Deletion of Smooth Muscle α-Actin Alters Blood–Retina Barrier Permeability and Retinal Function

Abstract

purpose. Vascular smooth muscle (SM) cells and pericytes are essential for normal vascular development. SM α-actin null mice were used to determine whether vascular SM and pericyte contractile functions, and not merely their presence, are necessary for vascular development, normal blood–retina barrier (BRB) permeability, and retinal function.

methods. Age-matched SM α-actin null and wild-type mice were analyzed. Retinal structure, vascular pattern, and SM cell and pericyte distribution were analyzed histologically. Retinal vascular permeability (RVP) was measured with the Evans blue dye method. Electroretinography (ERG) was performed to evaluate retinal function.

results. Deletion of SM α-actin did not result in any alterations in retinal morphology, vascular pattern, or SM cell and pericyte ensheathing of vessels in SM α-actin null mice. A significant increase in RVP was observed in SM α-actin null mice at both postnatal day (P)50 and P75 (P < 0.05 and P < 0.001, respectively). ERG analysis demonstrated a significant reduction in both rod and cone function in SM α-actin null mice at P22, P45, and P75 (P < 0.01 at all ages).

conclusions. These results demonstrate that SM α-actin in SM cells and pericytes is not necessary for the formation of a normal retinal vascular pattern; however, SM α-actin is necessary for SM cells and pericytes to interact with endothelial cells to form a fully functional BRB. These results are important in understanding the role of contractile gene expression in the maintenance and function of the BRB and may provide a model for studying pathologic conditions, such as diabetes, that alter the function of this barrier.

The formation and function of a normal retinal vasculature requires interactions between the inner endothelial cells and the surrounding mural cells, the vascular smooth muscle (SM) cells, and pericytes. ^{1 2} SM cells and pericytes express angiogenic factors that are critical in determining whether vessels will be stabile or will undergo branching and sprouting. ^{3 4} Blockage of pericyte recruitment during retinal vascular development results in profound aberrations in vessel patterning and endothelial hyperplasia. ^{5 6} Changes in retinal pericyte number and function appear early in diabetes, resulting in vessel destabilization with subsequent loss of normal retinal vasculature. The resultant ischemia often causes the abnormal neovascularization indicative of diabetic retinopathy. ^{2 7 8} Although it is clear that production of angiogenic factors by SM cells and pericytes is important in normal vascular development and function, little is known about the role SM α-actin may also play in this process. 

SM α-actin is one of six actin isoforms. It is the predominant actin isoform found in SM cells and is also expressed by pericytes. ⁹ In addition to SM α-actin, SM cells express SM γ-actin and the ubiquitous cytoplasmic actin isoforms β- and γ-actin. Pericytes do not express SM γ-actin but do express the cytoplasmic actins in addition to SM α-actin. The functional consequences of expression of these actin isoforms is only beginning to be understood. Recent studies have demonstrated that increased expression of SM α-actin results in increased contractile force generation. ¹⁰ In the SM α-actin null mouse, it has been demonstrated that SM cells lacking SM α-actin have decreased contractile force generation. ¹¹ Surprisingly, the only described phenotypic consequence of this decreased contractile force generation in SM cells of the SM α-actin null mouse is reduced peripheral blood pressure. 

Retinal blood vessels consist of an endothelial tube ensheathed by SM cells and pericytes. Ultrastructural studies of mature retinal blood vessels have identified SM cells as the cells surrounding arteries, arterioles, and veins and pericytes as the cells surrounding capillaries and venules. ^{12 13} In addition, retinal endothelial cells are connected by tight junctions that separate vascular and interstitial spaces and are responsible for the blood–retina barrier (BRB). ¹⁴ The primary function of SM cells in the mature retinal vascular is maintenance of blood flow and pressure through the contractility of these cells. Pericytes may also participate in the regulation of capillary blood flow through a contractile mechanism. ^{15 16} Unlike SM cells, all of which express SM α-actin, only a subpopulation of pericytes appears to express this actin isoform. ¹⁷ Pericytes nearer arterioles or venules appear to express more SM α-actin, whereas SM α-actin expression in pericytes throughout the capillary is very heterogeneous, and the extent of expression varies depending on species. ^{17 18}  

The expression of SM α-actin is a hallmark for the maturation of SM cells and pericytes in the retinal vasculature. This extensive expression of SM α-actin by these cells has been shown to be associated with vessel stability. ^{19 20} Previous studies have focused on the role angiogenic factors expressed by SM cells and pericytes have on vessel maturation and stabilization. ^{2 3 4 21} However, in addition to these angiogenic factors, studies have suggested that vessel stabilization may also result from a complex interplay of maturation processes, such as expression of SM α-actin and other SM-specific cytoskeletal proteins that contribute to the regulation of blood flow. ¹³  

To determine the role a SM α-actin in SM cells and pericytes in the formation and function of the retinal vasculature we have made use of a SM α-actin null mouse. We have found that, despite the appearance of a normal retinal morphology, vascular pattern, and pericyte ensheathing, SM α-actin null retinas have increased retinal vascular permeability and decreased retinal function. These results suggest that lack of expression of SM α-actin results in altered BRB function with a resulting increase in retinal vascular permeability and a decrease in retinal function. The expression of SM α-actin appears to be necessary for SM cells and pericytes to interact appropriately with endothelial cells to form a fully functional retinal vasculature. 

Methods and Materials

Animals

SM α-actin null and wild-type mice with the identical genetic background (C57Bl/6-sv129 hybrids) were used in this study. The gene-targeting strategy and phenotypes of the SM α-actin null mice have been described previously. ¹¹ Briefly SM α-actin null mice were generated by inserting a Pol II promoter neomycin hybrid gene into the SM α-actin gene locus at the +1 cap site. Previous studies have demonstrated that SM α-actin is not expressed in a variety of tissues in the null mouse. ^{11 22} For genotyping mice, genomic DNA was obtained from ear punches using the HotSHOT method ²³ and analyzed by RT-PCR. To determine the presence of the wild-type allele specific forward (5′-TTGTTCTGAGGGCTTAGGATGTT-3′) and reverse (5′-CTTTTCCAGTAAATCAAGCGTTGTT-3′) primers were made that flanked the region of the Pol2NeobpA cassette so that presence of the wild-type allele gave a 521-bp product. To determine the presence of the Pol2NeobpA cassette-specific forward (5′-TTGGGCAACACAGGCTGGTTAATC-3′) and reverse (5′-ATTGGAAGTAGCCGTTATTAGTGGA-3′) primers were made so that the forward primer was upstream of the insertion site and the reverse primer was in the Pol2NeoBPa cassette. The presence of the Pol2NeobpA cassette gene gave a 701-bp product. All genotyping was performed in duplicate. Care, use, and treatment of all animals in this study were in strict agreement with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research as well as the guidelines set forth by the University of Oklahoma Health Sciences Center. 

Western Immunoblot Analysis

Western immunoblot analysis was performed, to determine the levels of total actin, SM α-actin, and sarcomeric actin in null and wild-type mouse retinas as previously described. ²⁴ Retinas from three null and three wild-type mice were isolated, and total cellular proteins were solubilized by homogenizing the tissue in 2× SDS-PAGE sample buffer+2% β-mercaptoethanol followed by boiling for 3 minutes. As a control, mouse leg skeletal muscle was solubilized and treated similar to retinas. Total cellular protein concentration was determined (Protein dotMetric Kit; Geno Technology, Inc., St. Louis, MO). Equal amounts of protein were separated by SDS-PAGE (12.5%), transferred to polyvinylidene difluoride (PVDF) membranes (Micron Separations, Westboro, MA), blocked for 1 hour in blocking buffer (1× TTBS+2% casein), followed by overnight incubation with a mouse anti-SM α-actin monoclonal antibody (1:400 dilution, clone 1A4; Sigma-Aldrich, St. Louis, MO), a mouse anti-sarcomeric actin monoclonal antibody that recognizes both skeletal and cardiac α-actin ²⁵ (1:25,000 dilution, clone 5C5; Sigma-Aldrich) or a rabbit polyclonal anti-actin antibody that recognizes all actin isoforms (1:250 dilution; Sigma-Aldrich) followed by a goat anti-mouse or a goat anti-rabbit alkaline phosphatase (Sigma-Aldrich) for 1 hour and subsequently developed with chemiluminescent substrate (Immun-Star; Bio-Rad, Hercules, CA) for 5 minutes at room temperature (RT) and exposed to x-ray film (X-OMAT AR5; Eastman Kodak, Rochester, NY) for 5 to 120 seconds at 25°C, and the film was developed. 

Retinal Morphology and Immunohistochemistry

For examination of retinal morphology and immunolocalization of SM α-actin and glial fibrillary acid protein (GFAP), retinas from SM α-actin null and wild-type mice were isolated, fixed in 4% paraformaldehyde, embedded in paraffin, and sectioned. For morphology, sections were deparaffinized, rehydrated, stained with hematoxylin, and eosin, and analyzed. For immunohistochemistry paraffin sections were deparaffinized and rehydrated and blocked in normal goat serum (1:10 in PBS) for 30 minutes. For SM α-actin immunohistochemistry, sections were incubated with a monoclonal mouse anti-SM α-actin antibody conjugated to fluorescein (1:250 dilution; F3777; Sigma-Aldrich), followed by rabbit anti-fluorescein antibody (1:750 dilution; A889; Invitrogen, Eugene, OR) and subsequently incubated with biotinylated goat-anti-rabbit IgG (Vectastain ABC kit; Vector Laboratories, Burlingame, CA). The sections were then incubated with diluted ABC-alkaline phosphatase complex (Vector Laboratories). Control sections were treated identically, but without the primary antibody, and did not demonstrate any positive staining. 

For visualization of the retinal vascular pattern, whole eyes were collected and fixed in 4% paraformaldehyde for 30 minutes. Corneas and lenses were dissected, and eye cups were fixed an additional 8 hours. Whole retinas were blocked in horse serum (1:10 in PBS) for 1 hour, incubated in biotinylated Griffonia simplicifolia lectin II (1:200 dilution; B-1215; Vector Laboratories), followed by streptavidin conjugated to Alexa-fluor 488 (1:200 dilution; S-32,354; Invitrogen). 

To visualize the ensheathing of retinal vessels by SM cells and pericytes, we obtained whole retinas as described, blocked them in goat serum (1:10 in PBS), incubated them with anti-NG2 antibody (1:1000; obtained from William B. Stallcup, The Burnham Institute, La Jolla Cancer Research Center, La Jolla, CA) and subsequently with biotinylated goat-anti-rabbit IgG (Vecastain ABC kit; Vector Laboratories). The sections were then incubated with diluted ABC-alkaline phosphatase complex (Vector Laboratories). 

Measurement of Retinal Function

To assess the visual function of SM α-actin null and wild-type mice scotopic and photopic electroretinography (ERG) was measured (UTAS-E 3000 ERG system; LKC Technologies, Inc., Gaithersburg, MD), as previously described. ^{26 27} After overnight dark adaptation, mice were anesthetized with ketamine (80 mg/kg) and xylazine (16 mg/kg), placed on a heating pad and their pupils dilated (1% tropicamide; 2.5% phenylephrine HCl). For the assessment of rod photoreceptor function (scotopic ERG), a strobe flash stimulus was presented to the dark-adapted, dilated eyes in a Ganzfeld with a 138 cd · s/m² flash intensity. The amplitude of the a-wave was measured from the prestimulus baseline to the a-wave trough. The amplitude of the b-wave was measured from the trough of the a-wave to the peak of the b-wave. For the evaluation of cone function (photopic ERG), a strobe flash stimulus was presented to 5-minute light-adapted, dilated eyes in a Ganzfeld with a 79 cd · s/m² flash intensity. The amplitude of the cone b-wave was measured from the trough of the a-wave to the peak of the b-wave. 

Measurement of Vascular Permeability

Vascular permeability was quantified by measuring albumin leakage from blood vessels into the retina and iris using the Evans blue method according to documented protocols, with minor modifications. ^{28 29 30} In our previous studies using the Evans blue method we found very small individual differences in plasma levels that do not affect permeability results. ^{29 30} In addition, we have found that renal function is similar in wild-type and SM α-actin null mice based on urinary protein levels (Tomasek J, unpublished observations, 2005). Therefore, Evans blue levels in the plasma should be similar in both animals over the time span of the experiment. Evans blue dye (Sigma-Aldrich) was dissolved in normal saline (30 mg/mL). The mice were anesthetized, and Evans blue (30 mg/kg) was injected over 10 seconds through the femoral vein under microscopic inspection. Evans blue noncovalently binds to plasma albumin in the blood stream. ³¹ Immediately after Evans blue infusion, the mice turned visibly blue, confirming their uptake and distribution of the dye. The mice were kept on a warm pad for 2 hours to ensure the complete circulation of the dye, the chest cavity opened, and the mice perfused via the left ventricle with 1% paraformaldehyde in citrate buffer (pH 4.2), which was prewarmed to 37°C to prevent vasoconstriction. The perfusion lasted 2 minutes under a pressure of 120 mm Hg to clear the dye from the vessels. Immediately after perfusion, the eyes were enucleated, and the retina and iris were carefully dissected under an operating microscope. Evans blue dye was extracted by incubating each sample in 150 μL formamide for 18 hours at 70°C. The extract was centrifuged at 70,000 rpm (Rotor type: TLA-100.3; Beckman Coulter, Inc., Fullerton, CA) for 20 minutes at 4°C. Absorbance was measured using 100 μL of the supernatant at 620 nm. The concentration of Evans blue in the extracts was calculated from a standard curve and was normalized by the total protein concentration in each sample. Results were expressed in micrograms of Evans blue per milligram of total protein content. 

Results

Absence of SM α-Actin in the Retina

Previous studies have demonstrated that SM α-actin is not expressed in a variety of tissues in the SM α-actin null mouse. ^{11 22} To be certain that SM α-actin is not expressed in retinas of SM α-actin null mice and to characterize its distribution in wild-type retinas, we performed immunostaining and Western immunoblot analysis for SM α-actin in retinas from null and wild-type mice. Immunostaining for SM α-actin in wild-type retinas was only observed in perivascular cells and not other cell types (Figs. 1A 1B) , whereas no staining was observed in the null retinas (Fig. 1C) . Western immunoblot analysis also demonstrated SM α-actin expression in wild-type but not null retinas (Fig. 2A) . To confirm that SM α-actin null mice do not compensate for the lack of SM α-actin by expressing skeletal or cardiac α-actin retinas were analyzed by Western immunoblot analysis using an anti-sarcomeric α-actin antibody that recognizes these actin isoforms. ²⁵ No staining of either null or wild-type retinas were observed, whereas leg skeletal muscle from null mice stained strongly with this antibody (Fig. 2B) . 

Retinal Vascular Pattern and SM Cell and Pericyte Ensheathment

The interaction of SM cells and pericytes with endothelial cells is important in determining vessel stability and formation. ^{3 16 32 33} We examined whether lack of expression of SM α-actin in retinal SM cells and pericytes altered their ability to interact with endothelial cells and form a normal retinal vascular pattern. The vascular pattern of wild-type and SM α-actin-null retinas at P60 was examined by staining endothelial cells with Griffonia simplicifolia lectin. No differences in the vascular pattern of wild-type and SM α-actin retinas were observed (Figs. 3A 3B) . All areas of the null and wild-type retinas examined were vascularized, vessels appeared to be of similar size, density, and branching pattern, and there were no signs of neovascularization. SM cell and pericyte ensheathing was examined using anti-NG2 antibody. Although this assay does not quantify the number of pericytes, it allows for qualitative measurement of vessel ensheathing. ^{13 34} Vessels in both null and wild-type retinas were covered by SM cells or pericytes (Figs. 3C 3D) . There were no differences in the number of SM cells and pericytes (Figs. 3E 3F) . These results suggest that expression of SM α-actin is not necessary for the formation of retinal SM cells and pericytes and for these cells to interact with endothelial cells and participate in the development of a normal retinal vascular pattern. 

Retinal Vascular Permeability

The formation and maintenance of the BRB is critical to normal retinal function. ^{35 36 37} The zonular occludens junctions critical for BRB function are found between endothelial cells; however, it is unclear whether SM cells and pericytes may play a role in regulating the BRB. ^{38 39 40} We examined whether lack of expression of SM α-actin has an effect on BRB function by measuring retinal vascular permeability. Vascular permeability was quantified in postnatal day (P)50 and P75 SM α-actin null and wild-type mice by measuring protein leakage from blood vessels into the retina, using the Evans blue dye method. ^{28 29} Retinal vascular permeability was significantly increased in SM α-actin null mice compared with wild-type mice at both P50 and P75 (Fig. 4 ; P50, P < 0.05; P75, P < 0.001). The increased permeability of 1.6-fold of SM α-actin null to wild type at P75 is similar to the increase observed in diabetic over control rats. ²⁹ These results demonstrate that retinal vascular permeability is increased in SM α-actin null mice and strongly suggest that lack of expression of SM α-actin alters SM cells and pericytes interactions with endothelial cells and an abnormal BRB function. 

Decrease of Retinal Function

To determine the effect lack of expression of SM α-actin has on visual function scotopic and photopic ERGs were performed on SM α-actin null and wild-type mice. Rod and cone function in the SM α-actin null mice was significantly reduced compared with that in age-matched wild-type mice at all three ages examined (Fig. 5 ; P < 0.01). There was a 25% reduction in scotopic ERG a-wave and b-wave as early as P22 in SM α-actin null mice (Fig. 5 ; P < 0.01). An even greater loss of function of 45% was observed in the photopic ERG b-wave at P22 in SM α-actin null mice (Fig. 5 ; P < 0.01). There were slight changes in function in all three waveforms with age, increasing from P22 to P45 and decreasing from P45 to P75, suggesting a potential developmental delay; however, the only significant difference was in the photopic a-wave measurements between ages P45 and P75 (P < 0.05). These results demonstrate that lack of expression of SM α-actin in SM cells and pericytes alters retinal function. 

General Retinal Morphology

We examined whether lack of expression of SM α-actin and the correlated decrease in retinal function were associated with altered retinal morphology. Retinas from SM α-actin null and wild-type mice at P22, P45, and P75 (n > 2 for all points) were examined. No changes in retinal morphology were observed in SM α-actin null mice at any of the ages examined (Fig. 6) . We have also examined optic nerve structure in these eyes. No difference was noted between SM α-actin and wild-type eyes (not illustrated), suggesting that there has not been a change in intraocular pressure. These results suggest that lack of expression of SM α-actin does not result in a significant loss of cells in any of the layers of the retina and that the observed decreased retinal function are not the result of major changes in retinal morphology. 

Discussion

The interaction of SM cells and pericytes with endothelial cells is critical for vascular stabilization and formation. Although the importance of angiogenic factors produced by SM cells and pericytes in influencing retinal endothelial cell stability and proliferation has been demonstrated, ^{2 3 4 21} no studies have examined the role of SM α-actin in this process. In this study, we have examined the effect lack of expression of SM α-actin in SM cells and pericytes has on retinal vascular formation and function. We have found that the retinal vascular pattern and the ensheathing of blood vessels by SM cells and pericytes is normal in the SM α-actin null mice. However, SM α-actin null mice have increased retinal vascular permeability and decreased retinal function. We propose that lack of SM α-actin in SM cells and pericytes results in a dysfunctional BRB and the resulting increased retinal vascular permeability and decreased retinal function. These results suggest that the changes in SM cell and pericyte function due to the lack of SM α-actin are complex and may influence the final maturation of retinal blood vessels. Previous studies of diabetic retinopathy have demonstrated that the diabetic condition leads to a decreased number of functional pericytes, resulting in a dysfunctional BRB, increased retinal vascular permeability, and decreased retinal function. ^{35 36 37 41} The SM α-actin null mouse may provide a novel model for studying the underlying mechanisms by which changes in the contractile function of SM cells and pericytes result in alterations in BRB function, retinal vascular permeability, and retinal function that are important in diabetic retinopathy. 

In SM cells and pericytes the level of expression of SM α-actin is believed to be related to their level of contractile force generation. ^{9 11} It has been suggested that the maturation of SM cells and their consequent ability to regulate blood flow may play a key role in vessel stabilization. ¹³ It is clear from our data that SM α-actin expression is not necessary for vessels to develop normally and form a normal-appearing vasculature with no areas of vessel regression, neovascularization, or microaneurysms. However, it is not known whether these vessels would be more susceptible to remodeling in response to oxygen-induced retinopathy or diabetes. Our data demonstrate that expression of SM α-actin is more than just a maturation marker, as the lack of SM α-actin resulted in altered BRB function and decreased retinal function. It may be that the increased contractile force generation associated with expression of SM α-actin and associated capacity to regulate vascular tone is critical to the formation of fully mature blood vessels. 

Breakdown of the BRB is one of the most important pathophysiological changes in the early stages of diabetic retinopathy, as well as other ischemic or inflammatory retinal diseases ^{35 42} ; however, the exact mechanisms underlying the breakdown of the BRB and the increased permeability are largely unclear. Our results demonstrating increased retinal vascular permeability in SM α-actin null mice suggests that SM α-actin expression is necessary for the BRB to function properly. A functional BRB is dependent on the formation of tight junctions coupling together adjacent endothelial cells. ⁴³ SM α-actin is not expressed in endothelial cells; therefore, the increased retinal vascular permeability observed in SM α-actin null mice cannot be a direct effect on endothelial cells. Rather, lack of expression of SM α-actin must alter SM cells and pericyte function with a resultant effect on endothelial cells and BRB function. Decreased contractile force generation in SM cells and pericytes could have a dramatic impact on regulation of retinal blood flow and thereby influence pericyte-endothelial cell interactions critical to maintenance of the BRB. Mechanical stretching has been demonstrated in several cells to initiate intracellular signaling resulting in induction of gene expression, protein synthesis, and secretion of numerous growth factors. ⁴⁴ Decreased contractile force generation could effect the mechanical signaling of SM cells and pericytes or endothelial cells. Recently, it has been demonstrated that mechanical stretch of cultured pericytes induces expression of VEGF, ⁴⁵ a factor that increases retinal vascular permeability, suggesting a mechanism by which altered force generation could result in increased retinal vascular permeability. Understanding the mechanism by which lack of expression of SM α-actin increases retinal vascular permeability will help gain an understanding of how BRB function is altered in diabetic retinopathy and other ischemic and inflammatory retinal diseases. 

The loss of retinal function as measured by ERG is an early event in diabetes, with significant changes apparent within 2 weeks after diabetes onset in a rat model, ^{46 47} and often precedes the onset of microvascular lesions in patients with diabetic retinopathy. ⁴⁸ In the SM α-actin null mouse decreased retinal function was observed at P22, the earliest time point examined. This loss of function must be secondary to alteration in the function of SM cells and pericytes, since SM α-actin is not expressed in neuronal or neuronal supporting cells. It has been suggested in diabetes that decreased retinal function may be the result of metabolic changes due to altered glucose metabolism ⁴⁹ ; however, the lack of GFAP staining in retinal Müller cells (Tomasek J, unpublished observations, 2005) suggests that a similar metabolic change does not occur in SM α-actin null mice. The most likely cause of decreased retinal function is increased retinal vascular permeability and not the resultant changes in neuronal cells and the supporting neuronal cell metabolism. 

The SM α-actin null mouse provides the opportunity to study alterations in BRB and retinal function important in diabetic retinopathy in a nondiabetic model. Based on the results from this study, we propose the following model. The lack of expression of SM α-actin results in decreased contractile force generation in SM cells and pericytes. This reduction in force generation alters the interaction of these cells with endothelial cells in such a way as to result in loss of normal BRB function and increased retinal vascular permeability. We have observed a correlation between increased retinal vascular permeability and decreased retinal function, as evidenced by reduction in ERG. Abnormal BRB function has been associated with reduction in ERG function in patients with X-linked retinitis pigmentosa. ⁵⁰ However, whether increased retinal vascular permeability resulting for abnormal BRB function can lead to loss of function of the retina remains to be determined. This model also suggests that it is possible to separate BRB function and vessel stability. Vessels in nonstressed SM α-actin null mice appeared stabile; no evidence of vessel regression, neovascularization or microaneurysms was observed, and the vascular pattern appeared normal. In contrast, the BRB was dysfunctional, as evidenced by increased retinal vascular permeability. Expression of SM α-actin in SM cells and pericytes does not appear to be necessary for these cells to interact with endothelial cells to form stabile vessels with a normal pattern; however, expression of SM α-actin is necessary for a fully functional BRB to form. Future studies on the SM α-actin null mouse model should help elucidate mechanisms of BRB and retina function critical in understanding and treating diabetic retinopathy. 

 Supported by National Eye Institute Grants EY014435 (JJT), EY14052 (MRA), EY015650 (JXM), EY012231 (JXM), and EY12190 (Core), National Institutes of Health Research Resource Grant RR17703 (JDA); Research to Prevent Blindness (JDA); American Diabetes Association (JXM); and Juvenile Diabetes Research Foundation (JXM).

 Submitted for publication September 29, 2005; revised December 22, 2005; accepted March 23, 2006.

 Disclosure: J . J. Tomasek, None; C.J. Haaksma, None; R.J. Schwartz, None; D.T. Vuong, None; S.X. Zhang, None; J.D. Ash, None; J.-x. Ma, None; M.R. Al - Ubaidi, 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.

 

SM α-actin was found in perivascular cells in wild-type but not SM α-actin null retinas. (A) Immunostaining for SM α-actin was performed on a wholemount retina from a P24 wild-type mouse, demonstrating the pattern of perivascular cells in retinal vasculature. (B, C) Immunostaining for SM α-actin was performed on P45 retinal sections. In wild-type retinas SM α-actin was observed in perivascular cells in the retinal and choroidal vessels (B, arrows). No other SM α-actin-positive cells were observed in these sections. No SM α-actin staining was observed in perivascular cells in retinas from SM α-actin null mice demonstrating the lack of expression of SM α-actin (C). Scale bar: (A) 0.5 mm; (B, C) 50 μm.

Figure 2.

 

SM α-actin and sarcomeric actin was not expressed in SM α-actin null retinas. Retinas from P45 wild-type (WT) and SM α-actin null (Null) mice were dissected, solubilized, and analyzed by Western immunoblot. (A) Retinas contain equivalent levels of total actin (all actin isoforms). SM α-actin was found only in wild-type and not SM α-actin-null retinas (SM α-actin). (B) Sarcomeric actin was present in skeletal muscle tissue (SkM) but not in wild-type (WT) or SM α-actin null (Null) retinas.

Figure 3.

 

Retinal vascular pattern and SM cell and pericyte ensheathing was similar in wild-type and SM α-actin null mice. (A, B) The vascular pattern in retinas was examined by staining with Griffonia simplicifolia lectin. No noticeable differences in the retinal vascular pattern were observed between wild-type (A) and SM α-actin null mice (B). (C–F) The ensheathing of retinal blood vessels by SM cells and pericytes was analyzed by anti-NG2 antibody immunostaining. Immunostained SM cells were observed on larger vessels in both the wild-type (C) and SM α-actin null (D) retinas. Similarly, immunostained pericytes were observed on smaller vessels in both wild-type (E) and SM α-actin null (F) retinas. Scale bar: (A, B) 300 μm; (C, D) 200 μm; (E, F) 100 μm.

Figure 4.

 

BRB permeability is increased in SM α-actin null mice. BRB permeability was assessed using an Evans blue dye vascular permeability assay on P50 and P75 SM α-actin null and wild-type mice. Vascular permeability was significantly increased in SM α-actin null mice compared with wild-type mice at both P50 and P75.

Figure 5.

 

Retinal function is decreased in SM α-actin null mice. ERG was used to measure retinal function. Retinal function of SM α-actin null mice was normalized to the age-matched wild-type control and the data for SM α-actin null mice expressed as a percentage of the age-matched control. Retinal function was significantly decreased in SM α-actin null mice at all ages relative to the wild-type control (P < 0.01). There was a significant difference between photopic b-wave measurements at P45 and P75 (P < 0.05).

Figure 6.

 

SM α-actin null retinas demonstrated normal histology. Histologic analysis was performed on retinas from wild-type (A, C, E) and SM α-actin null (B, D, F) mice at P22 (A, B), P45 (C, D), and P75 (E, F). No alteration in retinal morphology was observed in SM α-actin null mice at any of the ages examined. Scale bar: 50 μm.

The authors thank Carla Hansens for assistance with general histology and Joel McRae for assistance with Western immunoblot analysis and mouse genotyping.