Neurochem. Int. Vol. 21, No. 1, pp. 129-133, 1992 Printed in Great Britain. All rights reserved

0197-0186/92 $5.00+0.00 Copyright © 1992Pergamon Press Ltd

GLIAL CELLS IN C O C U L T U R E CAN INCREASE THE ACETYLCHOLINESTERASE ACTIVITY IN H U M A N BRAIN E N D O T H E L I A L CELLS M . P,~K~SKI* a n d P. K~.SA Central Research Laboratory, Albert Szent-Gy6rgyi Medical University, Szeged, Hungary (Received 14 January 1991 ; accepted 4 November 1991)

Abstract--The elements of the cholinergic system (acetylcholinesterase and choline acetyltransferase) and butyrylcholinesterase were studied in human cortical capillary samples, brain-derived endothelial cell cultures and glial cell cultures. It was shown that the elements of the cholinergic system are present in the microvessels, but the choline acetyltransferase activity may be due to contamination with cholinergic nerve terminals since no choline acetyltransferase could be demonstrated in endothelial cell cultures. The present results revealed that the activity of acetylcholinesterase is reduced in the cortical endothelial cell cultures after longer culture times, while butyrylcholinesterase activity is not altered. In a system where endothelial cells were cocultured with embryonic human brain astroglial cells for 12 days in vitro, the acetylcholinesterase activity was increased 2-fold. These results support a glial influence on the enzyme activity of the cerebral endothelium.

The elements of the cholinergic system (acetylcholinesterase, ACHE; choline acetyltransferase, CHAT; and acetylcholine receptors, A C h R ) in cerebral microvessels have been studied by various methods in the central nervous system (CNS) of different animals. Histochemical studies indicate the presence of A C h E in capillaries in different areas of the brain of cat and guinea pig (Kreutzberg et al., 1979), while the enzyme has been demonstrated in nerve fibers in human brain cortical and hippocampal samples (Dob6 et al., 1992) and in the wall of intracerebral microvessels isolated from goat (Estrada et al., 1988). Muscarinic cholinergic receptors ( m A C h R ) could also be demonstrated in goat (Alborch, 1981), rat (Grammas et al., 1983) and bovine (Estrada et al., 1983) cerebral microvessels. The intraendothelial localization of C h A T in rat cortical capillaries was demonstrated by means of immunocytochemistry (Parnavelas et al., 1982; Arneric et al., 1988). The presence of C h A T in capillaries in the CNS was revealed biochemically, but the exact localization of the enzyme is still a matter of debate. Periendothelial (Estrada et al., 1983; Hamel et al., 1987) and/or intraendothelial (Gonz~tlez and SantosBenito, 1987) localizations are equally suggested. Earlier, we have reported the presence of some of the elements of the cholinergic system in the microvessels

*Author to whom all correspondence should be addressed.

and/or endothelial cells (ECs) of human brain (K~sa et al., 1991). Since during an in vitro culture the ECs can lose some of their biochemical properties (reduction of yglutamyl transpeptidase activity; Na +, K+-ATPase activity; A C h E activity: present experiment) the purpose of this study was to examine the effects of human glial cells (GCs) on the enzyme activities of ECs. EXPERIMENTAL PROCEDURES

Preparation o f human brain mierovessels Immediately following medically indicated artificial abortions, the brains from 15-18-week-old human embryos were removed and placed in cold Hank's balanced salt solution (HBSS). The cortical microvessels were prepared by using a modification of the method of Diglio et al. (1982). After homogenization and centrifugation (1000 g, 10 min), the pellet was resuspended in HBSS containing 15% dextran (124,000 mol. wt) and 5% fetal calf serum (FCS), and recentrifuged (2500 g, 20 min). To remove the myelin contamination totally, this latter centrifugation was repeated. An aliquot was transferred to a column of glass beads (0.250.30 mm diam) and the capillaries were collected. Preparation of embryonic human brain ECs cultures After 0.025% collagenase treatment of the capillaries, the ECs were plated onto collagen-coated Falcon plastic dishes. The cultures were maintained in Dulbecco's modified Eagle medium (DMEM) supplemented with 20% FCS, 1.25% glutamine (200 mM) and antibiotics in a 5% CO2-95% air atmosphere.

129

130

M. P~.KASK!and P. K,{SA

In addition to cortical ECs, small pial vessels were also removed, cut into pieces and plated onto collagen-coated plastic dishes. The pial ECs were maintained in a similar medium as mentioned above. After 8 10 days in Hlro, the ECs developed a monolayer and subcultures were provided by trypsinization. The biochemical and histochemical experiments were performed on the subcultures after different numbers of days in vitro (DIV). Preparation q/human astro#lial cultures

The GCs cultures were obtained from 15 18-week-old human fetal brain. Small pieces of embryonic human brain were triturated and centrifuged (1000 g, 10 min). The cells from the pellet were plated onto a collagen-coated special plastic holder (Pfikflski el al., 1990). The cultures were maintained in DMEM supplemented with 10% FCS, glutamine (I.25%, 200 mM) and antibiotics in a 5% CO_, 95% air atmosphere. Ellect O/GCs eullure.~ on A C h E actit!il.r ol ECs

To examine the effects of GCs on the enzyme activities of ECs, Ihe two types of cultures were maintained in a common medium, using a previously described method (Pfikfiski et al., 1990). When GCs confluently covered the surface of the special holder, which usually took DIV 10, this holder was transferred to a Petri dish where subcultures of cortical or pial ECs had started to develop (DIV 0). The two different cell populations (ECs and GCs) were not in contact, but they ~ere in a common medium.

llllmlOlo~3,1oJlemi~'g][ alld histochemical inl!esligglliOtl.s Both GCs and subcultures of ECs developing on collagencoated glass coverslips were identified immunocytochemically by using monoclonal antibody against glial fibrillary acidic protein (GFAP) or human factor VIII antigen, respectively. The immunostaining for GFAP and factor VIII was demonstrated by the peroxidase antiperoxidase method (Sternberger et al., 1970). The AChE histochemistry was performed according to Tsuji (1974). The reaction product was intensified in the samples by an incubation in Tris buffer conlaining 0.05% (w/v) diaminobenzidine, 0.15% (w/v) nickel chloride and 0.005% (v/v) hydrogen peroxide (E. Dob6, personal communication).

human cortical (Fig. 1) and pial ECs. To ascertain whether the ECs were contaminated with astrocytes, cultures were also stained for G F A P positivity. No cells positive to G F A P were revealed in the pial endothelial subculture. The cortical endothelial subculture contained G F A P - i m m u n o r e a c t i v e cells very seldom (Fig. 2). On immunostaining o f the human GCs with G F A P antisera, the perikarya o f the cells exhibited specific staining (Fig. 3). The A C h E histochemistry o f h u m a n cortical ECs showed specific staining in the cytoplasm o f the ECs (Fig. 4 and insert). ACHE, BuChE and C h A T activities in the microvessel samples, and in GCs and ECs cultures o f h u m a n fetal brain cortex arc presented in Table 1. The specific activity o f A C h E was similar in the capillary fraction and in the ECs obtained from the cortical capillaries. There was a significant difference (P < 0.01) between the BuChE activities o f the microvessel fraction and the ECs. A C h E and BuChE activities can not be detected in human GCs. The specific activity o f C h A T was very low (0.003_+0.001 nmol/mg protein/min) in the cortical capillaries, and it could not be detected in the ECs and GCs culture. A C h E and BuChE activities were also measured in the cortical and pial ECs subcultures on the 2nd, 7th and 12th D1V (Fig. 5). The A C h E activity o f the cortical ECs

Biochemical i#n:e.~ti#alions

r h e ChAT assay was performed by employing [~H]acetylCoA as described by Fonnum (t975). The activities of AChE and BuChE were measured spectrophotometrically with acetyl- or butyrylthiocholine as substrate (Ellman et al., 1961). Protein was determined by the method of Lowry et al. ( 1951).

4

,~vlaleriafis" DMEM, FCS and glutamine were purchased from Gibco (U.K.). Dextran and glass beads were from Sigma Chemical Company (U.S.A.). Collagenase was from Serva Feinbiochemica GMBH (Germany). [~H]Acetyl-CoA was obtained from Amersham (U.K.). The monoclonal antibody against yon Willebrand factor was from Dakopatts (Denmark). The monoclonal antibody against GFAP was kindly provided by Dr D. Dahl (Massachusetts, U.S.A.). R ES U 1.TS

Immunostaining with h u m a n factor VIII antisera revealed the presence o f factor VIII antigen in both

Fig. 1. lmmunocytochemical demonstration of factor Vlll antigen in lhe cytoplasm of ECs subculture derived from the human embryonic CNS at 4 days in t'ilro (D1V 4l. Bar = 70/tm.

Glial cells and AChE activity

131

1

Fig. 2. Immunohistochemical demonstration of GFAP in the subculture of ECs (DIV 7). Very seldom, GFAP positive GCs can be revealed (arrow). Bar = 70/lm.

Fig. 4. Histochemical demonstration of AChE activity in the human embryonic subculture of ECs (DIV 7). Bar = 50 lLm. At higher magnification (insert), the reaction end-product can be revealed in the cytoplasm of the ECs. Bar = 25 #m.

was significantly higher ( P < 0.0t) t h a n its B u C h E activity. In the cortical endothelial culture, the A C h E activity was substantially decreased after longer culture times (7th a n d 12th DIV), while the B u C h E activity was not altered significantly after the same time intervals (Fig. 5). In the pial ECs, the A C h E activity was continuously reduced during the first 7 DIV, but a further decrease

Table I. ACHE, BuChE and C h A T activities in capillary fraction of fetal h u m a n cortex, cortical endothelial and glial culture AChE

BuChE

ChAT

( n m o l / m g protein/min) Capillaries

28.87±5.55* n=5

2.39_+0.93** n=5

Endothelial culture (DIV2)

27.02_+2.68 n=5

6.84+0.93 n=5

Glial cell culture ( D I V 10)

Fig. 3. ]mmunocytochemical demonstration of GFAP in human embryonic GCs culture (DIV 10). Bar = 70 #m.

ND n=5

ND n=5

Values represent means ± SD. Student's t-test : * P < 0.001 ; ** P < 0.01. N D : N o t detectable ; n = n u m b e r o f experiments.

0.003+0.001 n-5 ND n=5 ND n=5

132

M.P.&K,&SKI and P. KASA [] • --~---e--

30

T

Pial EC Pial EC Cortical Cortical

(BuChE) (ACHE) EC (BuChE) EC (ACHE)

"~

2o

c

£ ~

N

_>~

N

6

X

2o

T

N

I--'~

N

N

>

lO

u.I Z '~

\\\

lo

uJ >N Z LLI

A

C

AChE

o 1 2

3

4

5

6

7

8

9 10 1 1 12 13

days in vitro Fig. 5. AChE and BuChE activities in a human cortical and pial endothelial culture at different time points. In cortical subcultures of ECs, the AChE activity is substantially decreased after longer culture times, while the BuChE activity is not altered significantly at different times. The points are means_+SD of 5 experimental values from different cultures. in the enzyme activity was not observed (Fig. 5). The BuChE activity was similar in the cortical and pial ECs at every examined culture time (Fig. 5). The A C h E activity was significantly higher (P < 0.002) than the BuChE activity in both endothelial cultures. The effects of GCs on the A C h E and BuChE activities of the cortical ECs subcultures were also examined. The ECs developing in the presence of GCs, had nearly twice the A C h E activity of the control ECs subcultures on the 12th DIV (Fig. 6). The GCs cultures had no effect on the BuChE activity of the cortical subcultures of ECs (Fig. 6).

DISCUSSION

The present report provides the first biochemical evidence that in coculture (ECs with GCs) GCs can increase the A C h E activity present in the ECs. Our histochemical study also shows the presence of A C h E in the cytoplasm of cultured human ECs. The results reveal that a very low C h A T activity can be demonstrated in the capillary fraction obtained from the cortical samples. A considerable activity of A C h E and a lower activity of BuChE were measured in the capillary fraction. The abundant A C h E activity of the microvessel preparation might stem from both ECs and red blood cells in it. The decrease in A C h E activity

A

0

BuChE

Fig. 6. (A) Activities of AChE and BuChE in GCs cultures (DIV 22). (B) Activities of AChE and BuChE in subcultures of ECs (DIV 12). (C) Activities of AChE and BuChE in ECs subcultures (D|V 12) in presence of GCs (DIV 22). Results are means + SD of 5 experimental values from different cultures. Statistical significance was determined by using Student's t-test : * P < 0.002: ** P > 0.1.

of a cortical ECs, during the period in culture, may be due to the fact that the cerebral endothelium loses different biochemical properties. It has been suggested that astrocytes may induce certain properties in the cerebral endothelium. Beck et al. (1986) found that the Na +, K+-ATPase and non-specific alkaline phosphatase of rat cerebral ECs were markedly increased when the ECs were cocultured with GCs. Maxwell et al. (1987) demonstrated that GCs release a protein which induces de novo synthesis of 7-glutamyl transpeptidase in cerebral microvessel ECs. In our experiments, the presence of GCs increased the A C h E activity of the cortical ECs, supporting the suggestion that GCs are not only structural elements in the CNS, but also have a functional (biochemical and physiological) role. It is well known that the most characteristic feature of Alzheimer's disease is the cholinergic hypofunction, but a blood brain barrier (BBB) dysfunction is additionally presumed (Glenner 1985; Hardy et al., 1986; Mooradian, 1988). To our knowledge, embryonic human brain ECs have not been applied for study of BBB dysfunction. Our results (Kfisa et al., 1991 : and presented here) lead us to suggest that embryonic human brain ECs cocultured with GCs may be a useful tool for studying the BBB dysfunction in different neurological diseases (such as Alzheimer's disease).

Glial cells and AChE activity REFERENCES Alborch E. (1981) Cholinergic receptors in the cerebral arteries of the goat. In : Cerebral Microcirculation and Metabolism (Cervos-Navarro J. and Fritschka E., eds), pp. 285 291. Raven Press, New York. Arneric S. P., Honig M. A., Milner T. A., Greco S., ladecola C. and Reis D. J. (1988) Neuronal and endothelial sites of acetylcholine synthesis and release associated with microvessels in rat cerebral cortex: ultrastructural and neurochemical studies. Brain Res. 454, 11 30. Beck D. W., Roberts R. L. and Olson J. J. (1986) Glial ceils influence membrane-associated enzyme activity at the blood-brain barrier. Brain Res. 381, 131-137. Diglio C. A., Grammas P., Giacomelli F. and Weiner J. (1982) Primary culture of rat cerebral microvascular endothelial cells. Lab. Invest. 46, 554-563. Dob6 E., Hlavati I. and K~isa P. (1992) Decreased cholinergic innervation of intracerebral microvessels in Alzheimer's disease brain tissue. Neurol. Rev. in press. Ellman G. L,, Courtney K. D., Andrews V. Jr and Featherstone R. M. (1961) A new and rapid colorimetric determination of acetyl cholinesterase activity. Bioehem. Pharmac. 7, 88 95. Estrada C., Hamel E. and Krause D. N. (1983) Biochemical evidence for cholinergic innervation ofintracerebral blood vessels. Brain Res. 266, 261 270. Estrada C., Triguero D., Munoz J. and Sureda A. (1988) Acetylcholinesterase-containingfibers and choline acetyltransferase activity in isolated cerebral microvessels from goats. Brain Res. 453, 275 280. Fonnum F. (1975) A rapid radiochemical method for the determination of choline acetyltransferase. J. Neurochem. 24, 407-409. Glenner G. G. (1985) On causative theories in Alzheimer's disease. Human Path. 16, 433-435. Gonzfilez J. L. and Santos-Benito F. F. (1987) Synthesis of acetylcholine by endothelial cells isolated from rat brain cortex capillaries. Brain Res. 412, 148 150. Grammas P., Diglio C. A., Marks B. H., Giacomelli F. and Wiener J. (1983) Identification of muscarinic receptors in rat cerebral cortical microvessels. J. Neuroehem. 40, 645~651.

133

Hamel E., Assumel-Lurdin Ch., Edvinsson L., Fage D. and MacKenzie E. T. (1987) Neuronal versus endothelial origin of vasoactive acetylcholine in pial vessels. Brain Res. 420, 391-396. Hardy J. A., Mann D. M. A., Wester P. and Winblad B. (1986) An integrative hypothesis concerning the pathogenesis and progression of Alzheimer's disease. Neurobiol. Aging 7, 48~502. K~tsa P., P~.k~iski M., Jo6 F. and Lajtha A. (1991) Endothelial cells from human fetal brain microvessels may be cholinoceptive, but do not synthesize acetylcholine. J. Neurochem. 56, 2143 2146. Kreutzberg G. W., Kaiya H. and T6th L. (1979) Distribution and origin of acetylcholinesterase activity in the capillaries of the brain. Histochemistry 61, 111-122. Lowry O. H., Rosebrough N. J., Farr A. L. and Randall R. J. (1951) Protein measurement with the Folin phenol reagent. J. biol. Chem. 193, 265---275. Maxwell K., Berliner J. A. and Cancilla P. A. (1987) Induction of 7-glutamyl transpeptidase in cultured cerebral endothelial cells by a product released by astrocytes. Brain Res. 410, 309-314. Mooradian A. D. (1988) Effect of aging on the blood-brain barrier. Neurobiol. Aging 9, 69 86. Pfik~ski M., Kfisa P., Jo6 F. and Wolff J. R. (1990) Cerebral endothelial cell-derived laminin promotes the outgrowth of neurites in CNS neuronal cultures. Int. J. devl. Neurosci. 8, 193 198. Parnavelas J. G., Kelly W. and Burnstock G. (1982) Ultrastructural localization of choline acetyltransferase in vascular endothelial cells in rat brain. Nature 316, 724 725. Sternberger L. A., Hardy P. H. Jr, Cuculis J. J. and Meyer H. G. (1970) The unlabelled antibody-enzyme method of immunocytochemistry. Preparation and properties of soluble antigen-antibody complex (horseradish peroxidase-antihorseradish peroxidase) and its use in the identification of spirochetes. J. Histochem. Cytochem. 18, 315 333. Tsuji S. (1974) On the chemical basis to thiocholine methods for demonstration of acetylcholinesteraseactivities. Histochemistry 42, 99-110.