Neuroscience Research, 13 (1992) 1-17 © 1992 Elsevier Scientific Publishers Ireland, Ltd. 0168-0102/92/$05.00 NEURES 00504

Research Reports

Adenosine and its nucleotides stimulate proliferation of chick astrocytes and human astrocytoma cells Michel P. Rathbone 1, Pamela J. Middlemiss t, J.-K. Kim 1, John W. Gysbers 1, Susan P. DeForge 1, R.W. Smith 2,3 and D.W. Hughes 3 i Departments of Biomedical Sciences (Neurosciences) and Medicine (Neurology), McMaster University Health Science Centre, 2 McMaster Regional Centre for Mass Spectrometry and 3 Department of Chemistry, McMaster University, Hamilton, Ontario (Canada) (Received 30 July 1991; Accepted 4 September 1991)

Key words." Adenosine; Adenine nucleotide; Astrocyte; Purine receptor; Adenosine receptor; Mitogen

SUMMARY

Aqueous extracts of the brains of 18-day-old white Leghorn chicken embryos contain several substances that stimulate proliferation of primary cultures of chick brain astrocytes. Most of the mitogens are peptides. Purification of one mitogenic fraction was obtained by centrifugation, passage through Amicon Diaflo membranes of nominal molecular weight cutoffs 30, 1 and 0.5 kDa, ion exchange chromatography and reverse phase high performance liquid chromatography (HPLC) using a Deltapak CIs column. The mitogenic fraction contained no amino acids. On the basis of its behaviour on thin layer chromatography, its ultraviolet absorption spectrum, its 1H and 31p nuclear magnetic resonance spectra and its behaviour on positive and negative ion fast atom bombardment mass spectrometry, the mitogenic material was identified as adenosine5'-monophosphate (AMP). Other adenosine compounds including adenosine, ADP and ATP also stimulated proliferation of and [3H]leucine incorporation into primary cultures of astrocytes. Nitrobenzylthyioinosine (NBTI), an inhibitor of nucleoside transport, did not prevent the stimulation of [3H]leucine incorporation into cultured astrocytes. Polyadenylic acid (Poly A), that mimics the effect of adenosine at adenosine receptors, also stimulated proliferation of the astrocytes. The effects of adenosine and Poly A were not inhibited by 1,3-dipropyl-8-(2-amino-4-chlorophenyl)xanthine (PACPX) but were inhibited by 1,3-dipropyl-7-methylxanthine (DPMX), indicating that adenosine and Poly A acted at the cell surface, likely through adenosine A 2 receptors. The stimulatory effect of ATP was biphasic. The proliferative effect of low, but not of high, concentrations of ATP were abolished by DPMX. The purinergic P2 receptor agonist 2-methylthioATP and, at higher concentrations, the P2y agonist, ot,3-methyleneATP also stimulated incorporation of [3H]thymidine. These data indicate that high concentrations of ATP stimulate cell proliferation through at a Pc, possibly a P2y receptor. These results have considerable biological significance. After brain injury, or when cells in brain die or become hypoxic, nucleotides and nucleosides are released from the cells. Their extracellular concentrations can exceed those required to stimulate astrocyte proliferation in vitro. Therefore they may be partly responsible for gliotic changes following cell death in brain.

Correspondence: M.P. Rathbone, McMaster University Health Science Centre, 4N25-1200 Main Street West, Hamilton, Ontario, Canada L8N 3Z5.

2

INTRODUCTION Astroglial cells play many key roles in the nervous system ~5,2,~,34 Astrocytes divide and become reactive in response to a variety of types of central nervous system injury 3,36,45. Widespread reactive astrogliosis occurs in association with degenerative diseases of the central nervous system such as Alzheimer's disease and Down's syndrome ~,12.32,46. Moreover, the most common forms of benign and malignant intrinsic brain tumours arise from cells of astrocytic lineage 4 Clearly, a knowledge of factors that control astrocyte proliferation and differentiation is important to an understanding of the various roles of these cells in the development of the nervous system and in its responses to disease and injury 15,29,34. A number of substances stimulate proliferation of astrocytes in cell culture. These include low molecular weight materials separated from embryonic chick brain ~', glial proliferation factors (GPF) 2 and 4 21, interleukin-I (II-1) 20, and fibroblast growth factor (FGF) 2~ Our objective has been to determine the nature of the stimuli that cause astrocytc mitosis under physiological and pathological conditions. As a first step towards this objective we isolated from brains of 18-day chick embryos a number of factors with molecular weights of less than 24000 Da that stimulate proliferation of chick astrocytcs in cell culture 5,6,49. Mitogenic activity in aqueous extracts of chick brains was purified through a number of steps 3~. Four fractions with molecular weight less than 2500 Da had mitogenic activity. Two of these contained no amino acids. Multiple physical chemical analyses identified one fraction as guanosine-5'-monophosphate (GMP)33 Guanosine, G D P and GTP also stimulated proliferation of cultured astroblasts. The proliferative effects of guanosine were inhibited in a dose-dependent fashion by theophylline, characteristic of phenomena mediated by purinergic receptors. Purinergic receptors are divided into two types, PI and P2. The natural ligands for P~ receptors are adenosine and possibly AMP, whereas ATP and ADP are believed to be the natural ligands for P2 receptors 3,53. PI receptors are divided on the basis of their affinity for various adenosine analogues and the ability of various substances to inhibit them into two subclasses, A l and A237'52. P2 receptors are also subdivided on the basis of their responsiveness to ATP analogues into P2x, P2y and P2z subclasses '~ Our data led us to investigate whether the other mitogenic fractions separated from brain were peptides or whether they, too, were nucleotides. In this paper we report the identity of a second mitogenic fraction that contained no amino acids as adenosine-5'mormphosphate (AMP). We also report that adenosine, ADP and ATP also stimulate proliferation of primary cultures of astrocytes. They appear to do so through activation of two separate types of purinergic receptors on the cell surface. Moreover, these compounds are highly active in stimulating proliferation of two clonal human astrocytoma cell lines.

MATERIALS AND METHODS

Materials Unless otherwise noted all laboratory chemicals were obtained from Fisher Scientific (Fairlawn, N J). Adenosine, AMP, ADP, a,/3-methyleneATP, nitrobenzylthioinosine (NBTI), and 5' polyadenylic acid (poly A) were obtained from Sigma Chemical Co. (St. Louis, MO). 1,3-dipropyl-7-methylxanthine (DPMX), 1,3-dipropyl-8-(2-amino-4-chlorophenyl)xanthine ( P A C P X ) a n d 2-methylthioATP were purchased from Research Biochemical Inc. (Natick, MA). All materials used in the tissue culture work, unless

otherwise indicated, were obtained from Gibco (Grand Island, NY). Methyl-[3H]thymi dine (specific activity 40-60 C i / m m o l ) was from ICN (Costa Mesa, CA). SKMG-1 and U373 are two human astrocytoma cell lines obtained from Dr. G. Cairncross (London Cancer Center, London, Ontario). SKMG-1 is a glial fibrillary acidic protein (GFAP) negative line, originally isolated by Pfreundschuh et al. 43 (originally designated AJ). U373 is a G F A P positive line, originally isolated by Jorgen Fogh of Sloan Kettering Institute ~6

Purification of mitogenic substances from chick-embryo brain The mitogenic substances were purified as described by Kim et al. 33. Essentially the supernatants were subjected to pressure-ultrafiltration using a series of membranes. The retentate was concentrated by lyophilization. The lyophilized retentate was then chromatographed, firstly by ion exchange chromatography, and secondly by reversed phase high pertbrmance liquid chromatography (HPLC). After each elution the fractions were pooled and their mitogenic activity was determined by bioassay. Finally, the PH3 fractions were rechromatographed by H P L C at least twice to purify them to homogeneity.

Preparation of astrocyte-like cells from chick brains Astrocytes from 10-day-old white Leghorn chick embryo brains were isolated as previously described 6,7,33. The meninges were removed from the brains that were separated from decapitated chick embryos. By a combination of trypsinization and mechanical sieving through 63/xm Nitex mesh (Small Parts Inc., Miami, FL), the brain tissue was digested and dispersed. The astrocytes were selected by differential adhesion 26 In order to destroy over 95% of any contaminating microglia, the cell suspension was incubated overnight at 37°C in Eagle's minimal essential medium (MEM) with 5 mM L-leucine methyl ester ~9,21 The astrocytes were then grown in Eagle's MEM containing 33.3 mM D-glucose, 5 /xg/ml human transferrin, 2 × 10 -8 M progesterone, 100 tzM putrescine, 3 x 10 -8 M sodium selenate, 5 / z g / m l insulin (all from Sigma Chemical Co.) and 10% fetal bovine serum (FBS) for 48 h. They were harvested by trypsinization, and subcultured. Into each tissue culture flask (250 ml) approximately 2 x 106 cells were added. They were grown for another 48 h in the same medium. The cells used in these studies had the shape of glial precursor cells 48. After 7 days in medium containing 10% FBS, over 90% of the cells stained positive for G F A P 6 G F A P may be present in astroblasts, astrocytes and senescent astrocytes. These cells are best termed astrocyte-like cells, but for convenience, we shall refer to them as astrocytes.

Bioassay of methyl-[ 3H]thymidine incorporation into astrocytes The bioassay was a modification 33 of an established method 23. Astrocytes (5 × 103/well) in 200 /xl MEM containing 10% FCS were transferred to 96-well Falcon microtest plates (Fisher Scientific). After 48 h they were serum-restricted by replacing the medium with MEM containing 0.2% or 0% FBS for another 48 h. That medium was then replaced with 200 #1 of MEM containing 1% FBS. Fresh medium was supplemented into the control wells with either 0%, 1% or 10% FBS. Varying concentrations of test substances or vehicle controls were added to the appropriate wells in a final volume of 10/xl. O n e / z C i of [3H]thymidine (10/xl/well) was then added. After 16 to 24 h the medium was removed, the wells rinsed, and the ceils harvested by the same

procedure as Gospodarowicz et al. :~. The cotton-tipped applicators containing the cells were processed through 10% and 5% TCA, then 95% ethanol and dried overnight and their radioactivity determined by liquid scintillation counting. In experiments using nitrobenzylthioinosine (NBTI), an adenosine uptake inhibitor, NBTI was added simultaneously with serum or the other test substances to each well for a final concentration of 1 /zM.

Assay of L-[ 3H]leucine incorporation into astrocytes Since NBTI interacts with the nucleoside transport system to inhibit nucleoside uptake 18 it also inhibits the uptake of [3H]thymidine by the cells. Therefore, rather than measuring the effect of NBTI on purine-stimulated increase in [3H]thymidine incorporation into the cells, we studied its effect on [3H]leucine incorporation into cellular proteins. We used a similar technique to that described above for thymidine incorporation except that we added 1 lzCi of L-[3H]leucine (161 C i / m m o l ) (Amersham), to each well 24 h before they were harvested. The cells were harvested in exactly the same manner as that described above for thymidine incorporation. The radioactivity in each fraction was determined by liquid scintillation counting.

Calculation of results Incorporation of [3H]thymidine or L-[3H]leucine into control cells, cells stimulated with 10% serum, with brain extracts, or with adenosine nucleotides increased linearly with time over a period from 8 to 30 h. Therefore, the ratios of counts between treated and control cultures during the periods of our experiments (16-24 h) remained unchanged. The mean radioactivity in the cells in each control well was approximately 4000 counts per minute (cpm), varying from experiment to experiment. The degree of stimulation of [3H]thymidine incorporation into the treated cells are expressed as a ratio: cpm in cells in medium with 1% FBS and test substances/cpm in cells with control medium containing 1% FBS. Inter-experimental variations due to differences in the absolute cpm incorporated into control and experimental cultures were minimized by this procedure.

Nuclear magnetic resonance (NMR) spectroscopy All NMR spectra were recorded on a Bruker AM-500 spectrometer. Proton spectra were acquired at 500.137 MHz using a 5 mm broadband inverse probe. Spectra were obtained in 16 scans for the 5'-AMP sample and 56 scans for the PH3 fraction sample in 16K data points over a 5.000 kHz spectral width (1.638 s acquisition time). Sample temperature was maintained at 30°C by a Bruker BVT-1000 variable temperature unit. The residual H D O resonance was suppressed by presaturation for 1.0 s prior to acquisition. The free induction decay (FID) was processed using exponential multiplication (line broadening: 0.2 Hz) and was zero-filled to 32K before Fourier transformation. Phosphorus-31 N M R spectra were recorded at 202.459 MHz using the 5 mm broadband inverse probe. The spectra were acquired in 400 scans for the 5'-AMP sample and in 1800 scans for the PH3 fraction sample over a 29.411 kHz spectral width in 16K data points (0.279 s acquisition time). The 31p pulse width was 5.0 /~s (40° flip angle). A 0.5 s relaxation delay was used. The FIDs were processed using exponential multiplication (line broadening: 5.0 Hz). Proton decoupling was achieved by composite pulse decoupling.

The compounds used in this study were dissolved in 99.996% DzO (MSD Isotopes). Chemical shifts for the 1H spectra are reported in ppm relative to TMS using the residual HDO signal at 4.60 ppm as internal reference. 31p chemical shifts are reported in ppm relative to external 85% H3PO 4 in D20. Fast atom bombardment mass spectrometry Fast atom bombardment mass spectrometry (FAB/MS) was performed with a VG Analytical ZAB-E double focusing mass spectrometer. Xenon was employed as the primary atom beam with 8 keV energy. Thioglycerol containing 5% trifluoroacetic acid or glycerol containing 20% trifluoroacetic acid were used for the positive ion experiments and either thioglycerol or triethanolamine for the negative ion experiments. Typically, 5/xl of extract was mixed with 10/xl of matrix and 3 /xl applied to the FAB probe.

RESULTS Aqueous extracts of the brains of 18-day-old white Leghorn chick embryos contained fractions that were mitogenic for primary cultures of chick astrocytes. We purified the low molecular weight mitogenic fractions using differential centrifugation, ultafiltration through Arnicon membranes (nominal molecular weight cut-offs of 30 and 1 kDa) and retention on a Amicon membrane with a nominal molecular weight cut-off of 0.5 kDa 6 The mitogenic fractions in the retentate were further purified by ion-exchange chromatography on DEAE cellulose columns and then by reversed-phase HPLC 33. The HPLC column eluate contained 4 mitogenic fractions (Fig. 1). Two of these, designated PH3 and PH4M, contained no amino acids. We had previously identified the PH4M fraction as GMP 33 We subjected the PH3 fraction to ultraviolet absorption spectroscopy. Absorption was maximal at 257.5 nm at pH 2.14, 259.5 nm at pH 10.71 and 259.5 nm at pH 8.24 (Fig. 2). These ultraviolet absorption spectra were identical to those obtained using commercial 5'-AMP and to published reference spectra of AMP 25 Additional proof of the structure of the PH3 peak was provided by 1H and 31p nuclear magnetic resonance (NMR) spectroscopy. The I H spectra of the PH3 fraction and 5'-AMP (Fig. 3a and b, respectively) were essentially identical. Chemical shifts and coupling constants presented in Tables I and II, respectively, showed a close correspondence between the two samples and were in accord with previously reported data for 5'-AMP 11. In the 31p NMR spectra the chemical shift of the PH3 fraction was 0.728 ppm whereas that of 5'-AMP was 0.933 ppm (Fig. 3c and d, respectively). These results are still compatible with PH3 fraction being 5'-AMP since 31p chemical shifts are very sensitive to a variety of sample conditions including pH, concentration, temperature and the presence or absence of buffers or salts 8. These factors will also influence the 1H chemical shifts particularly those of the purine H-8 and H-2 protons (Table I). As final confirmation of the identity of the PH3 peak we examined it using F A B / M S in the positive and negative ion mode employing the matrices described in the experimental section. In positive ion FAB, low intensity ions at m / z 348 (M + H) + were observed from both the unknown material and authentic 5'-AMP. In negative ion mode (Fig. 4), ions at m / z 346 (M-H)- were observed from both the unknown material and 5'-AMP. These data indicated that the PH3 fraction contained 5'-AMP. We next tested whether AMP increased solely the incorporation of [3H]thymidine into the cultured astrocytes or whether it concomitantly increased cell numbers. In

a

!

PH3

I o

!

10 2

o

o.s~

PH1

PH2

--

LL_

t'o

3~0

-~ I

ol

t

.

o 3~

PH4M

2~0 RETENTION TiME (Minutes)

b

222 1 Fig. 1. Separation on reversed phase HPLC column (Deltapak Cts) of mitogenic activities purified from brains of 18-day chick embryos as described in the text. (a) Ultraviolet absorption profile of the eluate at 260 nm: (b) Mitogenic activity of the various fractions as determined by their ability to stimulate [3H]thymidine incorporation into chick brain astrocytes as described in the text. Only the four named fractions contained significant activity. Samples taken earlier than eluation of the PH1 fraction, between the fractions or after the PH4M fraction did not significantlystimulate thymidine incorporation as shown by the left-hand and right-hand bars. confirmation of our earlier data for other mitogens 33, we found that addition of A M P to the astrocyte culture increased the number of cells concomitantly with increasing [3H]thymidine incorporation into the cells (data not shown). Since A M P was active in stimulating both cell proliferation and [3H]thymidine incorporation into the cells, we questioned whether other adenine nucleotides or adenosine were active. We found that each compound was active in a concentration-dependent manner (Table III). Both adenosine and A T P had biphasic activity-concentration curves. Most of the activity of adenosine was observed at between 0.01 and 1 nM. Adenosine had lesser activity when added to the cultures at concentrations from 1 / ~ M to 1 mM. The corresponding peaks for A T P had approximate maxima at 0.1 nM and 300 /zM ( 3 / z M - 3 m M ) respectively. A D P produced maximum stimulation of [3H]thymidine incorporation into the astrocytes at 30 /zM. Adenine also stimulated an increase in [3H]thymidine incorporation with maximal stimulation at 150/zM. We determined whether u p t a k e of adenosine into the astrocytes contributed to its ability to stimulate their proliferation. To examine this possibility: we treated the cultured astrocytes with nitrobenzylthioinosine (NBTI), a potent inhibitor of the nucleoside transport system ~8. At a concentration of 1 / z M , N B T I inh~ited nucleoside uptake into the cells by over 90% (data not shown). N B T I inhibited uptake not only of purine nucleosides but also of [3H]thymidine. Therefore, it was not meaningful to determine the effect of N B T I on adenosine-stimulated incorporation of [3H]thymidine into the

2.000

t.598 I

1.196

o 0.794

0.392

'~

'~... !

200

230

260

290

320

350

Wavelength

Fig. 2. Ultraviolet absorption spectroscopy of the PH3 fraction. Absorption maxima arc shown at 257.5 nm at p H 2.14 (solid line), 259.5 n m at pH 10.71 (dotted line) and 259.5 n m at pH 8.24 (dashed line). The spectra were determined using a Beckman DU-7 spectrophotometer.

DNA of cultured astrocytes. We determined, therefore, whether adenosine and ATP also stimulated incorporation of [3H]leucine into the proteins of cultured astrocytes (Fig. 5). Both adenosine and ATP stimulated incorporation of [3H]leucine into acid-insoluble material in the astrocytes. This stimulation was unaffected by NBTI (1 tzM). This indicated that adenosine did not need to be transported into the cells to stimulate astrocyte proliferation. Rather we concluded that it exerted its effects at the external surface of the astrocytes. We used another approach to confirm that adenosine indeed stimulated cell proliferation by acting at the astrocyte surface. Polyadenylic acid (poly A), an adenylic acid

TABLE I

1H A N D 31p C H E M I C A L SHIFTS O F 5 ' - A M P A N D PH3 F R A C T I O N IN D 2 0 Proton

Chemical shift (ppm) 5'-AMP

PH3 Fraction

H-8 H-2 H-I' H-2' H-3' H-4' H-5' H-5"

8.345 8.115 5.999 4.613 4.359 4.247 3.996 3.978

8.427 8.225 6.044 4.634 4.377 4.264 4.021 3.992

Phosphorus

Chemical shift (ppm) 5'-AMP

PH3 Fraction

5'-P

0.933

0.728

a

i I

i

I

J

80

i

7.5

7.0

6.5

6.0

i

5.5

5.0

4,5

40

PPM

C

d

15

10

5

0

-5

-10

15

PPM

Fig. 3. (a) Proton t H-NMR spectrum of PH3 fraction recorded at 500 MI-Iz in D20 at 30°C. The signals of the PH3 fraction at 8.427, 8.225, 6.044, 4.634, 4.377, 4.264, 4.021 and 3.992 ppm were assigned to H-8, H-2, H-I', H-2', H-3', H-4', H-5' and H-5", respectively. (b) These assignments were confirmed by comparison with a 1H-NMR spectrum of 5'-AMP. (c) 3tp-NMR spectrum of PH3 fraction recorded at 202 MHz in D20 at 300(2. A single peak was observed at 0.728 ppm and this indicates a single phosphorus atom in the molecule. (d) A sample of 5'-AMP produced a 3ip spectrum almost identical to that of the PH3 fraction.

homopolymer of mean chain length 200 bases, was added to the cultures. Its relatively high molecular weight prevents it from being taken up into the cells through the normal transport mechanisms. Poly A stimulates coronary flow in a manner similar to adenosine and ATP and is relatively stable in short term cultures of endothelial cells 4li Poly A stimulated [3H]thymidine uptake into chick astrocytes (Fig. 6).

M a n y effects o f extracellular adenosine and a d e n i n e nucleotides are mediated by their interactions with cell-surface purinergic receptors 53. T h e r e f o r e we tested w h e t h e r their mitogenic effects were also m e d i a t e d in this way. If they were, then specific antagonists of adenosine receptors should inhibit the mitogenic effects. T h e adenosine A~ r e c e p t o r antagonist, 1,3-dipropyl-8-(2-amino-4-chlorophenyl) xanthine ( P A C P X ) 47 did not significantly inhibit the stimulation of [3H]thymidine incorporation by adenosine (Fig. 7a) or by A T P (Fig. 7b). In contrast, however, the adenosine A 2 r e c e p t o r antagonist, 1,3-dipropyl-7-methylxanthine (DPMX)10,50,51 inhibited the stimulation of proliferation by adenosine (Fig. 7c) and by poly A (Fig. 6) as well as by low concentrations o f A T P . In contrast, D P M X did not abolish stimulation by high concentrations o f A T P (from 10 -6 to 10 -3 M) (Fig. 7d). This indicated that the effects of adenosine might be m e d i a t e d t h r o u g h A 2 receptors. It also implied that the mitogenic effects of low c o n c e n t r a t i o n s of A T P might be due to its d e g r a d a t i o n extracellularly to adenosine by

a

100

346

313

5G

,I, hi

427

lh I1,1I~tl,I IIIIIJr ~ll, 350

I It,I III~1 I lll Ill l

400

m/z

450

5OO

450

500

41 100 427 Z I11 1-.. W

_J W

5O

lJJ

IJ,Itlll 350

400

Fig. 4. In these negative ion FAB mass spectra the ions observed are characteristic of the test sample and the thioglycerol matrix (a) and thioglycerol only (b). Comparison of the ions observed from the test sample in (a), that is, m / z 346 (M-H)- with authentic 5'-AMP characterize the unknown material as 5'-AMP. All other ions observed in (a) are due to the matrix and are shown in (b). The vertical axis (% relative intensity) is a measure of the number of ions detected. The horizontal axis, the mass to charge ratio (m/z), represents the mass of the ions detected.

10 T A B L E II I H - I H A N D 1H-31P C O U P L I N G C O N S T A N T S O F 5'-AMP A N D PH3 F R A C T I O N IN D 2 0

Proton

Coupling constant (Hz) " 5'-AMP

PH3 Fraction

3j b 1',2' 2',3' 3',4' 4',5' 4',5"

5.8 5.1 3.7 4.8 5.3

5.6 5.3 3.t? 4.6 5.5

2j 5',5"

-11.8

-11.6

2.9 3.2

,~.'~0 3.3

3j 5',P 5",P

~' Coupling constants are accurate to 0.3 Hz. b j is the symbol for coupling constant and the superscript refers to the number of chemical bonds between the coupled nuclei.

ectoenzymes 2. In contrast, the stimulation of astrocyte proliferation by higher concentrations of ATP was unlikely to be mediated through A2 receptors. These data led us to the hypothesis that ATP exerted its principal effects through activation of purinergic P2 receptors. To test this, P2 agonists were added to the cultured astroblasts. The preferential P2y agonist, 2-methylthioATP (2-meSATP) stimulated [3H]thymidine incorporation and proliferation at concentrations from 10 -1° M to 10 -~ M (Fig. 8). The P2x agonist a,/3-methyleneATP was also active over this range of concentrations. It was slightly less active than the P2y agonist at the lower concentrations. These data indicate that the cells respond to stimulation by ATP through P: receptors that are known to exist in large numbers on the surface of astrocytes 4~ Most probably these are of the P2y subtype.

2.0

1.5

~

1.0

"~

0.5

0

0% FCS

1% FCS

Ado

Fig. 5. Effects of medium containing 0% FCS, 1% FCS and 1 0 / z M adenosine (Ado) with 1 /zM NBTI (black) and without NBTI (white) on [3H]leucine incorporation (vertical axis) into cultured astrocytes.

11 4

-

cO

ri/'l

0 LL

I

I

I

I

I

I

I

12

11

10

9

8

7

6

Concentration (-log molar)

Fig. 6. The effect of polyadenylic acid (Poly A) on [3H]thymidine incorporation into cultured chick astrocytes compared to astrocytes cultured in control medium. The fold stimulation compared to control medium is shown on the vertical axis and the concentration of Poly A added on the horizontal axis. White squares: Poly A alone. Black circles: Poly A + 10 ~ M DPMX.



IS~O ......................OiDENOSINE+PACPX •

2

• ADENOSINE

8 "5 E u)

,

1

o

.I

I

I

I

12

i

,

,

9

,

I

I

6

I

• ATP

,~ ~-

/

I,.

I!

b . . . . . . . . . . 12

3

-

9

~,Z,!

6

3

..........................

\I/

8 -5 E u) "o

\ T

,,o

~,,,~

d

0 i

t2

i

l

l

i

9 Concentration

i

i

6 (- log

i

I

I

3 molar)

I

I

,

12

,

,

,

.

.

9 Concentration

.

.

6 (- log

.

.

~

~ , 3

molar)

Fig. 7. Effects of PACPX and DPMX on stimulation of [3H]thymidine incorporation into cultured astrocytes by adenosine and ATP. (a) Stimulation of incorporation by adenosine alone or in combination with 10 izM PACPX. (b) Stimulation of incorporation by ATP alone or in combination with 10 /xM PACPX. (c) Stimulation of incorporation by adenosine alone or in combination with 10 /zM DPMX. (d) Stimulation of incorporation by ATP alone or in combination with 10 # M DPMX. Vertical axis indicates stimulation relative to [3H]thymidine incorporation into astrocytes cultured in medium without the above additions. Horizontal axis indicates concentrations of adenosine (a and c) or ATP (b and d).

12 T A B L E Ill S T I M U L A T I O N OF [ 3 H ] T H Y M I D I N E INTO A S T R O C Y T E S BY A U T t t E N T I C C O M M E R C I A L AI)t,NINE, A D E N O S I N E A N D A D E N I N E N U C L E O T I D E S Purine

Adenine Adenosine (major peak) Adenosine (minor peak) AMP ADP ATP (major peak) ATP (minor peak)

R a n g e of activity

1.5 ,aM- 1.5mM 0.01 n M - 1.0 nM 1.0 ,aM

Peak

Maximal fold stimulati~m (mean + SEMI

130 /xM 1.0 nM

1.84+(I.201n=8) 2.03 + (1.09 (~z = 9)

1.0 ,aM

1.64+1).17 (Jr = 9)

1.0 mM

50 n M - 5 0 ,aM 300 nM 3 mM 3.0 , a M - 3 , 0 m M 0.01 n M -

500 30 300

1.0nM

nM ,aM ,aM

1.87+0.26(n-12) 2.31 -+0.37(n 12) 2.47_+0.21(n=12)

1.0nM

1.52+0.11(n=12)

In each case the stimulation is expressed as a relative to the incorporation of [3H]thymidine into astroblast cultures treated identically except that the purine was omitted and instead only its solvenl (PBS) was added as described in the methods section.

We took precautions to eliminate contaminating microglial cells from the astrocyte cultures. But a small number of cells that did not stain positively for GFAP might have been microglia. Microglia produce several astroglial mitogens such as II-1, GPF 2 and GPF 4 20.21. To test whether adenosine and its nucleotides stimulated astrocyte proliferation by stimulating contaminating microglia to produce mitogens, or whether they could stimulate astrocytes directly, we obtained two clonal human astrocytoma cell lines. In this way we could test the effects of the purines on astrocyte-derived cells that contained no contaminating microglia. One cell line, U373, was GFAP +, the other, SKMG-1, was G F A P - . Both cell lines were stimulated to proliferate by adenosine and

3.0-

g

2.0

[

E ¢o O LL

T

& U,

1.0

Z

!i 10 8 6 Concentration (-log molar) Fig. 8. Stimulation of [3H]thymidine incorporation into astroblasts by 2-methylthioATP (white) and a,C3-methy l e n e A T P (hatched) at concentrations of 1 0 - m , 1 0 - s and 10 .-6 M.

13

12t

a

,° i 8

6--

4--

2q

i

0--

£3

I

I

I

I

I

I

I

I

I

I

I

12

11

10

9

8

7

6

5

4

3

Concentration (-log molar)

b

u-

4 2 0--1

r

I

I

I

I

I

J

I

I

I

12

11

10

9

8

7

6

5

4

3

Concentration (-log molar) Fig. 9. Stimulation o f [3H]thyrnidine incorporation into clona] human astrocytoma cell lines, U373 (a), and

SKMG-1 (b) by adenosine (black circle) and ATP (white square).

A T P (Fig. 9). In both cases the cells were stimulated to a greater extent by ATP than by adenosine. U373 was stimulated maximally over 10-fold by ATP and 6-fold by adenosine. Similarly, SKMG-1 was stimulated almost 5-fold by ATP and almost 3-fold by adenosine. These results indicate that adenosine and A T P can stimulate proliferation of astrocyte-derived cells even in the absence of microglia. DISCUSSION

O u r data indicate that in aqueous extracts of tissues, adenine nucleotides may co-purify closely with some low molecular weight peptide mitogens. Since purine nucleosides and nucleotides are themselves mitogenic, this has important implications for investigators who are purifying potentially mitogenic peptides from natural sources. It is clearly important to ensure that the materials are free of contaminating nucleotides. It is also important to exclude nucleotides from the culture medium when testing substances for their mitogenic activity in vitro.

14 The stimulation of cell proliferation by adenine and adenosine derivatives may bc all ubiquitous phenomenon in multiceIlular organisms. Certainly, N%substituted adenine and adenosine compounds are naturally occurring cytokinins in plants 3544. In contrast to the extensive literature on plant cytokinins, however, there are few reports of the mitogenic activity of adenosine and its derivatives on animal cells. Some investigators have reported that these compounds may play a nutritional role in stimulating cell proliferation in vitro H. Other investigators found that adenosine acts synergistically with other growth factors to stimulate cell proliferation. For example, N %(a2-isopentenyl) adenosine enhanced proliferation of phytohemagglutinin-stimulated lymphocytes iv. Hovi et al. 31 found that adenosine had a similar effect. Hcppel et al, :zs investigated the interaction of adenosine, ATP and other nucleotides with a series of mitogenic factors including epidermal growth factor, insulin and platelet derived growth factor. They found ATP alone had little mitogenic activity. It was, however, highly synergistic when added together with other mitogens. They reported that adenosine had little mitogenic activity on 3T3 cells and little or no synergism with other growth factors. Indeed, it was totally inactive when tested on Swiss 3T6 fibroblasts. Purine nucleosides and nucleotides arc also capable of stimulating proliferation of cultured cells directly rather than acting synergistically with other factors 3tl.33.3,~. Our data imply that adenosine, AMP, ADP and A T P can each alone stimulate astrocyte proliferation. In general adenosine, AMP, ADP and ATP stimulated proliferation to a lesser extent at their optimal concentrations than did their guanine-based counterparts 33. However, the optimal concentrations of the adenine-based compounds were lower than the guanine-based counterparts. This raised the possibility that guanosine and adenosine stimulate astrocyte proliferation through different mechanisms. Our laboratory is currently investigating this. Previously most investigators have not studied in any detail the receptors that mediate the purinergic stimulation of cell proliferation. Meininger and Granger 3,~ found that a non-specific adenosine receptor that had some pharmacological characteristics of both A 1 and A 2 receptors mediated the stimulation of endothelial cells by adenosine. The overall weight of their pharmacological and biochemical evidence indicated that it behaved more as an A= than an A I receptor. The effects of adenosine receptor antagonists in our experiments indicated that an A 2 receptor was likely involved. Moreover it provided evidence that when ATP was added to the cultures at low concentrations its effects could be blocked by an A receptor antagonist. This would be compatible with the hypothesis that at low concentrations ATP was being degraded extracellularly by ectoenzymes and converted to adenosine, which then exerted its effects through an adenosine receptor. Such an hypothesis has been previously proposed by Bruns 2. In contrast, the effects of higher concentrations of A T P were not inhibited by an A2 receptor antagonist, DPMX. Also relatively hydrolysis-resistant analogues of ATP, including 2-methylthioATP, a specific Pxy receptor agonist, stimulated astroblast proliferation. Therefore, it is probable that at higher concentrations ATP exerts its predominant effects through a P2 receptor. Both 2-methylthioATP and a,/3-methyleneATP were active at concentrations as low as 0.1 nM. This is similar to their potency in other systems, for example, regulating contraction of brain arteries 27. The differential effects of 2-methylthioATP and a,/3methyleneATP were not sufficiently great to permit us to ascribe a P2 receptor subtype to the effects of these compounds on astroblast proliferation. However, 2-methylthioATP was slightly more active at lower concentrations and a,/3-methyleneATP at higher

15 ones. This raises the possibility that the effects are mediated through a Pzy receptor. We are currently investigating this possibility in more detail. Adenosine stimulated astrocyte proliferation and [3H]leucine incorporation into proteins in the presence of NBTI, an inhibitor of the nucleoside transport system of cells. This implied that the mitogenic effect of adenosine was probably mediated through receptors on the surface of the astrocytes. This would certainly be compatible with the findings that its effects were blocked by antagonists of adenosine receptors. Moreover, Poly A, which is capable of mimicking adenosine at extracellular adenosine receptors and is not rapidly degraded in culture 39,40,41, also stimulated cell proliferation. A molecule of this size is unlikely to be internalized rapidly by cultured astrocytes. Furthermore the stimulation of astroblast proliferation by poly A was inhibited in a concentration-dependent manner by DPMX. This indicates that the effects of poly A were likely mediated through a n A 2 receptor. The observation that both adenosine and its nucleotides can stimulate proliferation of cultured astrocytes and astrocytoma cells, apparently through two separate purinergic receptors, has significant biological implications. After brief periods of brain anoxia, adenosine and purine nucleotides are released in concentrations that exceed micromolar 24. Indeed, this has led to the suggestion that adenosine may be one factor responsible for stimulation of cell proliferation after hypoxia and tissue injury 33,38,3,) Clearly, under conditions in which cells are not simply damaged but the continuity of their external membrane is breached, large quantities of nucleosides and nucleotides will be released into the extracellular fluid. Moreover, as cells die their ribonucleic and deoxyribonucleic acids are hydrolysed. These may also stimulate proliferation of adjacent astrocytes. Significantly, reactive gliosis is observed around areas of neuronal death or tissue injury in adult animals. In the central nervous system cell death is followed initially by microglial proliferation 22 and subsequently by astrogliosis. It is tempting to speculate that adenosine and adenine nucleotides released from dying cells stimulate proliferation of microglial cells as well as astrocytes. It is possible that they also stimulate production of IL-1 and other cytokines by the microglia. These cytokines would act synergistically with adenosine and its nucleotides in stimulating astroglial proliferation and reactive astrogliosis. One direct test of the hypothesis that purine nucleosides and nucleotides play a role in the glial response to cell injury would be to infuse these compounds into brain to determine their effect on astroglia and microglia. These experiments are currently under way in our laboratory. ACKNOWLEDGEMENTS

This work was supported by grants in aid of research from the Hospital for Sick Children Foundation (Toronto) and from the Hamilton Civic Hospitals Research Trust. We thank Brent Gabel for assistance with the tissue culture, Drs. J.G. Cairncross and D. Giulian tor their helpful discussions and suggestions, and Dr. J.G. Cairncross for the gift of SKMG-1 and U173 cells. REFERENCES 1 Allore, R.D., O'Hanlon, D., Price, R., Nielson, K., Willard, H.F., Cox, D.R., Marks, A. and Dunn, R.J., Gene encoding the beta subunit of SI00 protein is on chromosome 21: implications for Down syndrome, Science, 239 (1988) 1311-1313.

16 2 Bruns, R.F., Adenosine receptors: Roles and pharmacology, Ann. NYAcad. Sci., t~(13(199(D 211 22~. 3 Burnstock, G., Regulatory function of adenosine. In R.M. Berne, T.W. Rail and R. Rubio (l+ds.), Regulatoo' Function of Adenosine, Nijhoff, Boston, 1983, pp. 49-62. 4 Cairncross, J.G., Astrocytoma. In S. Federoff and A. Vernadakis (Eds.), Astrocytes: Cell Biology uml Patholo,q3' of Astrocytes. Academic Press Inc., Orlando, 1986, pp. 337-35{~. 5 Carlone, R.L. and Rathbone, M.P., Partial purification of a low-molecular weight growth factor from chicken brain, .I. Cell. Physiol., 124 (1985) 467-473. 6 Carlone, R.L., Kim, J-.K. and Rathbone, M.P., Purification of a chick brain-derived growth factor by reversed-phase high-performance liquid chromatography, J. Neurosci., 7 (1987) 2163-2167. 7 Carlone, R.L., Waters, B.W., Leonard, S.M. and Vijh, K.M., A low molecular weight chick brain derived growth factor is mitogenic for cultured astroglia from the chick embryo, Det,. Brain Res., 39 (1988) 97 1(14. 8 Cozzone, P.J. and Jardetsky, O., Phosphorus-31 fourier transform nuclear magnetic resonance study on mononucleotides and dinucleotides 1: chemical shifts, Biochernistry, 15 (1976) 4853-4859. 9 Cusack, N.J. and Hourani, S., Subtypes of P2-Purinoceptors: studies using analogues of ATP, Ann. N Y Acad. Sci., 603 (1990) 172-181. 10 Daly, J.W., Padgett, W., Shamim, M.T., Butts-Lamb, P. and Waters, .I., 1,3-Dialkyl-8-(psulfophenyl)xanthines: potent water soluble antagonists for A t and A , adenosine receptors, J. Med. CTwm., 28 (1985) 487-492. 11 Davies, D.B. and Danyluk, S.S., Nuclear magnetic resonance studies of 5'-ribo and deoxyribonucleotide structures, Biochemisto', 13 (1974) 4417-4434. 12 Duffy, P.E., Rapport, M. and Graf, L., Glial fibrillary acidic protein and AIzheimer-type senile dementia, Neurologg', 3(1 (19811) 778-782. 13 Eng, L.F., Astrocyte response to injury. In Reier, P.J., Bunge, R.P. and Seil, F.J. (Eds.), Current Issues in Neural Regeneration Research, Alan R. Liss, Inc., New York, 1988, pp. 247-255. 14 Engstrom, W. and Zetterberg, A., The relationship between purines, pyrimidines, nucleosides and glutamine for fibroblast cell proliferation, J. Cell Physiol., 120 (1984) 241. 15 Federoff, S. and Vernadakis, A., Astrocytes: (~'ll Biolo~,'y and Pathology o f Astrocytes, Vol. 3, Academic Press Inc., New York, 1986, pp. 129-472. 16 Fogh, J. and Trempe, G., Human Tumor Cells In Vitro, Plenum, New York, 1975, 115 pp. 17 Gallo, R.C., Whang-Pheng, D.J. and Perry, S., Isopentenyl adenosine stimulates and inhibits mitosis of human lymphocytes treated with phytohemagglutinin, Science, 165 (1969) 400-402. 18 Geiger, J.D. and Nagy, J.l., Adenosine deaminase and [~H]nitrobenzylthioinosine as markers of adenosine metabolism and transport in central purinergic systems. In: Williams, M. (Ed.), Adenosine and Adeno,~ine Receptors, Humana Press, N J, 199(I, pp. 225-288. 19 Giulian, D. and Baker, T.J., Characterization of ameboid microglia isolated from mammalian brain, J. Neurosci. 6 (1986) 2163-2178. 2(1 Giulian, D., Baker, T.J., Shih, I,.N. and Lacchman, L.B., Interleukin-I of the central nervous system is produced by ameboid microglia, J. Exp. Med., 164 (1986) 594-604. 21 Giulian, D., Allen, R.J., Baker, T.J. and Tomozawa, Y., Brain peptides and glial growth. 1. Gila-promoting factors as regulators of gliogenesis in the developing and injured central nervous system, J. Cell Biol., 102 (1986) 803-811. 22 Giulian, D., Chen, J., lngeman, J.B., George, J.K. and Noponen, M., The role of mononuclear phagocytes in wound healing after traumatic injury to adult mammalian brain, J. Neurosci., 9 (1989) 4416-4429. 23 Gospodarowicz, D., Bialecki, H. and Greenburg, G., Purification of the fibroblast growth factor activity from bovine brain, J. BioL Chem., 253 (1978) 3736-3743. 24 Hagberg, t1., Aderson, P., Lacarewicz, J., Jacobson, 1., Butcher, S. and Sandberg, M., Extracetlular adenosine, inosine, hypoxanthine, and xanthine in relation to tissue nucleotides and purines in rat striatum during transient ischemia, J. Neurochem., 49 (1987) 227-231. 25 tlaJl, R.H., The Modified Nucleosides in Nucleic Acids, Columbia University Press, New York, NY, 1971.55 Pp. 26 Hanson, G.R., lverson, P.L. and Partlow, L.M., Preparation and partial characterization of highly purified primary cultures of neuronal and non-neuronal (glial) cells from embryonic chick cerebral hemispheres and several other regions of the nervous system, Det:. Brain Res., 3 (1982) 529-545. 27 Hardebo, J.E., Kahrstrom, J. and Owman, C., PL and P2 purine receptors in brain circulation, Eur. J. Pharmacol., 144 (1987) 343-352. 28 Heppel, L.A., Wang, D., Huang, N., Gonzalez, F.A., Ahmed, A.H., Alfonzo, R.G. and Desheesh, M., Extracellular ATP is a mitogen for 3T3, 3T6, A431, DDTI-MF2, BALB/MK, NIE-115 and HFF cells. Ann. N Y Acad. Sci., 6(13 (1990) 432-434.

17 29 Hertz, L. and Richardson, J.S., Is neuropharmacology merely the pharmacology of neurons--or are astrocytes important too? Trends Pharmacol. Sci., 5 (1984) 272-276. 30 Hovi, T. and Vahern, A., Cyclic AMP and cyclic GMP enhance growth of chick embryo fibroblasts, Nature New Biol., 245 (1973) 175-176. 31 Hovi, T., Allison, A.C., Raivio, K.O. and Vaheri, A., Purine metabolism and control of cell proliferating. In: Elliott, K. and Fitzsimons, D.W. (Eds.), Purine and Pyrimidine Metabolism, Elsevier, Amsterdam, 1977, pp. 225-242. 32 Jorgensen, O.S., Brooksbank, B.W.L. and Balazs, R., Neuronal plasticity and astrocytic reaction in Down syndrome and Alzheimer disease, J. Neurol. Sei., 98 (1990) 63-79. 33 Kim, J-.K., Rathbone, M.P., Middlemiss, P.J., Hughes, D.W. and Smith, R.W., Purinergic stimulation of astroblast proliferation: guanosine and its nucleotides stimulate cell division in chick astroblasts, J. NeuroscL Res., 28 (1991) 442-455. 34 Kimelberg, H.K. and Norenberg, M.D., Astrocytes, Sci. Am., 260 (1989) 66-76. 35 Laloue, M., Terrine, C. and Guern, J., Cytokinins: Metabolism and activity of N6-(A2-isopententyl) adenosine and N6-(aZ-isopentenyl) adenine in tobacco cells and callus, Plant Physiol., 59 (1977) 478-483. 36 Lindsay, R.M., Reactive gliosis. In S. Federoff and A. Vernadakis (Eds.), Astrocytes: Cell Biology and Pathology cff'Astrocytes, Academic Press Inc., Orlando, 1986, pp. 231-262. 37 Londos, C., Cooper, D.M.F. and Wolff, J., Subclasses of external adenosine receptors, Proc. Natl. Acad. Sci. USA, 77 (1980) 2551-2554. 38 Meininger, C.J., Schelling, M.E. and Granger, H.J., Adenosine and hypoxia stimulate proliferation and migration of endothelial cells, Am. J. Physiol. (Heart Circ. Physiol.), 255 (1988) H554-H562. 39 Meininger, C.J. and Granger, J.H., Mechanisms leading to adenosine-stimulated proliferation of microvascular endothelial cells, Am. J. Physiol. (Heart Circ'. Physiol. 27), 258 (1990) H198-H206. 40 Nees, S., Herzog, V., Becker, B.F., Bock, M , DesRosiers, C. and Gerlach, E., The coronary endothelium: a highly active metabolic barrier for adenosine, Basic Res. Cardiol., 80 (1985) 515-529. 41 Nees, S., Coronary flow increases induced by adenosine and adenine nucleotides are mediated by the coronary cndothelium: a new principle of the regulation of coronary flow, Ent'r. Heart J., 10 (Suppl. F) (1989) 28-35. 42 Pearce, B., Murphy, S., Jeremy, J., Morrow, C. and Dandona, P., ATP-evoked Ca 2+ metabolism and prostanoid release from astrocytes: Pz-purinergic receptors linked to phosphoinositide hydrolysis, J. Neurochem., 52 (1989) 971-977. 43 Pfreundschuh, M., Shiku, H., Takahashi, T., Veda, R., Ransohoff, J., Oettgen, H.F. and Old, L.J., Serological analysis of cell surface antigens of malignant human brain tumors, Proc. Natl. Acad. Sci. USA, 75 (1978) 5122-5126. 44 Rathbone, M.P. and Hall, R.H., Metabolism and biological effects of N~'-(d2-isopentyl) adenosine on a cell line derived from circulating leukocytes of a patient with chronic myelongeous leukemia, Cancer Rel,., 32 (1972) 1647-1670. 45 Reier, P.J., Gliosis following CNS injury: The anatomy of astrocytic scars and their influences on axonal elongation. In S. Federoff and A. Vernadakis (Eds.), Astrocytes: Cell Biology and Pathology of Astrocytes, Vol. 3, Academic Press Inc., Orlando, 1986, pp. 262-324. 46 Schechter, R., Yen, S.H.C. and Terry, R.D., Fibrous astrocytes in senile dementia of the Alzheimer type, J. Neuroputhol. Exp. Neurol., 40 (1981) 95-101. 47 Schwabe, U., Ukena, D. and Lohse, M.J., Xanthine derivatives as antagonists at A~ and A 2 adenosine receptors, Naunyn Schrniedeberg's Arch. Pharmacol., 330 (1985) 212-22l. 48 Sensenbrenner, M., Dissociated brain cells in primary cultures. In S. Federoff and L. Hertz (Eds.), Cell, Tissue and Organ Culture in Neurobiology, Academic Press Inc., New York, 1977, pp. 191-213. 49 Szlapetis, G.K., Rathbone, M.P., Kim, J-.K., Del Maestro, R.F., Groves, J., Gilbert, J., Erola, K. and DeVilliers, R., Abstracts of the 4th Canadian Neuro-Oncology Meeting, Winnipeg, Manitoba, June 1-3, 1990. 50 Ukena, D., Jacobson, K.A., Padgett, W.L., Ayala, C., Shamim, M.T., Kirk, K.L., Olson, R.O. and Daly, J.W., Species differences in structure-activity relationships of adenosine and xanthine antagonists at brain A 1 adenosine receptors, FEBS Lett., 201 (1986) 122-128. 51 Ukena, D., Shamim, M.T., Padgett, W. and Daly, J.W., Analogs of caffeine: antagonists with selectivity for A 2 adenosine receptors, Life Sci., 39 (1986) 743-750. 52 van Calker, D., Muller, M. and Hamprecht, B., Adenosine inhibits the accumulation of cyclic AMP in cultured brain cells, Nature, 276 (1978) 830-841. 53 Williams, M., Purine receptors in mammalian tissues: pharmacology and functional significance, Annu. Ret'. Pharmacol. Toxicol., 27 (1987) 315-345.

Adenosine and its nucleotides stimulate proliferation of chick astrocytes and human astrocytoma cells.

Aqueous extracts of the brains of 18-day-old white Leghorn chicken embryos contain several substances that stimulate proliferation of primary cultures...
1007KB Sizes 0 Downloads 0 Views