February 2017
Volume 58, Issue 2
Open Access
Retina  |   February 2017
Blood Vessel Basement Membrane Alterations in Human Retinal Microaneurysms During Aging
Author Affiliations & Notes
  • Mariana López-Luppo
    Center of Animal Biotechnology and Gene Therapy, Universitat Autònoma de Barcelona, Bellaterra, Spain
  • Victor Nacher
    Center of Animal Biotechnology and Gene Therapy, Universitat Autònoma de Barcelona, Bellaterra, Spain
    Department of Animal Health and Anatomy, School of Veterinary Medicine, Universitat Autònoma de Barcelona, Bellaterra, Spain
  • David Ramos
    Center of Animal Biotechnology and Gene Therapy, Universitat Autònoma de Barcelona, Bellaterra, Spain
    Interdisciplinary Centre of Research in Animal Health, Faculty of Veterinary Medicine, Universidade de Lisboa, Lisbon, Portugal
  • Joana Catita
    Center of Animal Biotechnology and Gene Therapy, Universitat Autònoma de Barcelona, Bellaterra, Spain
    Department of Animal Health and Anatomy, School of Veterinary Medicine, Universitat Autònoma de Barcelona, Bellaterra, Spain
  • Marc Navarro
    Center of Animal Biotechnology and Gene Therapy, Universitat Autònoma de Barcelona, Bellaterra, Spain
    Department of Animal Health and Anatomy, School of Veterinary Medicine, Universitat Autònoma de Barcelona, Bellaterra, Spain
  • Ana Carretero
    Center of Animal Biotechnology and Gene Therapy, Universitat Autònoma de Barcelona, Bellaterra, Spain
    Department of Animal Health and Anatomy, School of Veterinary Medicine, Universitat Autònoma de Barcelona, Bellaterra, Spain
  • Alfonso Rodriguez-Baeza
    Department of Morphological Sciences, School of Medicine, Universitat Autònoma de Barcelona, Bellaterra, Spain
  • Luísa Mendes-Jorge
    Center of Animal Biotechnology and Gene Therapy, Universitat Autònoma de Barcelona, Bellaterra, Spain
    Interdisciplinary Centre of Research in Animal Health, Faculty of Veterinary Medicine, Universidade de Lisboa, Lisbon, Portugal
    Department of Morphology and Function, Faculty of Veterinary Medicine, Universidade de Lisboa, Lisbon, Portugal
  • Jesús Ruberte
    Center of Animal Biotechnology and Gene Therapy, Universitat Autònoma de Barcelona, Bellaterra, Spain
    Department of Animal Health and Anatomy, School of Veterinary Medicine, Universitat Autònoma de Barcelona, Bellaterra, Spain
    Interdisciplinary Centre of Research in Animal Health, Faculty of Veterinary Medicine, Universidade de Lisboa, Lisbon, Portugal
    CIBER de Diabetes y Enfermedades Metabólicas Asociadas (CIBERDEM), Spain
  • Correspondence: Jesús Ruberte, CBATEG, Edifice H - Campus UAB, 08193 Bellaterra, Barcelona, Spain; Jesus.Ruberte@uab.cat
Investigative Ophthalmology & Visual Science February 2017, Vol.58, 1116-1131. doi:10.1167/iovs.16-19998
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to Subscribers Only
      Sign In or Create an Account ×
    • Get Citation

      Mariana López-Luppo, Victor Nacher, David Ramos, Joana Catita, Marc Navarro, Ana Carretero, Alfonso Rodriguez-Baeza, Luísa Mendes-Jorge, Jesús Ruberte; Blood Vessel Basement Membrane Alterations in Human Retinal Microaneurysms During Aging. Invest. Ophthalmol. Vis. Sci. 2017;58(2):1116-1131. doi: 10.1167/iovs.16-19998.

      Download citation file:


      © 2017 Association for Research in Vision and Ophthalmology.

      ×
  • Supplements
Abstract

Purpose: Microaneurysms, considered a hallmark of retinal vascular disease, are present in aged retinas. Here, the basement membrane of human retinal microaneurysms has been analyzed during aging.

Methods: Retinas were obtained from 17 nondiabetic donors. Whole mount retinas and paraffin sections were marked immunohistochemically with antibodies against the main components of the blood basement membrane. Trypsin digestion and transmission electron microscopy also were performed.

Results: Small microaneurysms presented increased expression of collagen IV, laminin, fibronectin, nidogen, and perlecan, along with basement membrane thickening. Unexpectedly, crosslinked fibrils of collagen III, a type of collagen absent in retinal capillaries, were found specifically in small microaneurysms. This was parallel to enhanced lysyl oxidase-like (LOXL) 2 and 4 expression. Large microaneurysms showed diminution of protein content, as well as disorganization, in their basement membrane. This was concomitant with an increased expression of matrix-metalloproteinase (MMP)-9 and plasminogen activator inhibitor (PAI)-1. Pericyte coverage declined between small and large microaneurysms.

Conclusions: Thickening of the basement membrane in small microaneurysms by accumulation of matrix proteins probably produced by recruited pericytes, together with the appearance of crosslinked collagen III fibrils probably due to the action of LOXL2 and LOXL4, could be considered as compensatory mechanisms to strengthen the vascular wall in the early phase of microaneurysm formation. Later, increased activity of MMP-9 and PAI-I, which produce disruption of the blood basement membrane and expansion of microthrombi respectively, and loss of pericytes, which produces weakening of the vascular wall, could explain the wall dilation observed in the late phase of microaneurysm formation.

Retinal microaneurysms are dilations of capillaries that appeared as gross outpouchings of the vessel wall.1,2 The pioneer work of Ballantyne and Loewenstein,3 the first to use the term diabetic retinopathy, had led us to believe that such lesions were only rarely found apart from the diabetic retina. However, retinal microaneurysms are associated with other retinal conditions, such as venous thrombosis, glaucoma, Eales' disease, hypertension, arteriosclerosis, chorioretinitis, anemia, retinoblastoma, and, occasionally, with normal retina.4 
Microaneurysms are present in old human healthy retinas.5 Available epidemiologic data, using retinal photography, indicate a higher prevalence of microvascular abnormalities, including microaneurysms, in older adults without diabetes.69 More detailed postmortem studies showed microaneurysms in 24% of old retinas, in which no eye disease was suspected and in which no gross ophthalmoscopic findings were expected in life.10 
Different mechanisms have been involved in the development of microaneurysms. Potential contributors to the appearance of microaneurysms could be pericyte cell death,11,12 endothelial proliferation and impaired angiogenesis,1316 changes in intraluminal rheology,1719 and blood vessel basement membrane damage.20 At present, there still is a debate about the mechanism for microaneurysm formation, and knowledge about the role of the blood vessel basement membrane in the development of these vascular abnormalities is particularly scarce. 
In this study, we examined postmortem the presence of retinal microaneurysms in the eyes of a nonpreselected population of human old donors. Microaneurysms were classified according to their size, using a simplification of the scheme previously suggested by Moore et al.2 and Dubow et al.,21 as small and large microaneurysms. The expression of a selection of basement membrane proteins and the ultrastructure of the basement membrane were analyzed to understand the changes that occurred during microaneurysm formation. Increased expression of collagen IV, laminin, fibronectin, nidogen, and perlecan, along with basement membrane thickening, was found in small microaneurysms. Unexpectedly, crosslinked fibrils of collagen III, a type of collagen absent in retinal capillaries, were found specifically in small microaneurysms. This was parallel to enhanced lysyl oxidase-like (LOXL) 2 and 4 expression, which enzymatically crosslink collagen. If it could be assumed that small microaneurysms observed in this study are at an earlier stage of their development, our results suggested that during the first phases of microaneurysm formation a compensatory mechanism attempts to reinforce the vascular wall to stop the increased wall dilatation that occurs in later phases. Thereafter, the action of increased expression of matrix-metalloproteinase (MMP)-9, which produces a disorganization and a reduction in the content of basement membrane proteins, and the overexpression of plasminogen activator inhibitor (PAI)-1 that facilitates the expansion of microthrombi, could explain later the wall dilation observed in large microaneurysms. A decline in pericyte coverage also was evident between small and large microaneurysms, further contributing to vascular wall weakening during microaneurysm formation. 
Materials and Methods
Human Retinas
Human eye samples were obtained from voluntary body donations to the Faculty of Medicine at Universitat Autònoma de Barcelona (UAB). The procedure of body donation was approved 03-27-2015 by the Ethics Committee (CEEAH/UAB) and was in accordance with the Catalonian law (DECRET 297/1997, de 25 de Novembre). All subjects were treated in accordance with the Declaration of Helsinki. All participants in the study gave a written consent during lifetime for donation of their bodies after death for teaching and research purposes. Our laboratory received human eye samples without knowledge of the donor's identity, only biologic data, such as sex, age, clinical history, and cause of death, were available (Table 1). Eyes were obtained from 17 donors: 14 old aged donors (X̄ = 85.28 years old) and 3 middle aged donors (X̄ = 40 years old). The corneas were removed, and the retinas were examined using a dissecting microscope for pathologic evidence of neovascularization. No donor had a history of diabetic disease (Table 1), nor did they present with the characteristic retinal diabetic lesions of proliferative retinopathy. However, the possibility of undiagnosed asymptomatic nonproliferative diabetic retinopathy cannot be ruled out in this pool of old aged donors. Eyes were fixed in a solution of paraformaldehyde 4% with picric acid in 0.01 M PBS for 24 hours. The time for fixation after death was critical to obtain adequate samples for immunohistochemistry and electron microscopy. The average time for fixation in our study was 7 hours (Table 1). Next, the eyes were washed in PBS, partially dehydrated, and maintained in 70° alcohol solution at 4°C. The storage of prefixed eyes in this cold alcoholic solution allowed us to perform immunohistochemistry and electron microscopy during a long period of time. For immunohistochemical analysis, retinal fragments of approximately 3 mm2 were dissected out from the prefixed eyes and processed adequately for each experiment. 
Table 1
 
Sex, Age, Cause of Death, and Time Between Death and Eye Fixation
Table 1
 
Sex, Age, Cause of Death, and Time Between Death and Eye Fixation
Immunohistochemistry
Retinal fragments from all donors were used for immunohistochemistry (Table 1). Paraffin-sections and whole mount retinas were incubated overnight at 4°C with the following antibodies: mouse anti-human collagen III (Sigma-Aldrich Corp., St Louis, MO, USA) at 1:200 dilution, goat anti-mouse collagen IV (Merck Millipore, Billerica, MA, USA) at 1:200 dilution, rabbit anti-mouse collagen IV (Chemicon International, Temecula, CA, USA) at dilution 1:200, rabbit anti-human fibronectin (BD Pharmingen, CA, USA) at 1:100 dilution, rabbit anti-human laminin (Dako Cytomation, Glostrup, Denmark) at 1:200 dilution, mouse anti-human MMP-9 (Abcam, Cambridge, UK) at 1:100 dilution, rabbit anti-human nidogen (Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA) at 1:50 dilution, mouse anti-human PAI-1 (Abcam) at 1:200 dilution, rabbit anti-human perlecan (Santa Cruz Biotechnology, Inc.), mouse anti-human LOX (Santa Cruz Biotechnology) at 1:100 dilution, mouse anti-human LOXL1(Santa Cruz Biotechnology) at 1:100 dilution, mouse anti-human LOXL2 (Sigma-Aldrich Corp.) at 1:50 dilution, mouse anti-human LOXL3 (Santa Cruz Biotechnology) at 1:100 dilution, mouse anti-human LOXL4 (Sigma-Aldrich Corp.) at 1:200 dilution, goat anti-human α-smooth muscle actin (SMA; Abcam) at 1:100, and rabbit anti-human Ki-67 (Abcam) at 1:100. After they were washed in PBS, the retinas were incubated at 4°C overnight with specific secondary antibodies: anti-goat IgG Alexa 568 (Invitrogen, Carlsbad, CA, USA) at 1:250 dilution; anti-rabbit IgG Alexa 568 (Invitrogen) at 1:250 dilution, and biotinylated anti-mouse IgG (Vector Laboratories, Burlingame, CA, USA) at 1:100 dilution. For fluorescence visualization, streptavidin Alexa Fluor 488 (Invitrogen) at 1:100 dilution was used. Nuclear counterstaining was performed with To-Pro-3 iodide (Invitrogen) at 1:100 dilution for microscopic analysis using a laser-scanning confocal microscope Leica (TCS SP2; Leica Microsystems GmbH, Heidelberg, Germany). 
Histologic Protein Semiquantification and Colocalization Analysis
For histologic protein semiquantification, optical sections through the entire microaneurysm were made every micrometer. A summation of optical sections was calculated and an intensity profile, which estimates the amount of labeled protein using a color code, was obtained. 
Spatial colocalization between two fluorescently labeled proteins was determined by the Leica LAS Af Lite imaging software. Colocalization between collagen IV and other basement membrane proteins (laminin, fibronectin, nidogen, and perlecan) was analyzed, examining a total of 30 microaneurysms and 30 capillaries selected randomly in paraffin sections. The degree of colocalization between green fluorochrome (collagen IV) and red fluorochrome (fibronectin, laminin, perlecan, or nidogen) was quantified using the Pearson Coefficient,22 which compares the intensities of homologous pixels in which 1 is the maximal or perfect colocalization, 0 indicates randomly localization and −1, the minimal colocalization or perfect exclusion. 
Trypsin Digestion and Vascular Cell Quantification
For trypsin digestion, retinal fragments from the prefixed eyes of four aged donors (X̄ = 79 years old) were dissected and digested in 3% trypsin as described previously.23 The retinal vascular tree then was washed in filtered water, mounted to a clean slide and stained with PAS and hematoxylin. 
Total number of cells was determined in small microaneurysms (n = 11) and large microaneurysms (n = 4) from trypsin-digested retinas. Microaneurysm areas were calculated using an imaging analyzing system (ImageJ). Total cell numbers then were standardized to microaneurysm area (number of cells per mm2 of microaneurysm area). 
Transmission Electron Microscopic Analysis
Retinal fragments of 1 mm2, from the prefixed eyes of 3 old aged donors and 2 middle aged donors (Table 1), were fixed in 2.5% glutaraldehyde and 2% paraformaldehyde for 2 hours at 4°C. After washing in 0.1 M phosphate buffer, samples were postfixed in 1% osmium tetroxide, dehydrated through a graded acetone series and embedded in spurr resin. Ultrathin sections (70 nm) were stained using lead citrate and uranyl acetate. Then, sections were examined by a transmission electron microscope (H-7000; Hitachi Ltd., Tokyo, Japan). 
Statistical Analysis
Results were expressed as mean ± SD. Differences between groups were evaluated by the paired or unpaired Student's t-test, as appropriate. A P value <0.05 was considered significant. 
Results
Identification and Classification of Microaneurysms
Microaneurysms were identified as localized dilations of retinal capillaries in whole mount basement membrane immunostained retinas (Fig. 1A). These abnormalities were present in all the retinas obtained from old donors (n = 14), but were absent in all the retinas obtained from middle aged donors (n = 3). Using a simplification of the scheme previously suggested by Moore et al.2 and recently updated by Dubow et al.,21 we classified microaneurysms as small and large. Small microaneurysms, or focal bulges, are vascular dilations where the combined width of the focal bulge and the associated capillary was less than twice the width of an adjacent nonaffected capillary region. In large microaneurysms, the combined width of the vascular dilation and the associated capillary, was more than twice the width of an adjacent nonaltered capillary. The analysis of 110 randomly selected microaneurysms showed that small microaneurysms (54.5%) were more abundant than large microaneurysms (45.5%). Large microaneurysms frequently disrupted the retinal cellular architecture (Fig. 1B). 
Figure 1
 
Identification and classification of retinal microaneurysms. (A) Microaneurysms were identified as focal outpouchings of retinal capillaries in whole mount retinas immunohistochemically marked with antibodies against the blood basement membrane. A microaneurysm was classified as “small” when the combined width of the focal bulge and the associated capillary was less than twice the width of an adjacent nonaffected capillary region, and “large” when the combined width of the vascular dilation and the associated capillary was more than twice the width of an adjacent nonaltered capillary. (B) Comparison between a normal retinal topography without microaneurysms from a middle-aged donor and a disorganized retina with the presence of a large microaneurysms from an old aged donor. SM, small microaneurysm; LM, large microaneurysm; GCL, ganglion cell layer; INL, inner nuclear layer; ONL, outer nuclear layer. Nuclei were counterstained with ToPro-3 (blue). Scale bars: (A) = 34.2 μm; (B) = 30.68 μm; (C) = 28.26 μm.
Figure 1
 
Identification and classification of retinal microaneurysms. (A) Microaneurysms were identified as focal outpouchings of retinal capillaries in whole mount retinas immunohistochemically marked with antibodies against the blood basement membrane. A microaneurysm was classified as “small” when the combined width of the focal bulge and the associated capillary was less than twice the width of an adjacent nonaffected capillary region, and “large” when the combined width of the vascular dilation and the associated capillary was more than twice the width of an adjacent nonaltered capillary. (B) Comparison between a normal retinal topography without microaneurysms from a middle-aged donor and a disorganized retina with the presence of a large microaneurysms from an old aged donor. SM, small microaneurysm; LM, large microaneurysm; GCL, ganglion cell layer; INL, inner nuclear layer; ONL, outer nuclear layer. Nuclei were counterstained with ToPro-3 (blue). Scale bars: (A) = 34.2 μm; (B) = 30.68 μm; (C) = 28.26 μm.
Basement Membrane Protein Analysis in Small Microaneurysms
Blood vessel basement membranes are continuous self-assembled layers of proteins, glycoproteins, and proteoglycans, such as collagen IV, laminin, fibronectin, nidogen, and perlecan.24 Collagen IV, the major structural amorphous blood vessel basement membrane component,25 provides stiffness to the vessel wall26 and was overexpressed in small microaneurysms (Fig. 2A). Laminin, an adhesion glycoprotein that binds to collagen IV through nidogen and perlecan,27 also was overexpressed in small microaneurysms (Fig. 2B). Similar results were observed for other adhesion glycoproteins, such as fibronectin (Fig. 2C) and nidogen (Fig. 2D). Perlecan, a heparan sulfate glycoprotein involved in selective filtration,28 also was overexpressed in small microaneurysms. Altogether, these results indicated that the main components of blood vessel basement membrane were increased in small retinal microaneurysms during aging, and that this was associated with a thickening of the blood basement membrane demonstrated by transmission electron microscopy (Fig. 3). 
Figure 2
 
Analysis of basement membrane protein expression in small retinal microaneurysms. Small microaneurysms showed an increased expression of collagen IV (A), laminin (B), fibronectin (C), nidogen (D), and perlecan (E). IP, intensity profile from a maximum 2D projection obtained by making optical sections every micrometer through the entire microaneurysm length. Scale bars: (A) = 22.46 μm; (B) = 6.94 μm; (C) = 10.99 μm; (D) = 25.5 μm; (E) = 15.17 μm.
Figure 2
 
Analysis of basement membrane protein expression in small retinal microaneurysms. Small microaneurysms showed an increased expression of collagen IV (A), laminin (B), fibronectin (C), nidogen (D), and perlecan (E). IP, intensity profile from a maximum 2D projection obtained by making optical sections every micrometer through the entire microaneurysm length. Scale bars: (A) = 22.46 μm; (B) = 6.94 μm; (C) = 10.99 μm; (D) = 25.5 μm; (E) = 15.17 μm.
Figure 3
 
Ultrastructural analysis of the basement membrane (BM) in small retinal microaneurysms. The comparison between the vessel walls of a nonaffected capillary (A) and a small microaneurysm (B) was assessed by transmission electron microscopy. Magnification evidenced thickening of the basement membrane in small microaneurysms. As, astrocyte; EC, endothelial cell; Er, erythrocyte; Lu, lumen; Mü, Müller cell; Pe, pericyte. Scale bars: (A) = 9.97 μm; (B) = 7.13 μm.
Figure 3
 
Ultrastructural analysis of the basement membrane (BM) in small retinal microaneurysms. The comparison between the vessel walls of a nonaffected capillary (A) and a small microaneurysm (B) was assessed by transmission electron microscopy. Magnification evidenced thickening of the basement membrane in small microaneurysms. As, astrocyte; EC, endothelial cell; Er, erythrocyte; Lu, lumen; Mü, Müller cell; Pe, pericyte. Scale bars: (A) = 9.97 μm; (B) = 7.13 μm.
Under stress, astrocytes express collagen IV.29 To assure that collagen IV overexpression in small microaneurysms was not produced by neuroglial activation, a double immunostaining was performed in whole mount retinas with specific antibodies for collagen IV and glial fibrillary acidic protein (GFAP), a specific marker for astrocytes.30 The lack of colocalization between these two proteins (Fig. 4) suggested that the overexpression of collagen IV was a basement membrane intrinsic phenomenon associated with the genesis of microaneurysms. 
Figure 4
 
Analysis of colocalization between GFAP and collagen IV in small retinal microaneurysms. Collagen IV (red) in small microaneurysms did not colocalize with GFAP (green), suggesting that the overexpression of collagen IV in small microaneurysms is not related to formation of a glial scar. As expected, astrocytic end-feet processes marked by GFAP were in contact with the basement membrane of retinal capillaries. Furthermore, nuclei were counterstained with ToPro-3 (blue). Arrowhead, end-feed processes. Scale bar: 18.87 μm.
Figure 4
 
Analysis of colocalization between GFAP and collagen IV in small retinal microaneurysms. Collagen IV (red) in small microaneurysms did not colocalize with GFAP (green), suggesting that the overexpression of collagen IV in small microaneurysms is not related to formation of a glial scar. As expected, astrocytic end-feet processes marked by GFAP were in contact with the basement membrane of retinal capillaries. Furthermore, nuclei were counterstained with ToPro-3 (blue). Arrowhead, end-feed processes. Scale bar: 18.87 μm.
In healthy conditions, collagen III, a fibrous type of collagen,31 is not present in the basement membrane of retinal capillaries.32 Unexpectedly, in whole mount retinas collagen III was specifically overexpressed in small microaneurysms, but remained absent in the adjacent nonaffected capillary regions (Fig. 5A). In accordance with this, collagen fibrils showing a periodicity of 67 nm, a feature strongly suggestive of the presence of collagen III, were observed in the basement membrane of microaneurysms by transmission electron microscopy (Fig. 5B). Collagen III fibrils inside the basement membrane appeared parallel and, in some areas, crosslinked (Fig. 5B). 
Figure 5
 
Anomalous expression of collagen III in small retinal microaneurysms. (A) Laser confocal analysis of whole mount retinas evidenced the presence of collagen III in small microaneurysms. Adjacent nonaltered capillaries did not express collagen III. (B) Transmission electron microscopy revealed the existence of fibrils (arrows) in the thickened basement membrane of small microaneurysms. Fibrils have periodic cross-striations every 67 nm, which is compatible with the morphology of collagen III fibrils. In some regions, fibrils were crosslinked (circle). Healthy capillaries do not exhibit collagen fibrils in the basement membrane, presenting the amorphous characteristic aspect. Scale bars: (A) = 14.19 μm; (B) = 0.53 μm.
Figure 5
 
Anomalous expression of collagen III in small retinal microaneurysms. (A) Laser confocal analysis of whole mount retinas evidenced the presence of collagen III in small microaneurysms. Adjacent nonaltered capillaries did not express collagen III. (B) Transmission electron microscopy revealed the existence of fibrils (arrows) in the thickened basement membrane of small microaneurysms. Fibrils have periodic cross-striations every 67 nm, which is compatible with the morphology of collagen III fibrils. In some regions, fibrils were crosslinked (circle). Healthy capillaries do not exhibit collagen fibrils in the basement membrane, presenting the amorphous characteristic aspect. Scale bars: (A) = 14.19 μm; (B) = 0.53 μm.
To examine whether lysyl oxydases were involved in the crosslinking of collagen III fibrils in small microaneurysms, the expression of the following lysyl oxidases (LOX, LOXL1, LOXL2, LOXL3, and LOXL4) was analyzed. Lysyl oxidases are copper-containing amine oxidases that oxidize amine substrates to reactive aldehydes, producing the specific crosslinking of collagen and elastin.33 This crosslinking increases the strength of the vessel wall.34 Our observations indicated that only LOXL2 and LOXL4 were expressed all throughout the extension of large and small blood vessels, including capillaries, in aging human retinas (Figs. 6, 7). LOX, LOXL1, and LOXL3 expression was not observed in retinal blood vessels (data not shown). In addition, LOXL2 and LOXL4 colocalized with collagen IV in the blood vessel basement membrane (0.84 ± 0.02, n = 10 and 0.89 ± 0.01, n = 9, Pearson's arbitrary units, respectively), indicating the presence of lysyl oxydases in contact with their substrate and strongly suggesting that these lysyl oxydases were activated. Furthermore, LOXL2 and LOXL4 expression was increased in small microaneurysms when compared to adjacent nonaffected capillary regions (Figs. 6C, 7C). Together, these results pointed to LOXL2 and LOXL4 as possible players for the crosslinking of collagen III fibrils in small microaneurysms. 
Figure 6
 
Analysis of LOXL2 expression in old aged retinas. Immunodetection of LOXL2 (green) was observed in all the blood vessels (arrowheads), in whole mount (A), and paraffin sections (B), of human retinas. Few cells (arrows) in the inner nuclear layer also express LOXL2. Collagen IV (red) colocalized with LOXL2, suggesting that this lysil oxidase-like was activated. (D) In small microaneurysms, LOXL2 was overexpressed, compared to the low expression exhibited by the adjacent nonaffected capillary. a, arteriole. Nuclei were counterstained with ToPro-3 (blue). Scale bars: (A) = 122.92 μm; (B) = 29.17 μm; (C) = 14.67 μm.
Figure 6
 
Analysis of LOXL2 expression in old aged retinas. Immunodetection of LOXL2 (green) was observed in all the blood vessels (arrowheads), in whole mount (A), and paraffin sections (B), of human retinas. Few cells (arrows) in the inner nuclear layer also express LOXL2. Collagen IV (red) colocalized with LOXL2, suggesting that this lysil oxidase-like was activated. (D) In small microaneurysms, LOXL2 was overexpressed, compared to the low expression exhibited by the adjacent nonaffected capillary. a, arteriole. Nuclei were counterstained with ToPro-3 (blue). Scale bars: (A) = 122.92 μm; (B) = 29.17 μm; (C) = 14.67 μm.
Figure 7
 
Analysis of LOXL4 expression in old aged retinas. Immunodetection of LOXL4 (green) was observed exclusively in the blood vessels (arrowheads), in whole mount (A) and paraffin sections (B), of human retinas. Collagen IV (red) colocalized with LOXL4, suggesting that this lysil oxidase-like was activated. (D) In small microaneurysms, LOXL4 was overexpressed. Nuclei were counterstained with ToPro-3 (blue). Scale bars: (A) 187.35 μm; (B) = 25.78 μm; (C) = 13.83 μm.
Figure 7
 
Analysis of LOXL4 expression in old aged retinas. Immunodetection of LOXL4 (green) was observed exclusively in the blood vessels (arrowheads), in whole mount (A) and paraffin sections (B), of human retinas. Collagen IV (red) colocalized with LOXL4, suggesting that this lysil oxidase-like was activated. (D) In small microaneurysms, LOXL4 was overexpressed. Nuclei were counterstained with ToPro-3 (blue). Scale bars: (A) 187.35 μm; (B) = 25.78 μm; (C) = 13.83 μm.
Basement Membrane Protein Analysis in Large Microaneurysms
Basement membrane protein expression decreases in large microaneurysms in comparison with small microaneurysms (Figs. 2, 8). The greatest diminution was observed in nidogen and perlecan expression, which clearly was inferior in large microaneurysms when compared to adjacent nonaltered capillary regions (Figs. 7D, 7E) and with small microaneurysms (Figs. 2D, 2E). The expression pattern of basement membrane proteins was less homogeneous than in small microaneurysms, showing large regions of low expression with some small areas of high protein concentration (Fig. 8). Furthermore, colocalization of laminin, fibronectin, nidogen, and perlecan with collagen IV was significantly reduced in large microaneurysms when compared to adjacent nonaffected capillary regions (Table 2; Fig. 9). Altogether these results demonstrated that basement membrane organization was disrupted in large microaneuryms. 
Figure 8
 
Analysis of basement membrane protein expression in large retinal microaneurysms. Large microaneurysms showed a decreased expression of collagen IV (A), laminin (B), fibronectin (C), nidogen (D), and perlecan (E) in comparison with small microaneurysms (Fig. 2). Large microaneurysms showed extensive regions of low expression (asterisk) with some small fibrillary areas of high protein concentration (arrowhead). Scale bars: (A) = 26.19 μm; (B) = 19.8 μm; (C) = 13.03 μm; (D) = 28.8 μm; (E) = 20.31 μm.
Figure 8
 
Analysis of basement membrane protein expression in large retinal microaneurysms. Large microaneurysms showed a decreased expression of collagen IV (A), laminin (B), fibronectin (C), nidogen (D), and perlecan (E) in comparison with small microaneurysms (Fig. 2). Large microaneurysms showed extensive regions of low expression (asterisk) with some small fibrillary areas of high protein concentration (arrowhead). Scale bars: (A) = 26.19 μm; (B) = 19.8 μm; (C) = 13.03 μm; (D) = 28.8 μm; (E) = 20.31 μm.
Table 2
 
Quantification of Colocalization Between Collagen IV and the Main Basement Proteins
Table 2
 
Quantification of Colocalization Between Collagen IV and the Main Basement Proteins
Figure 9
 
Analysis of colocalization between nidogen and collagen IV in large retinal microaneurysms. Double immunostaining with antibodies against nidogen (green) and collagen IV (red) was performed in paraffin sections, to analyze protein colocalization by confocal microscopy. Large microaneurysms showed decreased colocalization between nidogen and collagen IV, when compared to adjacent nonaffected capillaries. Nuclei were counterstained with ToPro-3 (blue). *P < 0.05. Scale bar: 35.90 μm.
Figure 9
 
Analysis of colocalization between nidogen and collagen IV in large retinal microaneurysms. Double immunostaining with antibodies against nidogen (green) and collagen IV (red) was performed in paraffin sections, to analyze protein colocalization by confocal microscopy. Large microaneurysms showed decreased colocalization between nidogen and collagen IV, when compared to adjacent nonaffected capillaries. Nuclei were counterstained with ToPro-3 (blue). *P < 0.05. Scale bar: 35.90 μm.
Matrix metalloproteinases are a family of zinc-dependent proteases, which degrade components of the extracellular matrix, such as collagen, laminin, fibronectin, elastin, and other glycoproteins.35 To examine whether MMPs were involved in the decrease and disorganization of basement membrane proteins observed in large microaneurysms, the expression of MMP-9 was evaluated. Matrix metalloproteinase-9 was overexpressed in large microaneurysms but not in small microaneurysms in old human retinas (Fig. 10). 
Figure 10
 
Analysis of MMP-9 expression in retinal microaneurysms. Matrix metalloproteinase-9 was overexpressed in large microaneurysms (A) in comparison with the basal expression observed in small microaneurysms (B). Nuclei were counterstained with ToPro-3 (blue). Scale bars: (A) = 41.73 μm; (B) = 27.31 μm.
Figure 10
 
Analysis of MMP-9 expression in retinal microaneurysms. Matrix metalloproteinase-9 was overexpressed in large microaneurysms (A) in comparison with the basal expression observed in small microaneurysms (B). Nuclei were counterstained with ToPro-3 (blue). Scale bars: (A) = 41.73 μm; (B) = 27.31 μm.
Analysis of PAI-1 Expression in Small and Large Microaneurysms
In 1967, Cogan and Kuwabara proposed the idea that microthrombi could contribute to the development of microaneurysms.36 In our study, 18% of 110 analyzed microaneurysms showed their lumen blocked by thrombi, being the presence of microthrombi in large microneurysms (64%) much more frequent than in small microaneurysms (3%; Fig. 11A). Thus, suggesting that the formation of microthrombi could contribute to the wall dilation of microaneurysms. 
Figure 11
 
Analysis of microthrombi and PAI-1 expression in retinal microaneurysms. (A) The 64% of large microaneurysms are clogged by microthrombi. In contrast, only the 3% of small microaneurysms contains a microthrombus. Images are single confocal laser microscopy sections (B) PAI-1 was overexpressed in the microthrombi localized in large microaneurysms. Nuclei were counterstained with ToPro-3 (blue). Scale bars: (A) = 18 μm (left) and 15.5 μm (right); (B) 10.38 μm (upper) and 41.27 μm (bottom).
Figure 11
 
Analysis of microthrombi and PAI-1 expression in retinal microaneurysms. (A) The 64% of large microaneurysms are clogged by microthrombi. In contrast, only the 3% of small microaneurysms contains a microthrombus. Images are single confocal laser microscopy sections (B) PAI-1 was overexpressed in the microthrombi localized in large microaneurysms. Nuclei were counterstained with ToPro-3 (blue). Scale bars: (A) = 18 μm (left) and 15.5 μm (right); (B) 10.38 μm (upper) and 41.27 μm (bottom).
Plasminogen activator inhibitor-1 is the primary regulator of thrombus fibrinolysis by inhibiting plasminogen activators, a type of serine proteases that activate plasminogen to plasmin.37,38 To explore if the overexpression of PAI-1 may contribute to the dilation of microaneurysms, a double immunostaining was performed in whole mount retinas with specific antibodies anti-collagen IV and anti-PAI-1 (Figs. 11B, 11C). The results obtained showed that PAI-1 was overexpressed in the thrombi of large microaneurysms, thus inhibiting fibrinolysis, and facilitating the expansion of microthrombi and the subsequent dilation of the altered wall of retinal large microaneurysms. 
Pericyte Coverage in Small and Large Microaneurysms
The analysis of microaneurysms by the trypsin digestion method indicated that most of small microaneurysms contained a large number of dense and small nuclei within their wall, while the vascular wall of large microaneurysms remained hypocellular or even acellular (Fig. 12A). Furthermore, a quantitative analysis demonstrated that the number of nuclei in the wall of small microaneurysms (3296 ± 696.7 nuclei/mm2) was significantly higher than in the wall of large microaneurysms (637.5 ± 555.9 cells/mm2; P = 0.0431; Fig. 12B). 
Figure 12
 
Analysis of pericytes in retinal microaneurysms. (A) Small microaneurysms appeared hypercellular in trypsin digested retinal vasculature whereas the wall of large microaneurysms remained acellular. The presence of blood clots is confirmed in large microaneurysms. (B) Differences in cellularity between small and large microaneurysms were statistically significant. (C) Pericytes marked with anti-αSMA antibody were observed in the wall of small microaneurysms. In contrast, they were absent in large microaneurysms. As expected, smooth muscle cells of retinal arterioles (a) expressed αSMA heavily. (D) Pericytes (arrowheads) surrounded completely the wall of small microcaneurysms. Images are single confocal laser microscopy sections. Nuclei were counterstained with ToPro-3 (blue). *P < 0.05. Scale bars: (A) left = 23.94 μm; (A) right = 57 μm; (C) upper = 10.7 μm; (C) lower = 9.49 μm; (D) = 16.29 μm.
Figure 12
 
Analysis of pericytes in retinal microaneurysms. (A) Small microaneurysms appeared hypercellular in trypsin digested retinal vasculature whereas the wall of large microaneurysms remained acellular. The presence of blood clots is confirmed in large microaneurysms. (B) Differences in cellularity between small and large microaneurysms were statistically significant. (C) Pericytes marked with anti-αSMA antibody were observed in the wall of small microaneurysms. In contrast, they were absent in large microaneurysms. As expected, smooth muscle cells of retinal arterioles (a) expressed αSMA heavily. (D) Pericytes (arrowheads) surrounded completely the wall of small microcaneurysms. Images are single confocal laser microscopy sections. Nuclei were counterstained with ToPro-3 (blue). *P < 0.05. Scale bars: (A) left = 23.94 μm; (A) right = 57 μm; (C) upper = 10.7 μm; (C) lower = 9.49 μm; (D) = 16.29 μm.
Strong α-SMA reactivity, a validated pericyte biomarker,39 was observed in small but not in large microaneurysms (Fig. 12C). Pericytes completely surrounded the wall of small microaneurysms (Fig. 12D). Thus, altogether, our results indicated an increase in pericyte coverage of small microaneurysms while pericyte decline is observed in large microaneurysms. 
To unveil if pericytes located in small microaneurysms appear as a consequence of cellular division from preexisting capillary pericytes, the cellular proliferative marker Ki-6740 was tested in whole mount retinas (Fig. 13). Cellular proliferation was not observed in the wall of microaneurysms, despite that the anti-Ki-67 antibody worked in the same experiment staining the nuclei of the proliferating basal cells in the human epidermis (Fig. 13). These results suggested that pericytes could be recruited during the formation of small microaneurysms. Pericyte recruitment stimulates basement membrane matrix formation by endothelial cells41 and pericytes themselves produce collagen IV and laminin,42 which altogether could explain the increased expression of basement membrane proteins, along with basement membrane thickening, observed in small microaneurysms. Furthermore, pericytes stabilize the capillary vessel wall.43 Thus, the lack of pericytes observed in large microaneurysms could be a mechanism that explains capillary weakening and consequent wall dilation, which is observed in large microaneurysms. 
Figure 13
 
Analysis of cellular proliferation in retinal microaneurysms. Cellular division was not observed in the wall of small microaneurysms (A) and large microaneurysms (B) despite that the Ki-67 marker worked in the same experiment staining the nuclei (arrowheads) of proliferating basal cells in the human epidermis (C). Images are single confocal laser microscopy sections. Nuclei were counterstained with ToPro-3 (blue). Scale bars: (A) = 12.96 μm; (B) = 25.31 μm; (C) = 31.58 μm.
Figure 13
 
Analysis of cellular proliferation in retinal microaneurysms. Cellular division was not observed in the wall of small microaneurysms (A) and large microaneurysms (B) despite that the Ki-67 marker worked in the same experiment staining the nuclei (arrowheads) of proliferating basal cells in the human epidermis (C). Images are single confocal laser microscopy sections. Nuclei were counterstained with ToPro-3 (blue). Scale bars: (A) = 12.96 μm; (B) = 25.31 μm; (C) = 31.58 μm.
Discussion
Microaneurysms are considered a hallmark of diabetic retinopathy.21,4446 However, the impact of aging as a determinant factor in the development of retinal microaneurysms is not well understood. In this study, we observed few microaneurysms in all the retinas obtained from old aged donors, a feature not found in middle aged donors. This is in accordance with previous studies showing that the presence of human retinal microaneurysms was age-dependent.47,48 However, this fact is not unanimously accepted, because other previously reported clinical data4750 indicated that the percentage of old people with retinal microaneurysms was low, ranging from 6% to 36%. Differences between our data and other studies could be due to the number of individuals in the study groups. Clinical reports generally have a large number of patients, in contrast with our study, where only a relative small number of microaneurysms (110) from 14 old age donors were analyzed. Furthermore, age differences between study groups also may contribute to explain some of these inconsistencies. In this regard, whereas Sugi et al.47 include people older than 60 in their study group, our analysis only includes people older than 72 years. Methodologic reasons also could be a cause for discrepancies in the results observed. Clinical studies in older people used detailed grading high quality retinal photographs49,50 and even digital images taken directly from the fundus of both eyes.48 However, we have examined retinas by using confocal laser scanner microscopy, a technical approach with higher spatial resolution than that of ophthalmoscopy. Recent advances in retinal imaging, such as scanning light ophthalmoscopy,51 have increased the ability to image retinal vasculature, including microaneurysms. Under this perspective, in a near future, it could be possible to find a higher concordance between results obtained in clinical and histologic postmortem studies. However, at the present, the limited number of participants analyzed in our study, and the few number of preexisting reports that support our data, do not allow to consider that the prevalence of retinal microaneurysms is increased in the elderly. 
Since microaneurysms do not generally occur in rodent models, there have been relatively few histologic or ultrastructural studies of these lesions, and it is one of the reasons why the origin and pathogeny of microaneurysms remains incomplete and fragmentary.44 Furthermore, contradictory observations in the capillary wall have been reported during microaneurysm formation. Some investigators reported a narrowing in the capillary wall,52 while others evidenced a thickening in the capillary wall.53,54 Our study in small microaneurysms supported the idea that basement membrane thickening coincides with the early stage of microaneurysm formation. Furthermore, we described for the first time to our knowledge that basement membrane thickening in small microaneurysms is associated with an increased expression of collagen IV, laminin, fibronectin, nidogen, and perlecan, the main structural and adhesive components of the retinal capillary basement membrane. Unexpectedly, collagen III is expressed strongly in the basement membrane of small microaneurysms. This is a noteworthy event, since it has been observed previously that in normal conditions, collagen III is not present in retinal capillaries and its expression is restricted to larger retinal vessels, such as the arterioles and venules.32 Transmission electron microscopy showed that collagen III is forming fibrils in microaneurysms, the so-called reticular fibers. This result reinforces the observation made 40 years ago by Ashton and Tripathi, who showed argyrophilic reticular fibers encircling microaneurysms during diabetic retinopathy.55 Type III collagen fibrils are more abundant in tissues subjected to periodic stress, such as vasculature, confirming the role of collagen III in increasing vascular wall stiffness. In this regard, mice deficient for collagen III die perinatally presenting rupture of major blood vessels.56 
Collagen fibrils were found crosslinked in small microaneurysms during aging, a process that also takes place during diabetes.57 Advanced glycation end-products are the key players implicated in the nonenzymatic diabetic crosslinking of collagen in the vascular wall, which leads to basement membrane thickening and loss of vascular elasticity.57 However, in the absence of hyperglycemia, the collagen crosslinking should be enzymatic and mediated by lysyl oxidases. Our results confirmed this possibility and showed that LOXL2 and LOXL4 were overexpressed in microaneurysms during aging in nondiabetic conditions. Similar to our results, Akagawa et al.58 have reported an increased gene expression of LOXL2 in human intracranial aneurysms. In addition, saccular aneurysms in the thoracic aorta have been observed in mice lacking LOX.59 
In comparison with small microaneurysms, basement membrane composition and organization were altered in large microaneurysms, as showed by a diminution in protein expression and colocalization. The basement membrane disorganization in large microaneurysms was parallel to a change in the expression pattern of MMP-9, a MMP specifically associated with the formation of aortic aneurysms.60 The MMP-9 patchy expression pattern observed in large microaneurysms could explain the heterogeneous disorganization of their basement membrane and pointed to MMP-9 as player involved in basement membrane loss of stiffness and, consequently, a risk factor for microaneurysm dilation in aged retinas. 
Perhaps the strongest clinical evidence that microaneurysm formation is related to intravascular coagulation is the observation that aspirin, a recognized anticlotting agent, reduced the number of microaneurysms in diabetic patients.61,62 In our study, more than half (64%) of large microaneurysms have a microthrombus blocking their lumen, in contrast with only 3% of small microaneurysms that contained a microthrombus. Furthermore, PAI-1 was overexpressed in the microthrombi of large microaneurysms, which could facilitate the expansion of microthrombi and subsequent wall dilation observed in large microaneurysms. Interestingly, PAI-1 also is overexpressed during nonproliferative diabetic retinopathy,63 the phase of diabetic retinopathy where microaneurysms appear. 
Mechanisms of retinal microaneurysm formation during diabetes include focal endothelial proliferation, which gives rise to hypercellular microaneurysms.64,65 However, our results showed that during retinal aging, proliferative changes were not observed in small or in large microaneurysms. Even in small microaneurysms, where pericyte hypercellularity was detected, cellular proliferation using Ki-67 marker was not observed. Thus, these data could suggest that, unlike what happens in diabetic retinopathy, cellular proliferation is not a common mechanism involved in the formation of retinal microaneurysms during aging. 
Altogether, our data strongly suggested that basement membrane thickening in small microaneurysms by accumulation of collagen IV, laminin, fibronectin, nidogen, and perlecan, besides the appearance of crosslinked collagen III fibrils, probably due to the action of LOXL2 and LOXL4, could be considered as a compensatory mechanism to strengthen the vascular wall in the early phase of microaneurysm formation. Furthermore, pericyte hypercellularity observed in small microaneurysm also may contribute during early stages as a compensatory mechanism to delay the progression of microaneurysm formation. Later, in the long-term, the capacity of compensatory mechanisms could be exceeded leading to pericyte decline and increased activity of MMP-9 and PAI-I, which produce disruption of the blood basement membrane and expansion of microthrombi. These mechanisms would lead to wall weakening, thus explaining the wall dilation observed in the late phase of microaneurysm formation in the elderly. Nevertheless, we cannot rule out the possibility that, due to the uneven amount and organization of the basement membrane (increased and organized versus decreased and disorganized) and the different cellular wall composition (hypercellular versus hypocellular), small and large microaneurysms constitute two distinct types of microaneurysms and not two consecutive phases during the formation of a single microaneurysm. 
Acknowledgments
The authors thank Veronica Melgarejo, Lorena Noya, and Angel Vazquez for technical assistance. 
Supported by grants from Instituto de Salud Carlos III (PI16/00719 and PI12/00605), Ministerio de Ciencia e Innovacion, Spain and from Fundação para a Ciência e a Tecnologia (SFRH/BPD/102573/2014), Ministerio da Educação e Ciência, Portugal. 
Disclosure: M. López-Luppo, None; V. Nacher, None; D. Ramos, None; J. Catita, None; M. Navarro, None; A. Carretero, None; A. Rodriguez-Baeza, None; L. Mendes-Jorge, None; J. Ruberte, None 
References
Stitt AW, Gardiner TA, Archer DB. Histological and ultrastructural investigation of retinal microaneurysm development in diabetic patients. Br J Ophthalmol. 1995; 79: 362–367.
Moore J, Bagley S, Ireland G, McLeod D, Boulton ME. Three dimensional analysis of microaneurysms in the human diabetic retina. J Anat. 1999; 194: 89–100.
Ballantyne AJ, Loewenstein A. The pathology of diabetic retinopathy. Trans Ophthal Soc UK. 1943; 63: 95–115.
Wise GN. Retinal microaneurysms. AMA Arch Ophthalmol. 1956; 151–156.
Catita J, López-Luppo M, Ramos D, et al. Imaging of cellular aging in human retinal blood vessels. Exp Eye Res. 2015; 135: 14–25.
Klein R. Retinopathy in a population-based study. Trans Am Ophthalmol Soc. 1992; 90: 561–594.
Klein R, Klein BE, Moss SE, Wang Q. Blood pressure, hypertension and retinopathy in a population. Trans Am Ophthalmol Soc. 1993; 91: 207–226.
Klein R, Klein BE, Moss SE, Wang Q. Hypertension and retinopathy, arteriolar narrowing, and arteriovenous nicking in a population. Arch Ophthalmol. 1994; 112: 92–98.
Yu T, Mitchell P, Berry G, Li W, Wang JJ. Retinopathy in older persons without diabetes and its relationship to hypertension. Arch Ophthalmol. 1998; 116: 83–89.
Ashton N. Retinal micro-aneurysms in the non-diabetic subject. Br J Ophthalmol. 1951; 35: 189–212.
Cogan DG. Development and senescence of the human retinal vasculature. Trans Ophthalmol Soc UK. 1963; 83: 465–489.
Speiser P, Gittelsohn AM, Patz A. Studies on diabetic retinopathy III. Influence of diabetes on intramural pericytes. Arch Ophthalmol. 1968; 80: 332–337.
Wise GN. Arteriosclerosis secondary to retinal veins obstruction. Trans Am Ophthalmol Soc. 1958; 56: 361–380.
Yanoff M, Fine BS. Ocular Pathology: A Text and Atlas. New York: Harper and Row; 1975: 566–585.
Neuenschwander H, Julia C, Wyman M, Kador PF. Endothelial changes in galactose-fed dogs. Curr Eye Res. 1995; 14: 319–322.
van Eeden PE, Tee LB, Lukehurst S, et al. Early vascular and neuronal changes in a VEGF transgenic mouse model of retinal neovascularization. Invest Ophthalmol Vis Sci. 2006; 47: 4638–4645.
Little HL, Sacks AH. Role of abnormal blood rheology in the pathogenesis of diabetic retinopathy. Trans Sect Ophthalmol Am Acad Ophthalmol Otolaryngol. 1977; 83: 522–534.
Little HL. Alterations in blood elements in the pathogenesis of diabetic retinopathy. Ophthalmolology. 1981; 88: 647–654.
Ezra E, Keinan E, Mandel Y, Boulton ME, Nahmias Y. Non-dimensional analysis of retinal microaneurysms: critical threshold for treatment. Integr Biol. 2013; 5: 474–480.
Frank RN. On the pathogenesis of diabetic retinopathy. Ophthalmology. 1984; 91: 626–634.
Dubow M, Pinhas A, Shah N, et al. Classification of human retinal microaneurysms using adaptive optics scanning light ophthalmoscope fluorescein angiography. Invest Ophthalmol Vis Sci. 2014; 55: 1299–1309.
Manders EMM, Verbeek FJ, Aten JA. Measurement of co-localization of objects in dual-color confocal images. J Microscopy. 1993; 169: 375–382.
Chou JC, Rollins SD, Fawzi AA. Trypsin digest protocol to analyze the retinal vasculature of a mouse model. J Vis Exp. 2013; 76: e50489.
Baluk P, Morikawa S, Haskell A, Mancuso M, McDonald DM. Abnormalities of basement membrane on blood vessels and endothelial sprouts in tumors. Am J Pathol. 2003; 163: 1801–1815.
Inoue S. Ultrastructure of basement membranes. Int Rev Cytol. 1989; 117: 57–98.
Timpl R, Wiedemann H, vanDelden V, Furthmayr H, Kühn K. A network model for the organization of type IV collagen molecules in basement membranes. Eur J Biochem. 1981; 120: 203–211.
Hohenester E, Yurchenco PD. Laminins in basement membrane assembly. Cell Adh Migr. 2013; 7: 56–63.
Hayden MR, Sowers JR, Tyagi SC. The central role of vascular extracellular matrix and basement membrane remodeling in metabolic syndrome and type 2 diabetes: the matrix preloaded. Cardiovasc Diabetol. 2005; 4: 9.
Liesi P, Kauppila T. Induction of type IV collagen and other basement-membrane-associated proteins after spinal cord injury of the adult rat may participate in formation of the glial scar. Exp Neurol. 2002; 173: 31–45.
Sofroniew MV, Vinters HV. Astrocytes: biology and pathology. Acta Neuropathol. 2010; 119: 7–35.
Bailey AJ, Paul RG, Knott L. Mechanisms of maturation and ageing of collagen. Mech Ageing Dev. 1998; 106: 1–56.
Jerdan JA, Glaser BM. Retinal microvessel extracellular matrix: an immunofluorescent study. Invest Ophthalmol Vis Sci. 1986; 27: 194–203.
Smith-Mungo LI, Kagan HM. Lysyl oxidase: properties, regulation and multiple functions in biology. Matrix Biol. 1998; 16: 387–398.
Kagan HM, Li W. Lysyl oxidase: properties, specificity, and biological roles inside and outside of the cell. J Cell Biochem. 2003; 88: 660–672.
Woessner JFJr. Matrix metalloproteinases and their inhibitors in connective tissue remodeling. FASEB J. 1991; 5: 2145–2154.
Cogan DG, Kuwabara T. The mural cell in perspective. Arch Ophthalmol. 1967; 78: 133–139.
Saksela O, Rifkin DB. Cell-associated plasminogen activation: regulation and physiological functions. Annu Rev Cell Biol. 1988; 4: 93–126.
Pöllänen J, Stephens RW, Vaheri A. Directed plasminogen activation at the surface of normal and malignant cells. Adv Cancer Res. 1991; 57: 273–328.
Nehls V, Drenckhahn D. The versatility of microvascular pericytes: from mesenchyme to smooth muscle? Histochemistry. 1993; 99: 1–12.
Scholzen T, Gerdes J. The Ki-67 protein: from the known and the unknown. J Cell Physiol. 2000; 182: 311–322.
Stratman AN, Malotte KM, Mahan RD, Davis MJ, Davis GE. Pericyte recruitment during vasculogenic tube assembly stimulates endothelial basement membrane matrix formation. Blood. 2009; 114: 5091–5101.
Jeon H, Ono M, Kumagai C, Miki K, Morita A, Kitagawa Y. Pericytes from microvessel fragments produce type IV collagen and multiple laminin isoforms. Biosci Biotech Biochem. 1996; 60: 856–861.
Armulik A, Genové G, Betsholtz C. Pericytes: developmental, physiological and pathological perspectives, problems, and promises. Dev Cell. 2011; 21: 193–215.
Gardiner TA, Archer DB, Curtis TM, Stitt AW. Arteriolar involvement in the microvascular lesions of diabetic retinopathy: implications for pathogenesis. Microcirculation. 2007; 14: 25–38.
Curtis TM, Gardiner TA, Stitt AW. Microvascular lesions of diabetic retinopathy: clues towards understanding pathogenesis? Eye (Lond). 2009; 23: 1496–1508.
Stitt AW, Curtis TM, Chen M, et al. The progress in understanding and treatment of diabetic retinopathy. Prog Retin Eye Res. 2016; 51: 156–186.
Sugi K. Studies on the pathological changes in the retinal vessels of human eyes, using the trypsin digestion method. Jpn J Ophthalmol. 1966; 10: 252–266.
Qiu C, Cotch MF, Sigurdsson S, et al. Microvascular lesions in the brain and retina: the age, gene/environment susceptibility–Reykjavik study. Ann Neurol. 2009; 65: 569–576.
Yu T, Mitchell P, Berry G, Li W, Wang JJ. Retinopathy in older persons without diabetes and its relationship to hypertension. Arch Ophthalmol. 1998; 116: 83–89.
Wong TY, Klein R, Klein BEK, Tielsch JM, Hubbard L, Nieto FJ. Retinal microvascular abnormalities and their relationship with hypertension, cardiovascular disease, and mortality. Surv Ophthalmol. 2001; 46: 59–80.
Chui TYP, Mo S, Krawitz B, et al. Human retinal microvascular imaging using adaptive optics scanning light ophthalmoscopy. Int J Retina Vitreous. 2016; 2: 11.
Pope CHJr. Retinal capillary microaneurysms: a concept of pathogenesis. Diabetes. 1960; 9: 9–13.
Toussaint D. Aspects histologiques de la rétinopathie diabétique. Bull Soc Belge Ophthalmol. 1961; 129: 428–444.
Bloodworth JMJr, Molitor DL. Ultrastructural aspects of human and canine diabetic retinopathy. Invest Ophthalmol. 1965; 4: 1037–1048.
Ashton N, Tripathi R. Perivascular and intervascular reticular fibers of the retina. Am J Ophthalmol. 1975; 80: 337–350.
Arteaga-Solis E, Gayraud B, Ramirez F. Elastic and collagenous networks in vascular diseases. Cell Struct Funct. 2000; 25: 69–72.
Ulrich P, Cerami A. Protein glycation, diabetes, and aging. Recent Prog Horm Res. 2001; 56: 1–21.
Akagawa H, Narita A, Yamada H, et al. Systematic screening of lysyl oxidase-like (LOXL) family genes demonstrates that LOXL2 is a susceptibility gene to intracranial aneurysms. Hum Genet. 2007; 121: 377–387.
Mäki JM, Räsänen J, Tikkanen H, et al. Inactivation of the lysyl oxidase gene Lox leads to aortic aneurysms, cardiovascular dysfunction, and perinatal death in mice. Circulation. 2002; 106: 2503–2509.
Bunton TE, Biery NJ, Myers L, Gayraud B, Ramirez F, Dietz HC. Phenotypic alteration of vascular smooth muscle cells precedes elastolysis in a mouse model of Marfan syndrome. Circ Res. 2001; 88: 37–43.
The DAMAD Study Group. Effect of aspirin alone and aspirin plus dipyridamole in early diabetic retinopathy. A multicenter randomized controlled clinical trial. Diabetes. 1989; 38: 491–498.
Knott RM, Forrester JV. Pathogenesis of diabetic retinopathy and cataract. In: Pickup JC, Williams G, eds. Textbook of Diabetes, 2nd ed. Oxford: Blackwell Science; 1997: 45.1–45.19.
Grant MB, Ellis EA, Caballero S, Mames RN. Plasminogen activator inhibitor-1 overexpression in nonproliferative diabetic retinopathy. Exp Eye Res. 1996; 63: 233–244.
Frank RN. Etiologic mechanisms in diabetic retinopathy. In: Ryan SJ, ed. Retina. St. Louis: Mosby; 2001: 1259–1284.
Aguilar E, Friedlander M, Gariano RF. Endothelial proliferation in diabetic retinal microaneurysms. Arch Ophthalmol. 2003; 121: 740–741.
Figure 1
 
Identification and classification of retinal microaneurysms. (A) Microaneurysms were identified as focal outpouchings of retinal capillaries in whole mount retinas immunohistochemically marked with antibodies against the blood basement membrane. A microaneurysm was classified as “small” when the combined width of the focal bulge and the associated capillary was less than twice the width of an adjacent nonaffected capillary region, and “large” when the combined width of the vascular dilation and the associated capillary was more than twice the width of an adjacent nonaltered capillary. (B) Comparison between a normal retinal topography without microaneurysms from a middle-aged donor and a disorganized retina with the presence of a large microaneurysms from an old aged donor. SM, small microaneurysm; LM, large microaneurysm; GCL, ganglion cell layer; INL, inner nuclear layer; ONL, outer nuclear layer. Nuclei were counterstained with ToPro-3 (blue). Scale bars: (A) = 34.2 μm; (B) = 30.68 μm; (C) = 28.26 μm.
Figure 1
 
Identification and classification of retinal microaneurysms. (A) Microaneurysms were identified as focal outpouchings of retinal capillaries in whole mount retinas immunohistochemically marked with antibodies against the blood basement membrane. A microaneurysm was classified as “small” when the combined width of the focal bulge and the associated capillary was less than twice the width of an adjacent nonaffected capillary region, and “large” when the combined width of the vascular dilation and the associated capillary was more than twice the width of an adjacent nonaltered capillary. (B) Comparison between a normal retinal topography without microaneurysms from a middle-aged donor and a disorganized retina with the presence of a large microaneurysms from an old aged donor. SM, small microaneurysm; LM, large microaneurysm; GCL, ganglion cell layer; INL, inner nuclear layer; ONL, outer nuclear layer. Nuclei were counterstained with ToPro-3 (blue). Scale bars: (A) = 34.2 μm; (B) = 30.68 μm; (C) = 28.26 μm.
Figure 2
 
Analysis of basement membrane protein expression in small retinal microaneurysms. Small microaneurysms showed an increased expression of collagen IV (A), laminin (B), fibronectin (C), nidogen (D), and perlecan (E). IP, intensity profile from a maximum 2D projection obtained by making optical sections every micrometer through the entire microaneurysm length. Scale bars: (A) = 22.46 μm; (B) = 6.94 μm; (C) = 10.99 μm; (D) = 25.5 μm; (E) = 15.17 μm.
Figure 2
 
Analysis of basement membrane protein expression in small retinal microaneurysms. Small microaneurysms showed an increased expression of collagen IV (A), laminin (B), fibronectin (C), nidogen (D), and perlecan (E). IP, intensity profile from a maximum 2D projection obtained by making optical sections every micrometer through the entire microaneurysm length. Scale bars: (A) = 22.46 μm; (B) = 6.94 μm; (C) = 10.99 μm; (D) = 25.5 μm; (E) = 15.17 μm.
Figure 3
 
Ultrastructural analysis of the basement membrane (BM) in small retinal microaneurysms. The comparison between the vessel walls of a nonaffected capillary (A) and a small microaneurysm (B) was assessed by transmission electron microscopy. Magnification evidenced thickening of the basement membrane in small microaneurysms. As, astrocyte; EC, endothelial cell; Er, erythrocyte; Lu, lumen; Mü, Müller cell; Pe, pericyte. Scale bars: (A) = 9.97 μm; (B) = 7.13 μm.
Figure 3
 
Ultrastructural analysis of the basement membrane (BM) in small retinal microaneurysms. The comparison between the vessel walls of a nonaffected capillary (A) and a small microaneurysm (B) was assessed by transmission electron microscopy. Magnification evidenced thickening of the basement membrane in small microaneurysms. As, astrocyte; EC, endothelial cell; Er, erythrocyte; Lu, lumen; Mü, Müller cell; Pe, pericyte. Scale bars: (A) = 9.97 μm; (B) = 7.13 μm.
Figure 4
 
Analysis of colocalization between GFAP and collagen IV in small retinal microaneurysms. Collagen IV (red) in small microaneurysms did not colocalize with GFAP (green), suggesting that the overexpression of collagen IV in small microaneurysms is not related to formation of a glial scar. As expected, astrocytic end-feet processes marked by GFAP were in contact with the basement membrane of retinal capillaries. Furthermore, nuclei were counterstained with ToPro-3 (blue). Arrowhead, end-feed processes. Scale bar: 18.87 μm.
Figure 4
 
Analysis of colocalization between GFAP and collagen IV in small retinal microaneurysms. Collagen IV (red) in small microaneurysms did not colocalize with GFAP (green), suggesting that the overexpression of collagen IV in small microaneurysms is not related to formation of a glial scar. As expected, astrocytic end-feet processes marked by GFAP were in contact with the basement membrane of retinal capillaries. Furthermore, nuclei were counterstained with ToPro-3 (blue). Arrowhead, end-feed processes. Scale bar: 18.87 μm.
Figure 5
 
Anomalous expression of collagen III in small retinal microaneurysms. (A) Laser confocal analysis of whole mount retinas evidenced the presence of collagen III in small microaneurysms. Adjacent nonaltered capillaries did not express collagen III. (B) Transmission electron microscopy revealed the existence of fibrils (arrows) in the thickened basement membrane of small microaneurysms. Fibrils have periodic cross-striations every 67 nm, which is compatible with the morphology of collagen III fibrils. In some regions, fibrils were crosslinked (circle). Healthy capillaries do not exhibit collagen fibrils in the basement membrane, presenting the amorphous characteristic aspect. Scale bars: (A) = 14.19 μm; (B) = 0.53 μm.
Figure 5
 
Anomalous expression of collagen III in small retinal microaneurysms. (A) Laser confocal analysis of whole mount retinas evidenced the presence of collagen III in small microaneurysms. Adjacent nonaltered capillaries did not express collagen III. (B) Transmission electron microscopy revealed the existence of fibrils (arrows) in the thickened basement membrane of small microaneurysms. Fibrils have periodic cross-striations every 67 nm, which is compatible with the morphology of collagen III fibrils. In some regions, fibrils were crosslinked (circle). Healthy capillaries do not exhibit collagen fibrils in the basement membrane, presenting the amorphous characteristic aspect. Scale bars: (A) = 14.19 μm; (B) = 0.53 μm.
Figure 6
 
Analysis of LOXL2 expression in old aged retinas. Immunodetection of LOXL2 (green) was observed in all the blood vessels (arrowheads), in whole mount (A), and paraffin sections (B), of human retinas. Few cells (arrows) in the inner nuclear layer also express LOXL2. Collagen IV (red) colocalized with LOXL2, suggesting that this lysil oxidase-like was activated. (D) In small microaneurysms, LOXL2 was overexpressed, compared to the low expression exhibited by the adjacent nonaffected capillary. a, arteriole. Nuclei were counterstained with ToPro-3 (blue). Scale bars: (A) = 122.92 μm; (B) = 29.17 μm; (C) = 14.67 μm.
Figure 6
 
Analysis of LOXL2 expression in old aged retinas. Immunodetection of LOXL2 (green) was observed in all the blood vessels (arrowheads), in whole mount (A), and paraffin sections (B), of human retinas. Few cells (arrows) in the inner nuclear layer also express LOXL2. Collagen IV (red) colocalized with LOXL2, suggesting that this lysil oxidase-like was activated. (D) In small microaneurysms, LOXL2 was overexpressed, compared to the low expression exhibited by the adjacent nonaffected capillary. a, arteriole. Nuclei were counterstained with ToPro-3 (blue). Scale bars: (A) = 122.92 μm; (B) = 29.17 μm; (C) = 14.67 μm.
Figure 7
 
Analysis of LOXL4 expression in old aged retinas. Immunodetection of LOXL4 (green) was observed exclusively in the blood vessels (arrowheads), in whole mount (A) and paraffin sections (B), of human retinas. Collagen IV (red) colocalized with LOXL4, suggesting that this lysil oxidase-like was activated. (D) In small microaneurysms, LOXL4 was overexpressed. Nuclei were counterstained with ToPro-3 (blue). Scale bars: (A) 187.35 μm; (B) = 25.78 μm; (C) = 13.83 μm.
Figure 7
 
Analysis of LOXL4 expression in old aged retinas. Immunodetection of LOXL4 (green) was observed exclusively in the blood vessels (arrowheads), in whole mount (A) and paraffin sections (B), of human retinas. Collagen IV (red) colocalized with LOXL4, suggesting that this lysil oxidase-like was activated. (D) In small microaneurysms, LOXL4 was overexpressed. Nuclei were counterstained with ToPro-3 (blue). Scale bars: (A) 187.35 μm; (B) = 25.78 μm; (C) = 13.83 μm.
Figure 8
 
Analysis of basement membrane protein expression in large retinal microaneurysms. Large microaneurysms showed a decreased expression of collagen IV (A), laminin (B), fibronectin (C), nidogen (D), and perlecan (E) in comparison with small microaneurysms (Fig. 2). Large microaneurysms showed extensive regions of low expression (asterisk) with some small fibrillary areas of high protein concentration (arrowhead). Scale bars: (A) = 26.19 μm; (B) = 19.8 μm; (C) = 13.03 μm; (D) = 28.8 μm; (E) = 20.31 μm.
Figure 8
 
Analysis of basement membrane protein expression in large retinal microaneurysms. Large microaneurysms showed a decreased expression of collagen IV (A), laminin (B), fibronectin (C), nidogen (D), and perlecan (E) in comparison with small microaneurysms (Fig. 2). Large microaneurysms showed extensive regions of low expression (asterisk) with some small fibrillary areas of high protein concentration (arrowhead). Scale bars: (A) = 26.19 μm; (B) = 19.8 μm; (C) = 13.03 μm; (D) = 28.8 μm; (E) = 20.31 μm.
Figure 9
 
Analysis of colocalization between nidogen and collagen IV in large retinal microaneurysms. Double immunostaining with antibodies against nidogen (green) and collagen IV (red) was performed in paraffin sections, to analyze protein colocalization by confocal microscopy. Large microaneurysms showed decreased colocalization between nidogen and collagen IV, when compared to adjacent nonaffected capillaries. Nuclei were counterstained with ToPro-3 (blue). *P < 0.05. Scale bar: 35.90 μm.
Figure 9
 
Analysis of colocalization between nidogen and collagen IV in large retinal microaneurysms. Double immunostaining with antibodies against nidogen (green) and collagen IV (red) was performed in paraffin sections, to analyze protein colocalization by confocal microscopy. Large microaneurysms showed decreased colocalization between nidogen and collagen IV, when compared to adjacent nonaffected capillaries. Nuclei were counterstained with ToPro-3 (blue). *P < 0.05. Scale bar: 35.90 μm.
Figure 10
 
Analysis of MMP-9 expression in retinal microaneurysms. Matrix metalloproteinase-9 was overexpressed in large microaneurysms (A) in comparison with the basal expression observed in small microaneurysms (B). Nuclei were counterstained with ToPro-3 (blue). Scale bars: (A) = 41.73 μm; (B) = 27.31 μm.
Figure 10
 
Analysis of MMP-9 expression in retinal microaneurysms. Matrix metalloproteinase-9 was overexpressed in large microaneurysms (A) in comparison with the basal expression observed in small microaneurysms (B). Nuclei were counterstained with ToPro-3 (blue). Scale bars: (A) = 41.73 μm; (B) = 27.31 μm.
Figure 11
 
Analysis of microthrombi and PAI-1 expression in retinal microaneurysms. (A) The 64% of large microaneurysms are clogged by microthrombi. In contrast, only the 3% of small microaneurysms contains a microthrombus. Images are single confocal laser microscopy sections (B) PAI-1 was overexpressed in the microthrombi localized in large microaneurysms. Nuclei were counterstained with ToPro-3 (blue). Scale bars: (A) = 18 μm (left) and 15.5 μm (right); (B) 10.38 μm (upper) and 41.27 μm (bottom).
Figure 11
 
Analysis of microthrombi and PAI-1 expression in retinal microaneurysms. (A) The 64% of large microaneurysms are clogged by microthrombi. In contrast, only the 3% of small microaneurysms contains a microthrombus. Images are single confocal laser microscopy sections (B) PAI-1 was overexpressed in the microthrombi localized in large microaneurysms. Nuclei were counterstained with ToPro-3 (blue). Scale bars: (A) = 18 μm (left) and 15.5 μm (right); (B) 10.38 μm (upper) and 41.27 μm (bottom).
Figure 12
 
Analysis of pericytes in retinal microaneurysms. (A) Small microaneurysms appeared hypercellular in trypsin digested retinal vasculature whereas the wall of large microaneurysms remained acellular. The presence of blood clots is confirmed in large microaneurysms. (B) Differences in cellularity between small and large microaneurysms were statistically significant. (C) Pericytes marked with anti-αSMA antibody were observed in the wall of small microaneurysms. In contrast, they were absent in large microaneurysms. As expected, smooth muscle cells of retinal arterioles (a) expressed αSMA heavily. (D) Pericytes (arrowheads) surrounded completely the wall of small microcaneurysms. Images are single confocal laser microscopy sections. Nuclei were counterstained with ToPro-3 (blue). *P < 0.05. Scale bars: (A) left = 23.94 μm; (A) right = 57 μm; (C) upper = 10.7 μm; (C) lower = 9.49 μm; (D) = 16.29 μm.
Figure 12
 
Analysis of pericytes in retinal microaneurysms. (A) Small microaneurysms appeared hypercellular in trypsin digested retinal vasculature whereas the wall of large microaneurysms remained acellular. The presence of blood clots is confirmed in large microaneurysms. (B) Differences in cellularity between small and large microaneurysms were statistically significant. (C) Pericytes marked with anti-αSMA antibody were observed in the wall of small microaneurysms. In contrast, they were absent in large microaneurysms. As expected, smooth muscle cells of retinal arterioles (a) expressed αSMA heavily. (D) Pericytes (arrowheads) surrounded completely the wall of small microcaneurysms. Images are single confocal laser microscopy sections. Nuclei were counterstained with ToPro-3 (blue). *P < 0.05. Scale bars: (A) left = 23.94 μm; (A) right = 57 μm; (C) upper = 10.7 μm; (C) lower = 9.49 μm; (D) = 16.29 μm.
Figure 13
 
Analysis of cellular proliferation in retinal microaneurysms. Cellular division was not observed in the wall of small microaneurysms (A) and large microaneurysms (B) despite that the Ki-67 marker worked in the same experiment staining the nuclei (arrowheads) of proliferating basal cells in the human epidermis (C). Images are single confocal laser microscopy sections. Nuclei were counterstained with ToPro-3 (blue). Scale bars: (A) = 12.96 μm; (B) = 25.31 μm; (C) = 31.58 μm.
Figure 13
 
Analysis of cellular proliferation in retinal microaneurysms. Cellular division was not observed in the wall of small microaneurysms (A) and large microaneurysms (B) despite that the Ki-67 marker worked in the same experiment staining the nuclei (arrowheads) of proliferating basal cells in the human epidermis (C). Images are single confocal laser microscopy sections. Nuclei were counterstained with ToPro-3 (blue). Scale bars: (A) = 12.96 μm; (B) = 25.31 μm; (C) = 31.58 μm.
Table 1
 
Sex, Age, Cause of Death, and Time Between Death and Eye Fixation
Table 1
 
Sex, Age, Cause of Death, and Time Between Death and Eye Fixation
Table 2
 
Quantification of Colocalization Between Collagen IV and the Main Basement Proteins
Table 2
 
Quantification of Colocalization Between Collagen IV and the Main Basement Proteins
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×