Soluble Alzheimers β-amyloid constricts the cerebral vasculature in vivo

Neuroscience Letters 257 (1998) 77–80

Soluble Alzheimers b-amyloid constricts the cerebral vasculature in vivo Zhiming Suo*, James Humphrey, Amy Kundtz, Faisil Sethi, Andon Placzek, Fiona Crawford, Mike Mullan Roskamp Institute, 3515 E. Fletcher Avenue, University of South Florida, Tampa, FL 33613, USA Received 11 August 1998; accepted 28 September 1998

Abstract Bilateral temporoparietal hypoperfusion has been frequently observed early in the Alzheimer’s disease (AD) process. An increased b-amyloid (Ab) peptide is believed to play a central role in the pathogenesis of AD. In vitro experiments have shown that freshly solubilized Ab enhances constriction of cerebral and peripheral vessels. We propose that in vivo the Ab vasoactive property may contribute to cerebral hypoperfusion of AD patients. To test this hypothesis, we intra-arterially infused freshly solubilized Ab1–40 in rats and observed changes in cerebral blood flow and cerebrovascular resistance using fluorescent microspheres. We found that infusion of Ab in vivo resulted in a decreased blood flow and increased vascular resistance specifically in cerebral cortex but not in heart or kidneys. These data suggest that Ab has a direct and specific constrictive effect on cerebral vessels in vivo, which may contribute to the cerebral hypoperfusion observed early in the AD process.  1998 Elsevier Science Ireland Ltd. All rights reserved

Keywords: b-Amyloid; Alzheimer’s disease; Cerebral blood flow and cerebrovascular resistance

Alzheimer’s disease (AD) is a primary neurodegenerative dementia. Bilateral temporoparietal hypoperfusion is one of the major clinical features evident in the early phases of AD [4,15]. Also, reduction in regional cerebral blood flow (CBF) is closely associated with the severity of dementia [1,4]. It has been demonstrated that severe cerebrovascular hypoperfusion can result in subsequent neuronal degeneration [5,9]. Although in AD while it is generally believed that a decrease in metabolic demand by neurons leads to cerebrovascular hypoperfusion, it is possible that cerebrovascular hypoperfusion may initiate either disturbed neuronal function or even neuronal degeneration. As one of the pathological features, increased b-amyloid (Ab) production is believed to play a central role in the pathogenesis of AD. It is well known that Ab is deposited in the brain parenchyma of Alzheimer patients as senile plaques, and also in the walls of cerebral vessels as cerebral amyloid angiopathy (CAA). In spite of fibrillar Ab deposi* Corresponding author. Tel.: +1 813 9743722; fax: +1 813 9743915; e-mail: [email protected]

0304-3940/98/$ - see front matter PII S0304- 3940(98) 00814- 3

tion, levels of non-fibrillar Ab1–40 are increased in the leptomeninges and leptomeningeal vessels of AD cases compared to those of age-matched controls [3,12]. Since deposition of Ab fibrils also derives from the accumulation of the non-fibrillar form of Ab, non-fibrillar Ab may have early pathogenic consequences in the AD process prior to extensive Ab deposition in the brain parenchyma and cerebral vessels. Recent data from transgenic mice over-expressing Ab as a consequence of the bAPP 670/671 mutation supports a pathogenic role for non-fibrillar Ab. Cerebral vessels of transgenic mice aged 3–4 months demonstrate opposition to relaxation induced by the vasodilator sodium nitroprusside compared to their non-transgenic littermates, despite having no evidence of CAA. Furthermore, occlusion of the middle cerebral artery in these animals results in a larger infarct volume and reduced blood flow in the penumbra compared to non-transgenic littermates [17]. In vitro studies in our laboratories have previously demonstrated that freshly solubilized Ab can enhance vasoconstriction of rat aortae [2,14], and these effects have also been demonstrated

 1998 Elsevier Science Ireland Ltd. All rights reserved

78

Z. Suo et al. / Neuroscience Letters 257 (1998) 77–80

in vitro in bovine cerebral vessels [13]. We therefore propose that the vasoactive effects of Ab observed in vitro may also occur in vivo resulting in reduced CBF, and may contribute to the cerebrovascular hypoperfusion observed early in the AD process. To test this hypothesis, we intra-arterially infused rats with freshly solubilized Ab1–40 or controls, and followed this with an infusion of fluorescent microspheres, which have been proven to be a valid method of determining cerebrovascular blood flow and resistance [6]. Fluorometric quantification from isolated target tissues and reference blood samples enabled measurement of blood flow (BF) in cerebral cortex, heart and kidneys. Arterial blood pressure (BP) was also measured in order to evaluate vascular resistance (VR). Seventeen male Sprague–Dawley rats (7–8 months old) were divided into the following four groups: (1) control, (2) human Ab1–40 (AB), (3) rat Ab1–40 (rAB), (4) rat amylin (rAmylin). The BF and VR were determined separately for left and right cerebral cortices and for both kidneys. Since statistical analyses did not reveal any significant differences between left and right side for these organs, data from both sides were combined. As shown in Fig. 1, infusion of human Ab1–40 or rat Ab1–40 did not result in significant changes to BP, but infusion of rat amylin, a vasodilatory peptide [16] of similar size to Ab1–40 used as a methodological control, significantly decreased the BP in these animals (P , 0.001). By contrast, measurement of CBF during 14 min of treatment with either human or rat Ab1–40 revealed a significant (P , 0.05) decrease compared

Fig. 1. Changes in arterial blood pressure in Ab-infused rats. The animals within different treatment groups (control, n = 4; AB, n = 5; rAB, n = 4 and rAmylin, n = 4) were anesthetized (sodium pentobarbital 55mg/kg, i.p.), and cannulated (P-50) at the right common carotid artery, positioned at the aortic arch for arterial BP measurements and intra-arterial infusions. The average of mean BP measurements taken over a period of 4 min before (Pre) or after (Post) infusion of the peptide was used to evaluate the effects of the peptide on BP changes. Two-way ANOVA followed by post-hoc testing (Sheffe´) revealed the significant differences between groups as indicated.

to controls, but no such decrease occurred with rat amylin (Fig. 2A). Consequently, a calculation of cerebral vascular resistance (CVR) displayed significant (P , 0.05) increases for the Ab1–40 treatments (Fig. 2B). In addition, BF and VR in the heart and kidneys in animals treated with human Ab1– 40 were not significantly different from controls (Fig. 2C,D). These results demonstrated that in vivo infusion of Ab1–40 did not influence peripheral BP but did significantly reduce CBF, reflecting an increased CVR. However, BF in the heart and kidneys was not affected by Ab, indicating that Ab vasoactivity was specific to the cerebral vasculature in vivo. Both rat and human Ab1–40 exhibited similar potency on cerebral vessels suggesting that in vivo Ab vasoactivity did not reflect a species-specific immune response. Collectively, these results suggest that relatively low doses of freshly solubilized Ab1–40 have a direct and specific effect on cerebral vessel constriction in vivo. It is known that Ab1–40 is the predominant isoform found in the walls of cerebral vessels, though Ab1–42, the major isoform deposited in senile plaques, has been suggested to be more pathologic. We have previously demonstrated that Ab1–40 is significantly more vasoactive in vitro than Ab1–42 [2]. It has also been shown that intra-arterial infusion of Ab1–40 in guinea pigs results in an immediate cerebrovascular sequestration while Ab1–42 (infused in the same way) traverses the blood–brain barrier and is accumulated in the brain parenchyma [7]. Therefore, intra-arterial infusion of Ab1–40 is an appropriate method for assessment of the effects of increased levels of Ab in the vessel wall. Additionally, although in vitro experiments demonstrated Ab vasoactivity in both cerebral and peripheral vessels [2,13, 14], our study suggests that in vivo, Ab preferentially constricts cerebral microvessels. This may explain why, despite the fact that there may be increased levels of circulating Ab in AD patients, there appear to be predominantly cerebral rather than systemic vascular consequences. Although the mechanism of Ab vasoactivity is not known, it has been previously speculated that it may be mediated by an imbalance in free radicals, specifically superoxide and nitric oxide [14]. However, recent evidence suggests that the main product of this proposed imbalance, peroxynitrite, is not vasoconstrictive. The same study also shows that superoxide does not mediate Ab vasoactivity [10]. Additionally, brefeldin A completely blocks Ab vasoactivity suggesting that internalization of Ab or endoplasmic reticulum-Golgi apparatus transport may mediate this effect [11]. As none of the variables that we measured in the present experimental paradigm reflect activity of large vessels, the current study does not exclude the possibility that as well as affecting the microvasculature, circulating Ab also affects large and medium sized arteries, and perhaps veins. Moreover, peripheral blood pressure is mainly determined by changes in resistance vessels, and thus, the absence of change in peripheral blood pressure does not necessarily reflect an absence of vasoconstriction in large vessels.

Z. Suo et al. / Neuroscience Letters 257 (1998) 77–80

79

Fig. 2. BF and VR changes in cerebral cortex, heart and kidneys in Ab-infused rats. The surgical procedures for BF measurements were performed as described in Fig. 1. To obtain a reference blood sample, animals were also cannulated at the left femoral artery. Animals in treatment groups received 40 nmol/kg of either human Ab1–40 (QCB), rat Ab1–40 (QCB) or rat amylin (RBI), freshly solubilized in water (500 mM) and then diluted in 0.5 ml 0.9% NaCl solution. Control animals received 0.5 ml 0.9% NaCl alone. Four minutes after infusion, 1 ml of fluorescent microsphere (100 000 spheres/ml, 15 mm diameter; Molecular Probes) suspension in 0.9% NaCl solution was infused at the carotid cannula at a rate of 1 ml/min, while simultaneously withdrawing reference blood from the femoral artery at the same rate. After 10 min animals were sacrificed by decapitation and the target tissues (kidney, heart, and cerebral cortex) were dissected and then followed by digestion and fluorometric quantification. BF and VR for specific organs were calculated using the following established equations [8]: BF, (fluorescent counts for the tissue × rate of reference blood withdrawal × 100)/(fluorescent counts for reference blood × weight of the tissue); VR, mean BP/BF. One-way ANOVA followed by post-hoc testing (Sheffe´) revealed the significant between-groups differences indicated.

The cerebral specificity of these vasoconstrictive effects that we observe in vivo, has several possible explanations. There may be phenotypic differences in response to Ab stimulation between the vessels in peripheral and cerebral tissues. For instance, it is known that endothelin is the major vasoconstrictor controlling cerebral vascular tension whereas epinephrine and norepinephrine are major regulators of peripheral vasoconstriction. We have previously shown that the same dose of Ab enhances endothelin vasoconstriction much more (5–10 fold) than an epinephrine analogue (phenylephrine) induced vasoconstriction [2,14]. Therefore, it

may be that the selective cerebral effects are due to the potentiation of endothelin vasoconstriction. Other receptor systems of membrane-associated enzymes may be central or play an additional role. Although the specificity of our findings may suggest a receptor-related event, it is also possible that distribution of the infused Ab in vivo differs in cerebral and peripheral vessels. Additional investigation, including the examination of the kinetics of an Ab dose-dependent response, will be necessary to further characterize the mechanism of in vivo Ab vasoactivity, and clarify the differences observed between cerebral and peripheral tissues.

80

Z. Suo et al. / Neuroscience Letters 257 (1998) 77–80

We would like to thank Mr. and Mrs. Robert Roskamp for their generous support to this work. This work was also supported by an Alzheimer’s Association Zenith Award (96-025) to M.M. Also we would like to gratefully acknowledge the help of Dani Fallin, George Su, Patrick Pompl and Laila Abdullah. [1] Brown, D.R., Hunter, R., Wyper, D.J., Patterson, J., Kelly, R.C., Montaldi, D. and McCullouch, J., Longitudinal changes in cognitive function and regional cerebral function in Alzheimer’s disease: a SPECT blood flow study, J. Psychiatr. Res., 30 (2) (1996) 109–126. [2] Crawford, F., Suo, Z., Fang, C. and Mullan, M., Characteristics of the in vitro vasoactivity of b-amyloid peptides, Exp. Neurol., 150 (1998) 159–168. [3] Hamano, T., Yoshimura, M., Yamazaki, T., Shinkai, Y., Yanagisawa, K., Kuriyama, M. and Ihara, Y., Amyloid beta-protein (A beta) accumulation in the leptomeninges during aging and in Alzheimer disease, J. Neuropathol. Exp. Neurol., 56 (8) (1997) 922–932. [4] Hirsch, C., Bartenstein, P., Minoshima, S., Mosch, D., Willoch, F., Buch, K., Schad, D., Schwaiger, M. and Kurz, A., Reduction of regional cerebral blood flow and cognitive impairment in patients with Alzheimer’s disease: evaluation of an observerindependent analytic approach, Dement. Geriatr. Cogn. Disord., 8 (2) (1997) 98–104. [5] Lindsberg, P.J., Frerichs, K.U., Burris, J.A., Hallenbeck, J.M. and Feuerstein, G., Cortical microcirculation in a new model of focal laser-induced secondary brain damage, J. Cereb. Blood Flow Metab., 11 (1) (1991) 88–98. [6] Marcus, M.L., Heistad, D.D., Ehrhardt, J.C. and Abboud, F.M., Total and regional cerebral blood flow measurement with 7–10, 15-, 25-, and 50-mum microspheres, J. Appl. Physiol., 40 (4) (1976) 501–507. [7] Martel, C., Mackic, J., McComb, J., Ghiso, J. and Zlokovic, B., Blood–brain barrier uptake of the 40 and 42 amino acid sequences of circulating Alzheimer’s amyloid b in guinea pigs, Neurosci. Lett., 206 (1996) 157–160.

[8] Nakia, M., Tamaki, K., Yamamoto, J., Shimouchi, A. and Maeda, M., A minimally invasive technique for multiple measurement of regional blood flow of the rat brain using radiolabeled microspheres, Brain Res., 507 (1) (1990) 168–171. [9] Ni, J.W., Matsumoto, K., Li, H.B., Murakami, Y. and Watanabe, H., Neuronal damage and decrease of central acetylcholine level following permanent occlusion of bilateral common carotid arteries in rat, Brain Res., 673 (2) (1995) 290–296. [10] Paris, D., Parker, T.A., Town, T., Suo, Z., Fang, C., Humphrey, J., Crawford, F. and Mullan, M., Role of peroxynitrite in the vasoactive and cytotoxic effects of Alzheimer’s b-amyloid 1– 40 peptide, Exp. Neurol., 152 (1998) 116–122. [11] Paris, D., Town, T., Parker, T.A., Humphrey, J. and Mullan, M., Isoform-specific vasoconstriction induced by Apolipoprotein E and modulation of this effect by Alzheimer’s b-amyloid peptide, Neurosci. Lett., (1998) in press. [12] Shinkai, Y., Yoshimura, M., Ito, Y., Odaka, A., Suzuki, N., Yanagisawa, K. and Ihara, Y., Amyloid beta-proteins 1–40 and 1–42(43) in the soluble fraction of extra- and intracranial blood vessels, Ann. Neurol., 38 (3) (1995) 421–428. [13] Thomas, T., McClendon, C., Sutton, E.T. and Thomas, G., Cerebrovascular endothelial dysfunction mediated by betaamyloid, NeuroReport, 8 (6) (1997) 1387–1391. [14] Thomas, T., Thomas, G., McClendon, C., Sutton, T. and Mullan, M., b-Amyloid-mediated vasoactivity and vascular endothelial damage, Nature, 380 (1996) 168–171. [15] Waldemar, G., Hogh, P. and Paulson, O.B., Functional brain imaging with single-photon emission computed tomography in the diagnosis of Alzheimer’s disease, Int. Psychogeriatr., 9 (Suppl. 1) (1997) 223–227. [16] Westfall, T.C. and Curfman-Falvey, M., Amylin-induced relaxation of the perfused mesenteric arterial bed:meditatin by calcitonin gene-related peptide receptors, J. Cardiovasc. Pharmacol., 26 (6) (1995) 932–936. [17] Zhang, F., Eckman, C., Younkin, S., Hsiao, K.K. and Ladecoal, C., Increased susceptibility to ischemic brain damage in transgenic mice overexpressing the amyloid precursor protein, J. Neurosci., 17 (20) (1997) 7655–7661.