|
|
||||||||
1From the Howard Hughes Medical Institute, the 2Division of Epidemiology and Clinical Research, the 3Laboratory of Immunology, and the 5Office of the Scientific Director, National Eye Institute, National Institutes of Health, Bethesda, Maryland; and the 4Retina Group of Washington, Rockville, Maryland.
| Abstract |
|---|
|
|
|---|
METHODS. The following serum factors of 93 subjects were examined at the National Eye Institute (NEI) clinical center: the chemokines regulated on activation, normal T-cell expressed and presumably secreted (RANTES)/CCL5, epithelial neutrophil activator (ENA)-78/CXCL5, interferon-induced protein (IP)-10/CXCL10, stromal cellderived factor (SDF)-1
/CXCLl2, monocyte chemoattractant protein (MCP)-1/CCL2, macrophage inflammatory protein (MIP)-1
/CCL3, interleukin (IL)-8/CXCL8; the cytokine IL-6; the cell adhesion molecules intercellular adhesion molecule (ICAM-1/CD54) and vascular cell adhesion molecule (VCAM/CD106); and the growth factor vascular endothelial growth factor (VEGF). Logistic regression was performed to assess the association of these factors with age, sex, severity of retinopathy, hemoglobin A1C, total cholesterol, creatinine, duration of diabetes, and presence of macular edema. The outcome assessed was severity of retinopathy. Frozen sections of two donor eyes obtained at autopsy from a donor with documented severe nonproliferative diabetic retinopathy and diabetic macular edema and of a normal nondiabetic eye were processed by immunoperoxidase staining with primary antibodies against RANTES, MCP-1, ICAM-1, and LFA-1
/CD11a.
RESULTS. The levels of RANTES and SDF-1
were significantly elevated in patients with at least severe nonproliferative diabetic retinopathy compared with those with less severe diabetic retinopathy (P < 0.001 and 0.007, respectively). Positive immunostaining was observed in the inner retina for MCP-1 and RANTES of the patient with diabetes. Staining was strongly positive throughout the diabetic retina for ICAM-1. Normal retinal tissues showed little reactivity.
CONCLUSIONS. Serum chemokines were significantly elevated in patients with at least severe nonproliferative diabetic retinopathy compared with those who had less severe retinopathy. Elevated levels of the chemokines and cell adhesion molecules were also identified in eyes of a donor with ischemic diabetic retinopathy. These findings provide evidence to support the role of inflammation in the pathogenesis of diabetic retinopathy.
Although the pathogenesis of diabetic retinopathy is not known, diabetic retinopathy and nephropathy may have components of chronic inflammation. Increasing evidence comes from animal models of diabetic retinopathy, human tissues from patients with diabetic retinopathy and also studies measuring elevated inflammatory protein levels of cytokines, chemokines, and adhesion molecules in the vitreous of patients with diabetic retinopathy.3 4 5 6 7
In comparison, relatively few studies have examined chemokine levels in the serum of patients with diabetes. Evaluation of adhesion molecule levels in the serum of patients with diabetes has produced mixed results; this may be due to the differing comparison groups used in the experiments.8 9 10 11 12
In this study, we measured the serum levels of several chemokines, cytokines, adhesion molecules, and one growth factor in patients with diabetic retinopathy. We chose to study these inflammatory mediators, because they have been linked with diabetic retinopathy or to key etiologic components of diabetic retinopathy progression, such as hypoxia or angiogenesis. The association of these serum chemokines and cytokines with the increasing severity of diabetic retinopathy and the presence of diabetic macular edema was assessed. We further evaluated our results by performing immunohistochemistry on the retina of a deceased patient who had documented severe nonproliferative diabetic retinopathy, and these results were compared with results from the retina of a nondiabetic subject.
| Materials and Methods |
|---|
|
|
|---|
Patients, evaluated at the clinical center at the NEI, had complete eye examinations that included best corrected visual acuity, slit lamp biomicroscopy, tonometry, and dilated ophthalmoscopy. Stereoscopic fundus photographs of the retina in seven standard fields were performed and graded at the NEI using the final scale of the Early Treatment Diabetic Retinopathy Study (ETDRS) Classification.13
Demographic characteristics of the enrolled patients collected include age, gender, race, duration of diabetes, and age of onset of diabetes. Systolic and diastolic blood pressures, hemoglobin A1C, fasting serum total cholesterol, triglycerides, high-density lipoproteins, low-density lipoproteins, serum creatinine, and urinalysis were also measured. This study was approved by the institutional review board for human subjects and informed consents were obtained from all patients, in accordance with the Declaration of Helsinki.
Samples and Cytokine Measurement
Fasting serum levels of multiple factors were measured in patients with diabetes and various severities of diabetic retinopathy. Fresh serum samples were evaluated with ELISA kits (R&D Systems, Inc., Minneapolis, MN; and Endogen, Rockford, IL). Chemokines measured included regulated on activation, normal T-cell expressed and presumably secreted (RANTES)/CCL5, epithelial neutrophil activator (ENA)-78/CXCL5, interferon-induced protein (IP)-10/CXCL10, stromal cellderived factor (SDF)-1
/CXCLl2, monocyte chemoattractant protein (MCP)-1/CCL2, macrophage inflammatory protein (MIP)-1
/CCL3, interleukin (IL)-8/CXCL8; the cytokine IL-6; the cell adhesion molecules intercellular adhesion molecule (ICAM-1/CD54) and vascular cell adhesion molecule (VCAM/CD106); and the growth factor, vascular endothelial growth factor (VEGF). These chemokines, cytokines, and cell adhesion molecules were considered because of data from published studies linking each to diabetic retinopathy or to processes known to be involved in the development of diabetic retinopathy.11 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 RANTES was one of the factors that have not been implicated in diabetic retinopathy. It was included because of the previous study of histopathology that suggests macrophage function is important in the pathogenesis of diabetic retinopathy. This previous report showed increased macrophage recruitment in an eye with nonproliferative diabetic retinopathy and diabetic macular edema, suggesting that cytokines related to macrophage function should also be included.33
Immunohistochemistry
Frozen sections of a donor eye with severe nonproliferative diabetic retinopathy and diabetic macular edema and a normal donor eye were processed for immunohistochemical staining33 by the avidin-biotin-complex immunoperoxidase technique. The primary antibodies consisted of mouse monoclonal antibodies against two human chemokines, MCP-1 and RANTES, and the ICAM-1 and its primary ligand LFA-1
(R&D Systems, Inc., Minneapolis, MN). Frozen sections were fixed in acetone and absorbed in horse serum. After incubation with the primary antibody, the slides were incubated with biotin-labeled horse anti-mouse antibody (Vector Laboratory, Burlingame, CA). After amplification with avidin-biotin-complex (Vector Laboratory), slides were developed in 3,3' diaminobenzidine, nickel sulfate, and hydrogen peroxide.
Data Analysis
To evaluate the association of the serum factors with the severity of intraretinal diabetic retinopathy, patients were divided into two groups, depending on the severity. The patients with severe nonproliferative diabetic retinopathy (level 53), as assessed with the ETDRS-modified Airlie House grading system, had to have at least one of the following: four stereo fundus photographic fields with severe hemorrhages and microaneurysms, two fields with at least definite venous beading, or one field with at least moderate intraretinal microvascular abnormalities. The first group consisted of patients with retinopathy of this severity or worse. They were compared to patients with less severe diabetic retinopathy and those with none to mild or moderate nonproliferative changes caused by diabetic retinopathy.
Analyses were performed with logistic regression, to evaluate the association between the serum factors and the severity of the retinopathy. We adjusted for the following variables by including them in the statistical model: age, sex, total cholesterol, hemoglobin A1C, creatinine, duration of diabetes, type of diabetes, and the presence or absence of macular edema. Age, total cholesterol, hemoglobin A1C, creatinine, and duration of diabetes were treated as continuous variables. Macular edema was defined as retinal thickening affecting or threatening the center of the fovea and/or presence of focal laser photocoagulation. Several measurements of chemokine and cytokine levels were imputed, because their measured level was listed as below the predefined minimum level for that test (ENA-78, RANTES, IL-8, MIP-1
, and IL-6). For analysis purposes, such levels were imputed tobe half the predefined minimum, rather than zero. Some of the chemokinecytokine analyses contained a high number of these censored levels (IL-8, 67%; MIP-1
, 77%; and IL-6, 88%) and rather than analyzing their actual levels they were dichotomized. The analyses were performed by computer (SAS System, 8.2; SAS, Cary, NC).
| Results |
|---|
|
|
|---|
|
|
(OR = 5.15; 95% CI, 1.5617.02 per ng/mL unit increment of the log; P = 0.007; Table 4 ). No other significant differences were found. Despite the differences found, it should be noted that all the data obtained within our population of persons with diabetes were within the normal range found in normal control subjects within the laboratory where samples were measured.
|
|
was slightly increased in the retinas of the patient with diabetes compared with the retina from the person without diabetes.
|
| Discussion |
|---|
|
|
|---|
and RANTES in patients with severe nonproliferative diabetic retinopathy compared with patients who had less severe retinopathy. Each of the chemokines discussed has been implicated in studies examining various etiologic components of diabetic retinopathy, ranging from leukostasis to the hypoxic response to angiogenesis; however, none of these chemokines has been directly linked to diabetic retinopathy in the literature. RANTES/CCL5 is an infrequently studied chemokine that has not been evaluated for its potential role in retinal disease. It has been shown to have potentially angiogenic effects in various tumor model systems.34 35 The expression of RANTES has been associated with the expression of ICAM-1 in renal fibroblast cultures.36 The 28 G polymorphism in the RANTES promoter genotype has been associated with a twofold increase in the risk for diabetic nephropathy.37 A RANTES receptor antagonist has also been shown to reduce monocyte-induced renal damage during transplant rejection.38 39 Successful inhibition of pancreatic ß-cell destruction and diabetes has recently been reported in nonobese diabetic (NOD) mice treated with a neutralizing anti-CCR5 (the ligand of RANTES/CCL5) antibody.40
Our data indicate that RANTES is associated with more ischemic forms of diabetic retinopathy. In addition to inflammatory cells, RANTES is produced by retinal endothelial and pigment epithelial cells.21 41 We demonstrated the presence of RANTES in the retina with diabetic retinopathy. In this study, serum RANTES levels in the patients with less severe retinopathy were less than that in the normal control, this may not reflect a true significant variation. Other studies have shown that the normal control may have even lower levels of RANTES, with a mean of 900 pg/mL.42 Further studies are needed to clarify the potential role of RANTES in the development of diabetic retinopathy and other diabetic microangiopathies.
SDF-1
/CXCL12 has not been directly linked to diabetic retinopathy in previous studies. However, SDF-1
has been shown in several studies to be associated with key etiologic components of diabetic retinopathy.43 The receptor for SDF-1
CXCR4 is the predominant chemokine receptor expressed on inflammatory cells, and incubation of SDF-1
has been shown to promote intracellular signaling and chemotaxis in RPE cells.44 Hypoxia induces upregulation of SDF-1
in synovial fibroblasts,45 and SDF-1
has been shown to have angiogenic effects both in vivo and in vitro.30 46 A polymorphism of SDF-1
has been linked to decreased age of onset of diabetes in a population of Japanese males, and anti-SDF-1
has been linked to decreased incidence of diabetes in a murine model.47 48 SDF-1
may be an essential chemokine for trafficking and migration of autoreactive B cells in the development of diabetes.47 We have also found elevated levels of serum SDF-1
to be associated with the development of more ischemic forms of diabetic retinopathy. As just mentioned, SDF-1
has been linked to key processes involved in diabetic retinopathy.43 Results of previous studies and the data from the present study suggest a potential role for SDF-1
in the development of diabetic retinopathy.
Immunostaining of the retina of a patient with severe nonproliferative diabetic retinopathy and exudative macular edema demonstrated that ICAM-1/CD54 was strongly expressed throughout the retina of the patient with diabetes in comparison with its absence in a normal retina. These results are in agreement with previous examinations of ICAM-1 expression in the ocular tissue of patients with diabetes.11 48 49 ICAM-1 is an intracellular adhesion molecule necessary for the adhesion of leukocytes to capillary endothelium. It has been implicated in the pathogenesis of diabetic retinopathy in several studies. An examination of epiretinal membranes from patients with proliferative diabetic retinopathy revealed a strong ICAM-1 signal.50 51 It has also been implicated in the development of leukostasis, a prominent feature of diabetic retinopathy. An mAb to ICAM-1 blocked diabetes-induced leukostasis and decreased the breakdown of the bloodretinal barrier in a diabetic rat model.4
We found elevated MCP-1/CCL2 expression in the inner retina of a patient with severe intraretinal diabetic retinopathy in comparison with the normal retina. Although no elevation of serum MCP-1 was measured in the present study, MCP-1 is reported to be increased in the vitreous of patients with proliferative diabetic retinopathy.5 52 Previously, we have observed many macrophages in the retina of diabetic patients These infiltrating macrophages could produce MCP-1 in the retina.33 The inner retina is hypothesized to be the most hypoxic part of the diabetic retina. Measurements of oxygen tension in the retina of diabetic cats have shown this to be the case.53 Hyperglycemia has also been shown to increase the expression of MCP-1 by vascular endothelial cells.25 Studies of a hypoxia-induced ocular neovascularization mouse model found an increase in MCP-1 mRNA and protein expression after hypoxia induction. MCP-1 was found predominately in the inner retina in this model. Injection of anti-MCP-1 antibodies depressed the inflammatory neovascularization in this model.29 MCP-1 has been shown to induce ICAM-1 expression in renal tubular endothelial cells.54 It has been shown that MCP-1 is produced by retinal endothelial cells.21 These data from previous studies suggest a role for MCP-1 in the pathogenesis of diabetic retinopathy.
Limitations of our study include the fact that most of the data obtained from the patients are within normal levels found in our laboratory. Concurrent control subjects, unfortunately, were not evaluated. It is possible that the inflammatory process of microvascular abnormalities may only reflect mostly local changes within the ocular tissues and may not be reflected within the serum. Nevertheless, it is compelling to evaluate the differences in these serum risk factors in patients with more severe retinopathy. Our immunohistochemical examination of ocular tissue was also limited by the lack of additional patient samples. However, these data support our serum data, in that RANTES was also present in the ocular tissues. This is the first report of RANTES in the ocular tissues.
These data, however, suggest a further role for chemokines, cytokines, and cell adhesion molecules in the development of diabetic retinopathy and provide a potential tool for the assessment of risk in patients with diabetic retinopathy. The expression of RANTES and SDF-1
in the most hypoxic inner layers of the retina suggests a local response, which attracts leukocytes to the ischemic lesions.53 The more universal expression of ICAM-1 is most likely essential for the diapedesis and migration of leukocytes to the areas of ischemia. It is possible that RANTES and SDF-1
act in concert as part of the natural response to ischemia in the retina, to attract leukocytes that may play a role in propagating the damage in a series of self-sustaining paracrine loops. Our findings suggest roles for RANTES and SDF-1
in the development of more ischemic or severe diabetic retinopathy. Additional studies are needed to establish conclusively an association between these molecules and diabetic retinopathy.
| Footnotes |
|---|
Submitted for publication September 5, 2004; revised December 13, 2004, and May 18, 2005; accepted September 7, 2005.
Disclosure: A.D. Meleth, None; E. Agrón, None; C.-C. Chan, None; G.F. Reed, None; K. Arora, None; G. Byrnes, None; K.G. Csaky, None; F.L. Ferris III, None; E.Y. Chew, 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.
Corresponding author: Emily Y. Chew, Division of Epidemiology and Clinical Research, National Eye Institute, National Institutes of Health, Building 10, CRC, Room 3-2531, 10 Center Drive, MSC-1204, Bethesda, MD 20892-1204; echew{at}nei.nih.gov.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
X. Liu, M. G. Mameza, Y. S. Lee, C. I. Eseonu, C.-R. Yu, J. J. Kang Derwent, and C. E. Egwuagu Suppressors of Cytokine-Signaling Proteins Induce Insulin Resistance in the Retina and Promote Survival of Retinal Cells Diabetes, June 1, 2008; 57(6): 1651 - 1658. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Westerbacka, A. Corner, M. Kolak, J. Makkonen, U. Turpeinen, A. Hamsten, R. M. Fisher, and H. Yki-Jarvinen Insulin regulation of MCP-1 in human adipose tissue of obese and lean women Am J Physiol Endocrinol Metab, May 1, 2008; 294(5): E841 - E845. [Abstract] [Full Text] [PDF] |
||||
![]() |
H.-y. Zeng, W. R. Green, and M. O. M. Tso Microglial Activation in Human Diabetic Retinopathy Arch Ophthalmol, February 1, 2008; 126(2): 227 - 232. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Nakazawa, T. Hisatomi, C. Nakazawa, K. Noda, K. Maruyama, H. She, A. Matsubara, S. Miyahara, S. Nakao, Y. Yin, et al. From the Cover: Monocyte chemoattractant protein 1 mediates retinal detachment-induced photoreceptor apoptosis PNAS, February 13, 2007; 104(7): 2425 - 2430. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Muranaka, Y. Yanagi, Y. Tamaki, T. Usui, N. Kubota, A. Iriyama, Y. Terauchi, T. Kadowaki, and M. Araie Effects of Peroxisome Proliferator-Activated Receptor {gamma} and Its Ligand on Blood-Retinal Barrier in a Streptozotocin-Induced Diabetic Model. Invest. Ophthalmol. Vis. Sci., October 1, 2006; 47(10): 4547 - 4552. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |