; P < 0.01).
Retinas from P10 wild type mice demonstrated a dense superficial vascular bed, which nearly attained its mature radial dimension extending to the peripheral edge of the retina (Fig. 4A,C). Sprouts forming the deep vascular bed were also observed (Fig. 4E; arrowhead). In contrast, in P10 TSP1 transgenic mice, a substantial part of the peripheral retina was still avascular (Fig. 4B,D). The vascular bed appeared not only low in density but also as a meshwork of not fully developed, immature polygons. Collagen IV staining of the frozen sections prepared from P10 mice showed that no deep vascular plexuses had formed in retinas from TSP1 transgenic mice compared to wild type mice (Fig. 4E,F). The quantitative assessment of vascular densities in wholemount retinas, performed as described above, indicated a significant decrease in the vascular density of retinas from TSP1 transgenic mice compared to wild type mice (Fig. 4G; P < 0.01).
Wholemount PECAM-1 staining of the retinal vasculature of P21 mice also demonstrated a low vascular density in the superficial and the deep vascular beds of TSP1 transgenic mice compared to wild type mice (data not shown; Fig. 2D). Similar abnormalities in retinal vasculature were also observed in frozen sections prepared from P21 transgenic mice stained with antibody to collagen IV (data not shown). Retinal vascular density of P21 TSP1 transgenic mice determined as described above was approximately half that observed in the wild type mice, and was consistent with the results obtained by trypsin digests (data not shown; Fig. 2 and Table 1). Thus, lens-specific over-expression of TSP1 impacts all phases of retinal vascularization.
Increased Apoptosis in Retinal Vasculature of TSP1 Transgenic Mice
TSP1 inhibits angiogenesis by promoting apoptosis of EC both in vitro and in vivo (Jimenez et al.,2000; Nor et al., 2001). We next employed TdT-dUTP Terminal Nick-End Labeling (TUNEL) and anti-active caspase-3 staining to determine the rate of apoptosis in the developing retinal vasculature in P10 wild type and TSP1 transgenic mice. B4-lectin staining of retinal wholemounts was used to identify endothelial cells.
TUNEL staining of frozen retina sections from wild type and TSP1 transgenic mice is shown in Figure 5A and B, respectively. TUNEL-positive cells (arrows) were more abundant in sections prepared from TSP1 transgenic mice compared to wild type mice. The active caspase-3 wholemount staining of retinas from wild type and TSP1 transgenic mice is shown in Figure 5C and D. Arrows point to positive staining cells in the retinal vasculature. We consistently observed a 2.5-fold increase (P < 0.01) in the number of apoptotic nuclei in the retinal vasculature of the P10 TSP1 transgenic mice (average apoptotic cell density/μm2: 24.4 ± 4.5) compared to wild type mice (average apoptotic cell density/μm2: 9.7 ± 2.4; Fig. 5E). Examination of retinal sections demonstrated that the majority of the active-caspase-3 positive cells were associated with the vasculature (data not shown). Thus, increased expression of TSP1 in the lens was associated with increased numbers of apoptotic retinal EC.
Increased Proliferation in the Retinal Vasculature of TSP1 Transgenic Mice
Proliferation of retinal vascular EC in P10 TSP1 transgenic and wild type mice was determined by 5-bromo-2′-deoxyuridine-5′-monophosphate (BrdU) labeling followed by collagen IV staining of wholemount retinas. Figure 6 shows BrdU-positive cells in the retinal vasculature of wild type (Fig. 6A) and TSP1 transgenic (Fig. 6B) mice. We consistently observed a 1.5-fold increase (P < 0.01) in the number of proliferative nuclei observed in the retinal vasculature of the transgenic mice (average proliferative cell density/mm2:807 ± 132) compared to wild type mice (average proliferative cell density/mm2: 535 ± 93; Fig. 6C).
Increased Vessel Obliteration and Reduced Preretinal Neovascularization in TSP1 Transgenic Mice During Oxygen-Induced Ischemic Retinopathy
We next determined the response of TSP1 transgenic mice during oxygen-induced ischemic retinopathy. In this assay, 7-day-old mice are exposed to 75% oxygen for 5 days. Exposure to high oxygen negates the increase in VEGF expression that is required for normal vascularization and leads to underdeveloped, obliterated retinal vasculature (Smith et al.,1994; Alon et al.,1995; Pierce et al.,1995; Stone et al.,1996). Hyperoxia prevents new vessel growth and obliterates existing vessels by apoptosis (Alon et al.,1995; Pierce et al.,1996). The mice are then exposed to room air (20% oxygen) for 5 days. The relatively hypoxic room environment stimulates the production of VEGF, resulting in extensive growth of new vessels, which enter the vitreous and deep layers of the retina.
Retinas from P7 mice normally show a superficial layer of vessels that forms during the first week of life. Exposure to high oxygen prevents further development of retinal vessels and obliterates vessels in the central area of the retina. Figure 7A and B shows wholemount PECAM-1 staining of P12 retinal vasculature, while Figure 7C and D shows hematoxylin- and Periodic Acid-Schiff (PAS)-stained eye cross sections prepared from P17 wild type and TSP1 transgenic mice, respectively. TSP1 transgenic mice exhibited a more severe response during the hyperoxia phase resulting in significantly increased areas of nonperfusion compared to wild type mice (∼ 2-fold, P < 0.01). Figure 7E shows the quantitative measurements of nonperfused areas relative to the whole retina areas in P12 wild type and TSP1 transgenic mice. These studies demonstrate that over-expression of TSP1 in the lens promotes obliteration of retinal vessels during hyperoxia resulting in a larger nonperfused area.
Quantification of preretinal neovascularization of P17 mice was performed as recently described by us (Wang et al.,2003,2005). Figure 7F shows the mean number of vascular cell nuclei projecting into the vitreous (arrows in Fig. 7C and D) of eyes from wild type and TSP1 transgenic mice. There was a significant decrease (P < 0.01) in the number of vascular cell nuclei detected in the P17 TSP1 transgenic mice compared to the wild type mice.
Hyaloid Vasculature in TSP1 Transgenic Mice
The pupillary membrane and hyaloid vessels (hyaloid arteries, tunica vasculosa lentis, and vasa hyaloidea propria) provide nourishment to the immature lens, retina, and vitreous (Ito and Yoshioka,1999). However, hyaloid vessels regress during the later stages of ocular development and are completely resolved by 3–4 weeks of age. The status of hyaloid vasculature was determined by PECAM-1 staining of ocular specimens prepared from wild type and TSP1 transgenic mice. PECAM-1 stating of the ocular specimens prepared at P7, P14, and P21 showed a pattern of hyaloid vessels (tunica vasculosa lentis) that were more numerous in TSP1 transgenic mice (Fig. 8B,D,F) compared to wild type mice (Fig. 8A,C,E). We consistently observed an increase in the number of radiating blood vessels across the lens prepared from TSP1 transgenic mice. These differences became less pronounced as the vessels regressed. By six weeks of age, all the hyaloid vessels regressed in both wild type and TSP1 transgenic mice (data not shown). Therefore, early expression of TSP1 in the lens may impact the development of hyaloid vasculature but has minimal effect on its regression.
Histological Analysis of TSP1 Transgenic Eyes and ERG Analysis
To determine whether ocular development proceeded normally in TSP1 transgenic mice, histological examination of transgenic and wild type eyes was performed by hematoxylin and PAS staining of eye sections. Examination of eye sections prepared from all the transgenic lines indicated normal development without any adverse effects on lens and/or retinal development (Fig. 9A). This is consistent with slit-lamp analysis, which revealed no significant phenotype in the TSP1 transgenic mice compared to wild type mice.
We next examined retinal function by electroretinogram (ERG) analysis of wild type and TSP1 transgenic mice. Figure 9B shows intensity vs. response in the growth of ERG A- and B-wave with increased flash intensity in wild type (Fig. 9B, left) and TSP1 transgenic (Fig. 9B, right) mice. The prominent oscillatory potentials riding on the B-wave indicate normal retinal function in these mice. The numbers in the right of the curve are the neutral density filter values in decibels that were used to attenuate the standard flash intensity (1.4 Candela-Sec/m2). Thus, lens-specific over-expression of TSP1 has minimal effects on ocular development and the formation and function of the retina.
TSP1 is one of the first endogenous inhibitors of angiogenesis identified whose down-regulation is associated with an angiogenic switch during tumor progression (Rastinejad et al.,1989). It induces apoptosis of EC in vitro and in vivo through down-regulation of the cell death repressor bcl-2 expression and activation of caspases and the JNK MAP kinase pathway (Jimenez et al.,2000,2001; Nor et al.,2000). However, the role TSP1 plays during vascular development and angiogenesis requires further delineation. Here we show that the ectopic expression of TSP1 in the lens, prior to the development of retinal vasculature, severely compromises the development of retinal vasculature and prevents retinal neovascularization during OIR. Mice that over-express TSP1 in their lens have a significant decrease in retinal vascular density, and the retinal blood vessels appear disorganized and less mature. The transgenic mice are also more sensitive to hyperoxia-mediated vessel obliteration and, most importantly, fail to neovascularize their retina during OIR. These effects were mainly attributed to a reduced number of EC due to increased apoptosis in retinal vasculature of TSP1 transgenic mice. In contrast, the hyaloid vasculature was more prominent in TSP1 transgenic mice, especially at early time points. However, their regression was minimally affected by increased expression of TSP1.
The development of mouse retinal vasculature occurs in two main phases. In the first phase (P0–P7), vessel development occurs radially from the optic nerve to the ora serrata within the most superficial layer of the retina. In the second phase (P7–P21), the vessels start to sprout downward, into the inner plexiform layer, where they establish the deep vascular plexuses parallel to the superficial layer (Wang et al.,2003). It is well accepted that the establishment of the deep vascular plexuses in the inner plexiform layer occurs by angiogenesis (Cuthberson et al.,1986; Connolly et al.,1988; Dorrel et al.,2002; Fruttiger,2002; Wang et al.,2003). However, it remains controversial as to whether primary vascular development across the inner surface of the retina occurs by vasculogenesis or angiogenesis (McLeod et al.,1987; Hughes et al.,2000; Chang-Ling et al.,1990,2004). Recent evidence suggests that the primary vascular development across the inner surface of the retina, at least in mice, also occurs by angiogenesis (Fruttiger,2002; Wang et al.,2003).
In our study, over-expression of TSP1 disrupted the development of both the primary and the deep retinal vascular plexuses. We observed a significant increase in apoptosis of retinal EC in TSP1 transgenic mice. This increased apoptosis was compensated, at least in part, by increased cell proliferation in these mice. We recently showed TSP1 expression is essential for appropriate development of retinal vasculature. In its absence, retinal vasculature fails to undergo appropriate pruning and remodeling during later stages of retinal vascularization resulting in increased retinal vascular density (Wang et al.,2003). Therefore, inappropriate expression of TSP1 prior to initiation of retinal vascularization halts the progressive development of retinal vasculature by inducing excessive vascular pruning and remodeling, which results in reduced vascular density.
Our retinal trypsin-digests indicated that a decrease in the number of both EC and pericytes contributes to decreased retinal vascular density in TSP1 transgenic mice (Fig. 2 and Table 1). What is not clear is whether the reduced number of pericytes is secondary to the reduced number of EC and/or is the result of the direct effects of TSP1 on these cells. In smooth muscle cells/pericytes, TSP1 acts as an autocrine factor and promotes their proliferation, especially in the presence of platelet-derived growth factor (PDGF) (Majack et al.,1988). We recently showed that the lack of TSP1 does not affect the number of pericytes in the developing retinal vasculature of TSP1−/− mice compared to wild type mice (Wang et al.,2003). Therefore, proliferation of smooth muscle cells/pericytes is not dependent on TSP1 expression in vivo. Consistent with our results, Isenberg et al. (2005) recently showed that endogenous TSP1 is not necessary for proliferation but is permissive for smooth muscle cell migratory and proliferative responses to PDGF in culture. Smooth muscle cells/pericytes express many of the TSP1 receptors expressed on the EC including CD36, CD47, and αvβ3 integrin. However, the effects of TSP1 on smooth muscle cells/pericytes through these receptors are distinct from those on EC. Furthermore, TSP1 can affect both cell types positively and negatively depending on the levels of TSP1 and the combination of the receptors involved (Isenberg et al.,2005). Therefore, over-expression of TSP1 may inhibit proliferation and migration of both cell types, contributing to the antiangiogenic activity of TSP1 and the reduced vascular density observed here.
We next examined the effects of TSP1 over-expression on retinal neovascularization during OIR. The P7 wild type and TSP1 transgenic mice were exposed to 75% oxygen for 5 days. Hyperoxia negates the increase in VEGF expression during vascular development resulting in underdeveloped retinal vasculature and apoptotic obliteration of the existing vessels. Thus, excessive vascular pruning may be the primary event in the pathogenesis of retinopathy of prematurity (Yamada et al.,1999; Keshet,2001). We recently demonstrated that TSP1 expression in wild type mice does not change during OIR (Wang et al.,2003). Furthermore, retinas of TSP1−/− mice are less susceptible to hyperoxia-mediated vessel obliteration. These studies indicate that a dramatic decrease in the levels of VEGF tips the angiogenesis balance in favor of TSP1, promoting EC apoptosis during hyperoxia. Thus, vessel obliteration that accompanies exposure to high oxygen requires down-regulation of VEGF levels and TSP1 expression (Wang et al.,2003). Our results show that TSP1 over-expressing transgenic mice are more susceptible to hyperoxia-mediated vessel obliteration and, as a result, the nonperfused area is significantly larger when compared to wild type mice. These observations are consistent with the antiangiogenic activity of TSP1, and they further emphasize the important role of TSP1 in vascular pruning and remodeling under hyperoxia.
The exposure of mice to room air for 5 days (normoxia) during OIR stimulates the growth of new vessels due to increased production of VEGF and tips the balance in favor of angiogenesis. However, over-expression of TSP1 in the transgenic mice opposes the effects of increased VEGF expression and thereby inhibits retinal neovascularization when compared to wild type mice. Our results suggest increased expression of TSP1 interferes with normal development of retinal vasculature and inhibits retinal neovascularization during OIR. Shafiee et al. (2000) recently showed antiangiogenic peptides from TSP1 inhibit retinal neovascularization in a rat model of OIR. Therefore, TSP1 is an important regulator of retinal vascular homeostasis, and its alterations, along with those of VEGF, under pathological conditions may contribute to retinal vasculopathies and neovascularization.
Bcl-2 is a cell death repressor and protects cells from apoptosis in response to various stimuli. Bcl-2 is a common target of many pro- and anti-angiogenic factors. Up-regulation of bcl-2 expression by proangiogenic factors such as VEGF promotes EC survival, while its down-regulation by antiangiogenic factors such as TSP1 promotes EC apoptosis (Nor et al.,2000). We recently showed that bcl-2−/− mice exhibit a delay in the development of retinal vasculature and reduced vascular density (Wang et al.,2005). This was associated with increased rates of apoptosis and proliferation in the retinal vasculature. Bcl-2−/− mice also failed to elicit a neovascular response during OIR despite an expression of VEGF similar to that of wild type mice. These defects are very similar to abnormalities observed in TSP1 over-expressing mice. Therefore, the effects of TSP1 over-expression on retinal vasculature may be mediated, at least in part, through down-regulation of bcl-2 expression in TSP1 transgenic mice.
The pupillary membrane and hyaloid vessels (hyaloid arteries, tunica vasculosa lentis, and vasa hyaloidea propria) provide nourishment to the immature lens, retina, and vitreous (Ito and Yoshioka,1999). However, they are known to regress during the later stages of ocular development by apoptosis. The contribution of TSP1 to these processes has not been previously addressed. We have shown TSP1 is present in the vitreous at relatively high amounts (Sheibani et al.,2000). Therefore, the presence of TSP1 in the vitreous may promote regression of hyaloid vessels that are intimately in contact with the vitreous. Our recent studies with TSP1−/− mice showed that the regression of hyaloid vessels, mainly the tunica vasculosa lentis, is significantly delayed in the absence of TSP1 compared to wild type mice (Wang et al.,2003).
However, TSP1 overexpressing mice consistently exhibited a more prominent hyaloid vasculature during early postnatal time points. The increase in hyaloid vasculature may be an attempt to compensate for reduced retinal vascular density in transgenic mice. However, the impact of TSP1 overexpression on regression of hyaloid vessels was minimal. It is possible that hyaloid vessel regression may be occurring at a maximum rate, and additional TSP1 would not have a significant impact on this process. However, we cannot rule out the possibility that increased TSP1 expression may influence hyaloid vessels function and/or organization.
In summary, these studies demonstrate that TSP1 is an important modulator of retinal vascular development and neovascularization. Alterations in TSP1 expression during development and/or under pathological conditions, such as diabetes and ischemia, may contribute to retinal vasculopathies and neovascularization. Therefore, TSP1 may provide an important target for manipulation of retinal vascularization under pathological conditions.
Generation and Screening of Transgenic Mice
The plasmid containing the 409 bp of murine αA-crystalline, 221 bp of the human β-globin intervening sequence 2, and the 237 bp SV40 polyadenylation signal (pαIpA-IVS2b) was obtained from Dr. Anne E. Griep (University of Wisconsin; Stolen et al.,1997). The cDNA for human TSP1 was removed from the pBSKS6TXE plasmid obtained from Dr. William A. Frazier (Washington University Medical School; Sheibani and Frazier,1995) by SacI and XbaI digestion, blunted by T4 DNA polymerase, and cloned into pαIpA-IVS2b at the unique BamHI site that was blunted by Klenow DNA polymerase (Fig 1A). The position and the integrity of the transgene unit was analyzed by restriction enzyme digestion and DNA sequence analysis. The transgene αATSP1 was excised by HindIII and KpnI digestion, isolated by agarose gel electrophoresis, and purified for microinjection by extraction using QIAEX gel extraction kit (Qiagen, Valencia, CA). Transgenic mice were generated on the C57BL/6j background at the University of Wisconsin Biotechnology Center's Transgenic Facility. Each founder transgenic mouse was bred to nontransgenic C57BL/6j mice to establish several lines of transgenic mice. Transgenic lines were maintained on the C57BL/6j background.
Genomic DNA was prepared from tail biopsies of mice, and transgenic mice were identified by PCR screening with primers hTSP3S (5′-CAGACTCAGCATTCAGC CTC-3′ from human TSP1 3′-UTR) and SVPA-1 (5′- TTTTCACTGCATTCTAGTTGTGG-3′ from SV40 sequences). This primer pair specifically amplified a 672-bp fragment from the genomic DNA of transgenic mice. The amplified fragments were electrophoresed in agarose gels and visualized by ethidium bromide staining. For Southern blot analysis, genomic DNA (5 μg) was digested with BamHI, electrophoresed on 0.8% agarose gels, transferred to Zeta-probe membrane (Bio-Rad, Hercules, CA), and hybridized with digoxigenin random-prime-labeled probe as described by the supplier (Roche Applied Science, Indianapolis, IN). The probe was a 1.3-Kbp BamHI 5′-internal fragment of human TSP1 cDNA prepared from pBSKS6TXE (Sheibani and Frazier,1995). This probe hybridized to the 1.3-Kbp BamHI fragment of αA-TSP1 transgene but not the endogenous mouse TSP1.
Clinical Evaluation of Murine Eyes
Mouse eyes were examined for ocular abnormalities using a Kowq SL-2 hand-held slit lamp and an ophthalmoscope. Mice were physically restrained during the analysis and their eyes were dilated with 1% tropicamide at least 15 min before evaluation. Both nontransgenic (10 eyes) and transgenic mice (36 eyes from six different TSP1 transgenic lines) were analyzed without the observers' prior knowledge of genotype. We also examined the retinal functions in wild type and transgenic mice by ERG analysis comparing growth of ERG A- and B-wave in response to increased flash intensity (Mizota and Adachi-Usami,2002).
TSP1 Western Blot Analysis
Eyes from 4-week-old mice (3 eyes from 3 different mice) were removed and lysed in 0.2 ml of RIPA buffer (50 mM Tris, pH 7.4, 150 mM NaCl, 1% Triton X-100, 1% sodium deoxycholate, 0.1% sodium dodecyl sulfate) containing a cocktail of protease inhibitors (Roche Applied Science, Indianapolis, IN), briefly sonicated, and centrifuged at 16,000g at 4°C. The supernatant was recovered and the protein concentration was determined using the BCA protein assay (Pierce, Rockford, IL). Samples (30 μg) were combined with equal volumes of 2× SDS loading buffer, boiled, electrophoresed in 4–20% polyacrylamide gel (Invitrogen, Carlsbad, CA), and electrotransferred to nitrocellulose membrane. The membrane was blocked for 1 h in blocking solution (3% BSA and 3% nondairy creamer), prepared in Tris-Buffered Saline (TBS, 20 mM Tris pH 7.6, 150 mM NaCl) containing 0.1% Tween 20 (TBST), and then incubated with a mouse monoclonal anti-human TSP1 antibody (A6.1; Neo Marker, Fremont, CA) for 1 h at room temperature. We have shown that this antibody reacts with TSP1 under nonreducing conditions in conditioned medium prepared from wild type but not TSP1−/− retinal endothelial cells (Su et al.,2003). The immune complexes were detected with a goat-anti-mouse horseradish peroxide (HRP)-conjugated secondary antibody (1:10,000; Pierce, Rockford, IL) followed by ECL detection (Amersham Biosciences, Piscataway, NJ).
Wild type and transgenic mice were bred for different experimental time points. All experiments were carried out in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Retinal vasculature was examined in retinal wholemounts prepared from mice grown in room air or during oxygen-induced ischemic retinopathy (OIR). For OIR, 7-day-old (P7) pups and their mother were placed in an airtight incubator and exposed to an atmosphere of 75 ± 0.5% oxygen for 5 days. Incubator temperature was maintained at 23 ± 2°C, and oxygen was continuously monitored with a PROOX model 110 oxygen controller (Reming Bioinstruments Co., Redfield, NY). Mice were brought to room air for 5 days, and then pups were sacrificed for retinal wholemount preparations as described below.
Trypsin-Digested Retinal Vessel Preparation
Eyes were enucleated from P21 or P42 mice and fixed in 4% paraformaldehyde for at least 24 h. The eyes were bisected equatorially, and the entire retina was removed under the dissecting microscope. Retinas were washed overnight in distilled water and incubated in 3% trypsin (Trypsin 1:250, Difco, Fisher Scientific) prepared in 0.1 M Tris, 0.1 M maleic acid, pH 7.8, containing 0.2 M NaF for approximately 1–1.5 h at 37°C. Following completion of digestion, retinal vessels were flattened by four radial cuts and mounted on glass slides for periodic acid-schiff (PAS) and hematoxylin staining. Nuclear morphology was used to distinguish pericytes from EC. The nuclei of EC are oval or elongated and lie within the vessel wall along the axis of the capillary, while pericyte nuclei are small, spherical, stain densely, and generally have a protuberant position on the capillary wall. The stained and intact retinal wholemounts were coded, and subsequent counting was performed masked.
The numbers of endothelial cells and pericytes were determined by counting respective nuclei under the microscope at a magnification of 400×. A mounting reticle (10 μm × 10 μm) was placed in one of the viewing oculars to facilitate counting. Only retinal capillaries were included in the cell count, which was performed in the mid-zone of the retina. We counted the number of endothelial cells and pericytes in four reticles from the four quadrants of each retina. The total number of endothelial cells and pericytes for each retina was determined by adding the numbers from the four reticles. The ratio of endothelial cells to pericytes was then calculated. To evaluate the density of cells in the capillaries, the mean number of endothelial cells or pericytes was recorded in four reticles from the four quadrants of each retina.
Visualization of Retinal Vasculature and Quantification of Avascular Areas
Vessel obliteration and the retinal vascular pattern were analyzed using retinal wholemounts stained with PECAM-1 antibody as described previously (Wang et al.,2003). At designated postnatal times, the mouse eyes were enucleated and briefly fixed in 4% paraformaldehyde (10 min on ice). The eyeballs were fixed in 70% ethanol for at least 24 h at −20°C. Retinas were dissected in PBS and then washed with PBS three times, 10 min each. Following incubation in blocking buffer (50% fetal calf serum, 20% normal goat serum in PBS) for 2 h, the retinas were incubated with rabbit anti-mouse PECAM-1 (prepared in our laboratory and diluted 1:250 in PBS containing 20% fetal calf serum, 20% normal goat serum) at 4°C overnight. Retinas were then washed three times with PBS, 10 min each, incubated with secondary antibody, Alexa 594 goat-anti-rabbit (Invitrogen, Carlsbad, CA; 1:500 dilution prepared in PBS containing 20% FCS, 20% NGS) for 2 h at room temperature, washed four times with PBS, 30 min each, and mounted on a slide with PBS/glycerol (2 vol/1 vol). Retinas were viewed by fluorescence microscopy, and images were captured in digital format using a Zeiss microscope (Carl Zeiss, Chester, VA). The central capillary dropout area was quantified as a percentage of the whole retinal area from the digital images, in masked fashion, using Axiovision software (Carl Zeiss, Chester, VA).
Quantification of Neovascular Proliferative Retinopathy
Quantification of vitreous neovascularization at P17 was performed as previously described (Wang et al.,2003). Briefly, mouse eyes were enucleated, fixed in formalin for 24 h, and embedded in paraffin. Serial sections (6 μm thick), each separated by at least 40 μm, were taken from around the region of the optic nerve. The hematoxylin and PAS-stained sections were examined in masked fashion for the presence of neovascular tufts projecting into the vitreous from the retina. The neovascular score was defined as the mean number of neovascular nuclei per section found in eight sections (four on each side of the optic nerve) per eye.
BrdU Labeling and Collagen IV Staining of Wholemount Retinas
The detection of cell proliferation on the retinal blood vessels was assessed by immunohistochemistry for 5-bromo-2-deoxyuridine (BrdU) incorporation followed by PECAM-1 staining of blood vessels. Mice were injected intraperitoneally with BrdU (Sigma, St. Louis, MO) at 120 mg/Kg of body mass dissolved in water. One and a half hours later, the animals were sacrificed; their eyes were removed and fixed immediately in 4% paraformaldehyde for 3 min on ice. Eyes were then transferred to 70% ethanol (v/v) and stored at −20°C for 2 to 72 h. Retinas were dissected in PBS, washed for 30 min in PBS containing 1% Triton X-100 to permeabilize cell membranes, and placed in 2 M HCl at 37°C for 1 h. Each retina was then washed in 0.1M sodium borate for 30 min to neutralize the HCl. Retinas were then washed in PBS containing 1% Triton X-100 for 15 min and incubated with a monoclonal antibody to BrdU (Roche; diluted 1:250 in PBS containing 1% bovine serum albumin, BSA) at 4°C overnight or at room temperature for 2 h. Following incubation, retinas were washed for 10 min in PBS containing 1% Triton X-100 and incubated with anti-mouse Alexa 488 antibody (Invitrogen, Carlsbad, CA) diluted 1:200 in PBS containing 1% BSA for 2 h. Retinas were then washed in PBS and stained with anti-collagen IV antibody as described above for staining of wholemount retinas. The collagen IV and secondary antibodies were used at 1:500 dilutions. After a final wash in PBS for 30 min, the retinas were mounted with the ganglion cell layer uppermost in PBS:glycerol (2 vol/1 vol). Retinas were viewed by fluorescence microscopy, and images were captured in digital format using a Zeiss microscope (Carl Zeiss). For quantification, the numbers of BrdU-positive nuclei on the blood vessels were determined per retina and recorded as number of proliferating cells/mm2 of retina.
Immunohistochemical Staining of the Frozen Sections
Mouse eyes were enucleated and embedded in optimal cutting temperature (OCT) compound at −80°C. Sections (9 μm) were cut on a cryostat, placed on glass slides, and allowed to dry for 2 h. For fluorescence microscopy, sections were fixed in cold acetone (4°C) on ice for 10 min, followed by three washes with PBS, 5 min each. Sections were incubated in blocker (1% BSA, 0.2% skim milk, and 0.3% Triton X-100 in PBS) for 15 min at room temperature. Sections were then incubated with rabbit anti-mouse type IV collagen (Chemicon Intenational, Inc., Temecula, CA; 1:500 dilution prepared in blocking solution) overnight at 4°C in a humid environment. After three washes in PBS, 5 min each, sections were incubated with secondary antibody Alexa 594 goat-anti-rabbit (Invitrogen, Carlsbad, CA; 1:500 dilution prepared in blocking solution). Sections were washed three times in PBS, covered with PBS:glycerol (2 vol/1 vol), and mounted with a coverslip. Retinal sections were viewed by fluorescence microscopy, and images were captured in digital format using a Zeiss microscope (Carl Zeiss).
Staining of the Hyaloid Vasculature
Following the removal of eyes, the sclera, choroids, and retinas were dissected anteriorly from the optic nerve to the limbus. The remaining wholemount specimen was stained with PECAM-1 antibody as described above to visualize the hyaloid vasculature. Wholemount specimens were viewed by fluorescence microscopy, and images were captured in digital format using a Zeiss microscope (Carl Zeiss). The dissection resulted, in most cases, in the loss of hyaloid artery and vasa hyaloidea propria vessels. However, the tunica vasculosa lentis vessels were clearly visible following PECAM-1 staining.
TdT-dUTP Terminal Nick-End Labeling and Active Caspase-3 Staining
Apoptotic cell death on the retinal vasculature was assessed by two different methods. TdT-mediated dUTP-biotin nick-end labeling (TUNEL) staining was performed on 9-μm frozen eye sections with the “MEBSTAIN Apoptosis Kit II” (Medical & Biological Laboratories, Watertown, MA) as recommended by the supplier. Briefly, the slides were preincubated with the TdT buffer II for 20 min at room temperature, and subsequently incubated with TdT solution (mixture of TdT buffer, Biotin-dUTP, and TdT) for 2 h at 37°C. The reaction was terminated by incubation with blocking solution at RT for 20 min. The TUNEL labeling was visualized by incubation with avidin-FITC II. TUNEL-positive cell nuclei were visualized using a fluorescence microscope and photographed in digital format as described above. As an alternative, activated caspase-3 was detected by immunohistological staining of wholemount retinas. After fixation as above, retinas were dissected in PBS and permeabilized in PBS with 1% (w/v) Triton X-100 for 45 min. The wholemount retinas were incubated overnight at 4°C with a cleaved activated caspase-3 antibody at 1:100 dilution (rabbit polyclonal, Asp 175; Cell Signaling Technology, Beverley, MA) as described above. Alex 594 goat anti-rabbit IgG (1:400 dilution, Molecular Probes) was used as a secondary antibody. To identify whether cells showing caspase-3 immuno-reactivity are the cells on the retinal vessels, retinal vasculature was also labeled with FITC-conjugated B4-lectin (1:100 dilution, Sigma, St. Louis, MO) and visualized as above. The retinas were examined using a 40× objective under the fluorescence microscope and the number of caspase-3-positive cells in the retinal vasculature was determined in each retina, as described for wholemount retinal trypsin digests and recorded as the number of apoptoic cells/μm2.
Statistical differences between groups were evaluated with Student's unpaired t-test (two-tailed). Means ± SD are shown. P ≤ 0.05 is considered significant.
We thank Dr. Anne E. Griep for providing us with the crystalline promoter and Dr. James N. Ver Hoeve for help with ERG analysis. We also thank Drs. Arthur Polans and Carol Kiekhaefer for critical reading of the manuscript and Norma Woods for editorial assistance. This work was funded in part by National Institutes of Health (EY13700, (N.S.) and DK67120 (C.M.S) and the Retina Research Foundation (N.S.). N.S. is a recipient of a Career Development Award from the Research to Prevent Blindness Foundation. Z.W. was supported in part by a fellowship from the Chinese Government.