Molecular Brain Research 62 Ž1998. 101–105
Short communication
Vascular endothelial growth factor in Alzheimer’s disease and experimental cerebral ischemia R.N. Kalaria a
a,)
, D.L. Cohen a , D.R.D. Premkumar a , S. Nag b, J.C. LaManna a , W.D. Lust
a
Departments of Neurology, Neurosciences, Neurological Surgery and Pathology, Case Western ReserÕe UniÕersity, CleÕeland, OH 44106, USA b Department of Neuropathology, Toronto Western Hospital, Toronto, Canada Accepted 14 July 1998
Abstract Several growth factors have been implicated in the pathogenesis of Alzheimer’s disease ŽAD.. We considered whether the vascular endothelial growth factor ŽVEGF. is involved in the vascular pathology associated with most cases of AD. We observed enhanced VEGF immunoreactivity in clusters of reactive astrocytes in the neocortex of subjects with AD compared to elderly controls. VEGF reactivity was also noted in walls of many large intraparenchymal vessels and diffuse perivascular deposits. In addition, we established that astrocytic and perivascular VEGF reactivity was enhanced in cerebral cortex of rats subjected to cerebral ischemia and to chronic hypoxia; experimental conditions known to be associated with astrogliosis and angiogenesis. We suggest the increased VEGF reactivity, also observed in infarcted human brain tissue, implicates compensatory mechanisms to counter insufficient vascularity or reduced perfusion Žoligemia. apparent in AD. q 1998 Elsevier Science B.V. All rights reserved. Keywords: Alzheimer’s disease; Astrocyte; Glioblastoma multiforme; Growth factor; Hypoxia; Immunocytochemistry; Ischemia; Vascular endothelial growth factor; Vascular smooth muscle cell
In addition to nerve growth factor, several growth factors have been implicated in the neurodegenerative processes associated with dementing disorders. Some of these factors along with the growth promoting properties of the amyloid precursor protein have been suggested to play various roles in the pathogenesis of Alzheimer’s disease ŽAD. w25x. Previous studies have described the localization of epidermal growth factor and its receptor w23x, basic fibroblast growth factor w6,22x, transforming growth factorß w26x, platelet-derived growth factor ŽPDGF. w15x, tumor necrosis factor-a w4x and the cytokine, interleukin-1ß w7x in lesions or cells undergoing changes associated with AD. PDGF is associated with alterations in neurons as well as perivascular glia in AD w15x. Some of the factors also appear to be localized to extracellular matrix proteins bound to amyloid ß deposits. Remarkably, many appear to be produced by glia such as reactive astrocytes and microglia w6,7,26x. )
Corresponding author. CBV Path Group, MRC Neurochemical Pathology Unit, Newcastle General Hospital, Westgate Road, Newcastle upon Tyne NE4 6BE. Fax: q 44-191-272-5291; E-m ail:
[email protected]
Vascular endothelial growth factor ŽVEGF. or vascular permeability factor is a 41–46 kDa dimer of two identical disulphide-linked subunits. To date four isoforms of VEGF are recognized, which are encoded by four alternatively spliced mRNAs comprising 121, 165, 189, and 206 amino acid residues w3,24x. All forms are mitogenic for endothelial cells, exhibit vascular permeability activity and variably expressed in different tissues w3x. VEGF binds with high affinity to specific receptors such as the flk-1 tyrosine kinase type localized on endothelial cells. Recent studies suggest VEGF to be the main regulator of hypoxia and tumor-induced angiogenesis w19,21x. VEGF can also regulate the ubiquitous glucose transporter, GLUT1 and the tissue factor known to be involved in initiating the extrinsic coagulation pathway w2x. Interestingly, both GLUT1 and tissue factor or tissue thromboplastin, have been previously implicated to play a role in the vascular pathology associated with AD w8,16x. These previous observations and our interests in the status of the cerebral microvasculature in AD w8x prompted us to examine whether VEGF has a role in the vascular pathology w10x associated with AD. Consistent with previous evidence for profound degeneration and apparent growth of the microvasculature in AD
0169-328Xr98r$ - see front matter q 1998 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 9 - 3 2 8 X Ž 9 8 . 0 0 1 9 0 - 9
102
R.N. Kalaria et al.r Molecular Brain Research 62 (1998) 101–105
R.N. Kalaria et al.r Molecular Brain Research 62 (1998) 101–105
w9x, we considered the possibility that VEGF might be activated in AD. Here, we used immunocytochemical techniques to examine VEGF protein in AD and experimental conditions associated with vascular and glial proliferation. Human brain tissue from a total of 12 subjects with AD and six age-matched controls was sampled at autopsy within 2 to 10 h after death. The mean Ž"S.E.M.. age of the AD subjects was 76 " 2 and that of the controls was 71 " 3 years. The postmortem intervals between death and tissue retrieval were not different for the two groups. The mean Ž"S.E.M.. interval hours for AD subjects was 4.7 " 0.4 and that for controls was 5.7 " 0.7. The diagnosis of AD, suspected clinically, was confirmed by histological examination which satisfied the consortium to establish a registry for Alzheimer’s disease ŽCERAD. criteria revealing the presence of neuritic plaques in the neocortex and neurofibrillary tangles in the hippocampus w12,20x. Brain tissues were also obtained from other sources. A specimen of glioblastoma multiforme from a 40 year old man, occipital cortex from a 76 year old man diagnosed with borderline AD and multi-infarct dementia, temporal cortex from a 60 year old woman with cerebrovascular disease, frontal cortex from a rhesus monkey and cerebral cortex from rats subjected to focal stroke or hypobaric hypoxia. Experimental cerebral ischemia was induced in spontaneously hypertensive Wistar rats by occlusion of the middle cerebral artery ŽMCA. essentially as described previously w11x. Prolonged hypoxia was induced by maintaining adult Wistar rats in hypobaric chambers at 0.5 atm for a period of 2 weeks w14x. For immunocytochemistry, fresh tissue blocks were either immersion fixed in paraformaldehyde lysine periodate or 8% buffered formalin, cryoprotected and cut into 15–20 mm sections for immunostaining w20x. Tissue sections were incubated with affinity purified antiserum against VEGF1 – 20 carboxyl-terminal peptide Ždilution 1:1000– 5000, Santa Cruz Biotechnology, CA.. Adjacent unstained tissue sections or VEGF immunostained sections were incubated with antibody to GFAP ŽDako, 1:3000., HLADR ŽDako, 1:100., collagen IV ŽChemicon, 1:1000., glucose transporter, GLUT1 ŽCalbiochem, 1:2000. or Aß protein Ž4G8, 1:2000.. Primary antibody bound antigens were localized using the avidin–biotin complex ŽABC.
103
method w12x. In some experiments, paraffin sections of 6 mm thickness were immunostained in the same manner. To demonstrate specificity of staining, control sections were incubated with absorbed antibody. VEGF antiserum was incubated with the 15–20 mgrml of peptide for 14–16 h at 48C and used as primary antibody. Specificity was also evaluated by incubating sections with either irrelevant antibodies or normal rabbit preimmune serum, and in the absence of primary antibody. Immunocytochemical staining of neocortical tissues from neurologically normal subjects or adult rats showed negligible VEGF immunoreactivity, which was largely restricted to astrocyte-like cells. There was no other distinct pattern of cellular or extracellular localization in most of the normal aging controls without neurological disease. Immunocytochemical analyses of neocortical tissues from AD subjects revealed prominent VEGF reactivity in numerous cells resembling astrocytes that were often apparent as clusters ŽFig. 1a.. Double immunostaining with antibodies to VEGF and GFAP or HLA-DR confirmed the astrocytic staining. VEGF was not evident in microglial cells as revealed by HLA-DR immunoreactivity ŽFig. 1b.. Occasionally profiles of pyramidal neurons were lightly stained but this appeared to be a non-specific reaction Žnot shown., which was also evident in three of the six controls. Examination of temporal and occipital cortex from all 12 AD subjects indicated VEGF reactivity was apparent in approximately 40% of the reactive cells, often localized around vessels ŽFig. 1c–e.. The reactivity was less apparent in the frontal cortex. We also observed strong VEGF reactivity in walls of many large intraparenchymal vessels ŽFig. 1c.. Such reactivity was associated with the normal appearing medial smooth muscle cell layers. It was evident that the presence of VEGF immunoreactivity in cases with moderate to severe cerebral amyloid angiopathy ŽCAA. was more intense than those with minimal CAA. However, we often noted diffuse perivascular deposits associated with both large and smaller vessels that were VEGF positive ŽFig. 1f and g.. These were often evident in samples that exhibited vascular abnormalities. The perivascular staining was readily identified by examination of double immunostained sections with antibodies to collagen IV Žor GLUT1. and VEGF ŽFig. 1g.. In addition to the
Fig. 1. Immunocytochemical localization of VEGF reactivity in cerebral cortex of human subjects with AD and of rats subjected to experimental focal ischemia. Ža. Clusters of VEGF Žbrown. immunoreactive astrocytes and perivascular cells Žarrows. in temporal cortex from an AD subject. Note, isolated GFAP stained cells Žlight purple. are not VEGF reactive. Žb. VEGF reactive astrocytes Ždark brown. around a diffuse plaque with a microvessel. VEGF reactivity was not present in the HLA-DR Žpurple. positive microglia Žarrows.. Žc. Numerous perivascular astrocytes labelled by VEGF antibodies Žbrown.. Intense VEGF reactivity was also observed in the large vessel wall. Žd. Adjacent section Žto c. showing almost complete absorption with VEGF carboxyl-terminal peptide. Že. VEGF reactive astrocyte Žbrown. juxtaposed to a microvessel stained for collagen IV Žpurple.. Žf and g. Perivascular deposits containing VEGF reactivity Žbrown. associated with large vessels Žf. and capillaries Žg. stained with collagen IV antibodies Žpurple.. Amyloid ß was not consistently present in these deposits. Žh. Positive VEGF reactivity in neoplastic glial cells in glioblastoma multiforme. Some multi-nucleated giant cells were also positive. In adjacent sections, similar profile of cells was stained for GFAP Žnot shown.. Ži. VEGF reactive astrocytes in the perifocal region of infarcted ipsilateral cortex of a rat with MCA occlusion. Žj. Adjacent section showing absorption of reactivity by preincubation with VEGF peptide. Magnification bar, a s 300 mm, b–g s 60 mm, h–j s 80 mm.
104
R.N. Kalaria et al.r Molecular Brain Research 62 (1998) 101–105
absorption of VEGF immunoreactivity by the respective peptide ŽFig. 1d., the specificity of the observations was demonstrated by positive staining of neoplastic cells ŽFig. 1h. in sections of glioblastoma multiforme w19x. In parallel experiments, we observed sporadic VEGF positive astrocytes in two cases of subcortical gliosis and in astrocytes along the perifocal region of cerebral infarct in the occipital and temporal cortex from subjects with cerebrovascular disease or multi-infarct dementia and AD Žnot shown.. We also observed immunoreactivity in vascular smooth muscle cells and some adventitial fibrocytes associated with the leptomeninges. Similarly, intense VEGF immunoreactivity was observed in the ipsilateral neocortex of rats with MCA occlusion and in rats subjected to chronic hypoxia. In rats with MCA occlusion, the perifocal regions and some midline structures were intensely immunostained by VEGF antibodies ŽFig. 1i and j.. The peak response was evident at 3–5 days after the occlusion. Our observations invariably demonstrated intense VEGF immunoreactivity to be predominantly localized in perivascular reactive astrocytes. Extracellular VEGF immunoreactive deposits were also associated with cerebral vessels in AD. Consistent with previous observations w1,19x, we further demonstrated VEGF immunoreactivity in glial neoplasm. That not all GFAP positive astrocytes were VEGF reactive suggests that only a subclass of those closely associated with the vasculature switch on production of VEGF, which can be readily detected by immunochemical procedures. This view is compatible with an increased VEGF mRNA in neocortex that is presumably localized to perivascular astrocytes. Of relevance here is a parallel study in which we observed VEGF mRNA to be significantly increased by 40% as determined by reverse-transcription polymerase chain reaction and in situ hybridization techniques Žw13x and DRDP and RNK, unpublished observations.. While it is possible that VEGF is also induced by the co-existent cerebrovascular vascular pathology in terms of periventricular ischemic lesions and microinfarcts often present in late-onset AD w20x, the consistent localization of protein reactivity and increased soluble VEGF in tissue determined by ELISA ŽRNK and A.B. Pax, unpublished observations. in the cases studied suggests VEGF to be an important factor associated with the cerebral microvasculature in AD. The astrocytic reactivity may induce vascular growthrrepair mechanisms by upregulation of VEGF to increase vascular permeability or endothelial transport of nutrients such as glucose w2x. Our findings also showed that cerebral vessels with smooth muscle and vascular smooth muscle cells derived from vascular fractions and placed in culture express VEGF Žand its mRNA w13x.. This is in agreement with the previously reported expression of VEGF isoforms in smooth muscle cells of peripheral vessels w27x. We suggest that the increased expression of protein and presumably its mRNA in cerebral vessels is related to the pathogenetic events responsible for decreased cerebral perfusion or oxy-
gen extraction and glucose metabolism, and transport apparent in AD w8,9x. This is not inconsistent with the proposal that vascular smooth muscle cell VEGF might be regulated by tissue oxygen-sensing mechanism during prolonged hypoxic episodes w5x. However, our results may relate to the intriguing observations of Masliah et al. w15x showing increased immunoreactivity of PDGF-AA in cerebral microvessels in AD. VEGF is known to bind with extracellular matrix proteins w24x like other factors such as the fibroblast growth factor. It is not clear whether the intensive VEGF immunoreactive deposits around vessels represent immobilized VEGF secreted from perivascular astrocytes. It is likely that binding to extracellular matrix components resulting from damaged or regenerating vessels inactivates the factor. Nevertheless, the enhanced VEGF reactivity around vessels may promote active proliferation of capillary endothelial cells and smooth muscle cells in certain vessels. Whether such a response is limited by concomitant deposition of extracellular amyloid or aging per se is unclear, but these observations are in accord with evidence for smooth muscle cell proliferation in human coronary arteries that develop atherosclerotic lesions w10x. In view of previous evidence indicating increased perivascular VEGF reactivity in glial neoplasia w1,19x and hypoxia-induced angiogenesis w18,21x, our observations imply that vascular proliferation or repair processes occur to some degree in chronic neurodegenerative disorders such as AD. The increased presence of VEGF mRNA in rat brain w17x from animals subjected to ischemic and hypoxic insults and the localization of the factor in astrocytic processes in the perifocal region of ischemic tissue also indicates the initiation of mechanisms for vascular growth and repair in damaged brain tissue w11x. We found enhanced VEGF immunoreactivity to be associated with reactive astrocytes and cerebral vessels in the neocortex of AD subjects. These observations were supported by the distinct localization of VEGF reactivity in glioblastoma multiforme and within reactive astrocytes in brains of rats subjected to ischemia and hypobaric hypoxia. We suggest the increased reactivity in VEGF implicates regulatory mechanisms to compensate for insufficient vascularity and reduced cerebral perfusion in AD.
Acknowledgements We thank our colleagues in the Neuropathology Division, CWRU for help in acquisition of human brain tissue, and Andrea Pax for assistance with some of the initial experiments. We also thank Arthur Oakley for reproducing the microphotographs provided in Fig. 1. We are also grateful for support from the UH Alzheimer Center, the Alzheimer Association ŽZenith Award., Chicago and the USPHS for grants AG 08012 and AG 10030.
R.N. Kalaria et al.r Molecular Brain Research 62 (1998) 101–105
References w1x R.A. Berkman, M.J. Merrill, W.C. Reinhold, W.T. Monacci, A. Saxena, W.C. Clark, J.T. Robertoson, I.U. Ali, E.H. Oldfield, Expression of the vascular permeability factorrvascular endothelial growth factor gene in central nervous system neoplasms, J. Clin. Invest. 91 Ž1993. 153–159. w2x D.T. Connolly, Vascular permeability factor: a unique regulator of blood vessel function, J. Cell. Biochem. 47 Ž1991. 219–223. w3x N. Ferrara, K. Houck, L. Jakeman, D.W. Leung, Molecular and biological properties of the vascular endothelial growth factor family of proteins, Endocr. Rev. 13 Ž1992. 18–32. w4x H. Fillit, W.H. Ding, L. Buee, J. Kalman, L. Altstiel, B. Lawlor, G. Wolf-Klein, Elevated circulating tumor necrosis factor levels in Alzheimer’s disease, Neurosci. Lett. 129 Ž1991. 318–320. w5x M.A. Goldberg, T.J. Schneider, Similarities between the oxygensensing mechanism regulating the expression of vascular endothelial growth factor and erythropoietin, J. Biol. Chem. 269 Ž1994. 4355– 4359. w6x C.W. Gray, A.J. Patel, Regulation of beta-amyloid precursor protein isoform mRNAs by transforming growth factor-beta 1 and interleukin-1 beta in astrocytes, Mol. Brain Res. 19 Ž1993. 25–256. w7x W.S.T. Griffin, L. Stanley, C. Ling, L. White, V. Macleod, L.J. Perrot, C.L. White III, C. Araoz, Brain interleukin 1 and S-100 immunoreactivity are elevated in Down syndrome and Alzheimer disease, Proc. Natl. Acad. Sci. U.S.A. 86 Ž1989. 7611–7615. w8x R.N. Kalaria, The blood-brain barrier and cerebral microcirculation in Alzheimer’s disease, Cerebrovasc. Brain Metab. Rev. 4 Ž1992. 226–260. w9x R.N. Kalaria, Cerebral vessels in ageing and Alzheimer’s disease, Pharm. Therap. 72 Ž1996. 193–214. w10x R.N. Kalaria, S.N. Kroon, Expression of leukocyte antigen CD34 by brain capillaries in Alzheimer’s disease and neurologically normal subjects, Acta Neuropathol. 84 Ž1992. 606–612. w11x R.N. Kalaria, S.U. Bhatti, E.A. Palatinsky, D.H. Pennington, E.R. Shelton, H.W. Chan, G. Perry, W.D. Lust, Accumulation of the ß amyloid precursor protein at sites of ischemic injury in rat brain, NeuroReport 4 Ž1993. 211–214. w12x R.N. Kalaria, A.B. Pax, D.R.D. Premkumar, I. Lieberburg, Production and increased detection of amyloid ß-protein and amyloidogenic fragments in brain microvessels, meningeal vessels and choroid plexus in Alzheimer disease, Mol. Brain Res. 35 Ž1995. 58–68. w13x R.N. Kalaria, D.L. Cohen, A.S. Al-Mudallal, W.D. Lust, D.R.D. Premkumar, VEGFrVPF expression in ischemic brain injury and human neurodegenerative disorders, J. Cereb. Blood Flow Metab. 15 Ž1995. S599.
105
w14x J.C. LaManna, L.M. Vendel, R.N. Farrell, Brain adaptation to chronic hypobaric hypoxia in rats, J. Appl. Physiol. 72 Ž1992. 2238–2243. w15x E. Masliah, M. Mallory, M. Alford, R. Deteresa, T. Saitoh, PDGF is associated with neuronal and glial alterations of Alzheimer’s disease, Neurobiol. Aging 16 Ž1995. 549–556. w16x R.D. McComb, K.A. Miller, S.D. Carson, Tissue factor antigen in senile plaques of Alzheimer’s disease, Am. J. Pathol. 139 Ž1991. 491–494. w17x W.T. Monacci, M.J. Merrill, E.H. Oldfield, Expression of vascular permeability factorrvascular endothelial growth factor in normal rat tissues, Am. J. Physiol 264 Ž1993. C995–C1002. w18x J. Pe’er, D. Shweiki, A. Itin, I. Hemo, H. Gnessin, E. Keshet, Hypoxia-induced expression of vascular endothelial growth factor by retinal cells is a common factor in neovascularizing ocular diseases, Lab. Invest. 72 Ž1995. 638–645. w19x K.H. Plate, G. Breier, H.A. Weich, W. Risau, Vascular endothelial growth factor is a potential tumor angiogenesis factor in human gliomas in vivo, Nature 359 Ž1992. 845–848. w20x D.R.D. Premkumar, D.L. Cohen, P. Hedera, R.P. Friedland, R.N. Kalaria, Apolipoprotein E ´4 alleles in cerebral amyloid angiopathy and cerebrovascular pathology in Alzheimer’s disease, Am. J. Pathol. 148 Ž1996. 2083–2095. w21x D. Shweiki, A. Itin, D. Soffer, E. Keshet, Vascular endothelial growth factor induced by hypoxia may mediate hypoxia-initiated angiogenesis, Nature 359 Ž1992. 843–845. w22x E.G. Stopa, A.M. Gonzalez, R. Chorsky, R.J. Corona, J. Alvarez, E.D. Bird, A. Baird, Basic fibroblast growth facto in Alzheimer’s disease, Biochem. Biophys. Res. Commun. 171 Ž1990. 690–696. w23x S.D. Styren, E.J. Mufson, G.C. Styren, W.H. Civin, J. Rogers, Epidermal growth factor receptor expression in demented and aged human brain, Brain Res. 347 Ž1990. 347–352. w24x E. Tischer, R. Mitchell, T. Hartman, M. Silva, D. Gospodarowicz, J.C. Fiddes, J. Abraham, The human gene for vascular endothelial growth factor. Multiple protein forms are encoded through alternative exon slicing, J. Biol. Chem. 266 Ž1991. 11947–11954. w25x M.H. Tuszynski, F.H. Gage, Neurotrophic factors and neuronal loss: potential relevance to Alzheimer disease, in: R.D. Terry, R. Katzman, K.L. Bick, ŽEds.., Alzheimer Disease. Raven Press, New York, 1994, pp. 405–417. w26x E.A. van der Wal, F. Gomez-Pinilla, C.W. Cotman, Transforming growth factor-beta 1 is in plaques in Alzheimer and Down pathologies, NeuroReport 4 Ž1993. 69–72. w27x K. Weindel, D. Marme, H.A. Weich, AIDS-associated Kaposi’s sarcoma cells in culture express vascular endothelial growth factor, Biochem. Biophys. Res. Commun. 183 Ž1992. 1167–1174.
