Purines stimulate chick neuroblasts proliferation in culture

Neuroscience Letters, 41 (1983)325-330 Elsevier Scientific Publishers Ireland Ltd.

PURINK'~ STIMULATE CHICK NEUROBLASTS PROLIFERATION IN CULTURE

!. BARAKAT, M. SENSENBRENNER* and G. LABOURDETTE

Centre de Neurochimie du CNRS, 5, rue Blaise Pascal, 67084 Strasbourg Cedex (France) (Received August 19th, 1983; Accepted August 31st, 1983)

Key words: neuroblast culture - prohferation - purines

Neuroblasts from cerebral hemispheres of 6-day-old ~:hick embryo are able to divide to a certain extent under suitable culture conditions. It was found that addition of purine bases to the culture medium induced an increase of tritiated thymidJne incorporation into the cells, resulting from a stimulation of neuroblast proliferation. Most purines elicited a stimulation, but guanine compounds were the most active. inosinic acid (IMP), the first purine synthesized, was also active. Folic acid was inactive. These results suggest that neuroblasts in culture are defective in the biosynthesis of purines and that this deficiency is not due to a lack of folic acid. Some other cell types were also tested including glial cells, meningeal cells, whole embryo fibroblasts and heart fibroblasts. Only the latter did not respond to purine bases. These observations show that different cell types in primary culture need exogenous purines for maximal growth.

Proliferation of cells in primary culture can depel, d on or be stimulated by purines [4, 5, 7, 8, 12] or thymidine derivatives [8, 12, 13, 21], but this stimulation occurs only in the absence of folic acid [8, 12] or in the presence of an insufficient level of folic acid, as in medium 199 [7, 13, 21]. Dependence of growth on the uptake of exogenous nucleosides has also been shown in the presence of methotrexate which inhibits tetrahydrofolate synthesis [15, 22]. Thus, it seems that all cultured cells so far studied are able to synthesize nucleotides for a maximum proliferation rate when enough folic acid is present in the culture medium. However, it is known that in vivo some cells and tissues, including leukocytes [18], bone marrow [6], and gastrointestinal mucosa [14], are totally incapable of, or at least markedly defective in, de novo purine synthesis. Although :he brain contains the enzymes for de novo synthesis of the nucleotides and salvage pathways [3, 9, 14, 16, 17, 20], it has been reported that cortical de novo nucleotide synthesis in nervous tissue is quite slow compared with that found in other tissues [6]. This observation prompted us to investigate the effects of *Author for correspondence. 0304-3940/83/$ 03.00 © !9,-.~ Elsevier Scientific Publishers Ireland Ltd.

326

nucleotide bases on neural cells in culture, and particularly on neuroblast proliferation. Cultures enriched in neuroblasts can be obtained from brains of young chick embryos. These cells are able to divide in culture to some extent [19]. For the present study cells from cerebral hemispheres of 6-day-old chick embryos were used. We have shown previously that, under our culture conditions, cultures contain mainly neuronal cells for the first 3 days, and during this period proliferation of neurobJ.asts mainly occurred. Thereafter, proliferation of glioblasts begins [1]. In a first set of experiments cells were treated with 4 bases, cytosine, hypoxanthine, adenine and guanine, at various concentrations (Fig. 1). Treatment began at 24 h. Tritiated thymidine was added subsequently at 48 h and its incorporation into

16o

1so .E E

j.. Z

130

"6

~. 0

~'

°

110

11111 90 0

0125

Q25

(15

1

Concentration (raM)

Fig. 1. Effect of different bases on ['H]thymidine incorporation in chick neuronal cultures. Neuronal cuhures x~ere prepared as previously described [1] with some modifications. Briefly, dissociated cells from cerebral hemispheres of 6-day-old chick embryos were cultured on a layer of 1% collagen. The autrient culture medium consisted of Eagle's MEM (Gibco) supplemented with 50/0 fetal calf serum (FCS) (Gibco). Cells were cultured in Falcon plastic Petri dishes (60 mm diameter). Culture medium was changed after 24 h. The new culture medium contained the various bases at different concentrations. After 24 h incubation in the presence of the different bases (48 h in culture) [methyl-~H]thymidine was added at a final concentration of 0.5 ~Ci/ml. 24 h later (72 h in culture), medium was removed and dishes were washed twice with phosphate-buffered saline (0.14 M NaCl, 70 mM potassium phosphate, pH 7.2). Cells were scraped off into 4 ml of 0. ! % EDTA in the same buffer and the suspension was filtered through borosilicate filter discs (Whatman GF/B). Discs were washed 4 times with 10 ml of cold 5% trichloroacetic acid and dried at 80°C for 30 min. Radioactivity was determined in a Beckman LS 9000 scintillation counter using Rotiszin II (Roth) scintillation mixture. Results are expressed as percent relative to the untreated cultures. They are means of 7 different dishes from two independent experiments. Bars represent S.E.M., , , guanine; u, hypoxanthine; A, adenine; o, cytosine (from top to bottom).

327

DNA was determined at 72 h. Such a pretreatment before determination of the thymidine incorporation resulted in higher stimulation. An increase of thymidine incorporation was observed only ia those cultures treated with the purine bases. The guanine base elicited the strongest effect. In the cultures of chick embryo neuroblasts some glioblasts are present. To deter, mine if the increase of thymidine incorporation was due only to a proliferation of neuronal cells two different approaches we're undertaken. The first combined autoradiography [11] and identification of neuronal cells by staining with toluidine blue. The principle of the experiment was to label the culture with ['H]thymidine at the beginning of the culture when the effects of bases were observed. Then excess [3H]thymidine was washed off and cultures were grown up to 10 days after seeding. At that time the labeled cells could be identified by their morphology on the autoradiographic preparations. For these experiments cells were grown in the presence of guanine (0.2 mM) and were labeled for 24 h periods from time 0 until day 4 after seeding. On day 5 all media were changed for a medium supplemented with 20°7o FCS instead of 5°70, because in the presence of 20o70 FCS cells are more dispersed and can be easily identified. In control and guanine-treated cultures, labeling was found in neuroblasts; almost no glial cells were labeled in 10-day-old cultures, after addition of radioactive thymidine up to day 3 after seeding. When [3H]thymidine was administered from day 3 to 5, more and more glial cells were found to be labeled after 10 days in both conditions. These observations show that the stimulation of thymidine incorporation by guanine is due essentially to neuroblast proliferation. In the second approach cells were treated with guanine and labeled with thymidine as described above. Then they were rinsed several times and treated with an antiserum against the neuronal D2 protein [2] with complement. After 1 h incubation at 37°C most neuronal cells underwent lysis. Cultures were then incubated with DNAase for 15 min at 37°C. Non-neuronal cells which remained were scraped off. Their radioactivity in control as well as in guanine treated cultures was only about 3if/0 of that of the whole culture. Since the stimulation of thymidine incorporation by guanine is at least 50°7o, non-neuronal cells cannot be responsible fo~ such an increase. These two sets of data show that guanine stimulates the proliferation of neuroblasts in culture. The fact that guanine elicits a higher stimulation than adenine suggests that guanine is a limiting metabolite for neuroblast proliferation. To investigate further the limiting step the cells were treated with different purine nucleotides and nucleosides (Fig. 2). All guanine derivatives tested gave a high stimulation. Adenine derivatives, hypoxanthine, inosine and inosinic acid were less active. Xanthine, xanthosine and XMP were inhibitory. The activity of adenine nucleotides was lower compared with that of the guanine nucleotides and this would be consistent with their possible partial conversion to guanine nucleotides.

328

Adenine 149

I--lypo~Of~hlne _ _ , . 141

•.1~

!i

d,e~Lmogr~ 111

~'~lenos~ne t 54

•. 5

*_15

,,

,,

d At,A P 80 ~!

At,,IP 1253 iO

dACcF'

It

.........

ADP

*-5

__~..

IT

it

Guon~ne 11:32 ~2

iT

Inos~ne 136

IMP 166 ,l~J

~

--5

IT

Xant~osune 44

_,17

_~. - ; ~

Xanthlne 93

Gu~osme 203 t4

t5

-~, _~..i_~.

II

XM P 86 i4

~,_---~

dGuo, n o s m e

li

il

CoMP 188 ~17

GL-)P I¢6

dGMP 188 t2

__

..

dODP

t

1'

I*

d AT ,:'

Ar ~

',.:] P

d C t ~)

t !

,16

,,12

t t5

Fig. 2. Effect of various bases and derivatives of the purine nucleotides triphosphate pathways on proliferation of chick neuroblasts in culture. Major pathways of purine metabolism in mammals are shown. Cultures, treatments and incorporation were as described for Fig. 1. All compounds were added at a final concentration of 0.2 raM. Results show the percenl of thymidine incorporation compared with untreated cultures (100%). They are tile means of at least 3 different dishes, up to 10, with _+ S.D. Large arrows indicate metabolic entry and exitpoints.

Interestingly, dGTP gives a high stimulation and dATP gives an inhibition. Since deoxyribonucleotides do not convert easily or at all to ribonucleotides, this result would suggest that dGTP acts by direct incorporation into DNA. The dATP would be inhibitory because the level reached, when added to the dATP already present in the cells, would be high enough to inhibit the synthesis of all other nucleotides. Since inosinic acid, the first purine synthesized, is active, the limitation might occur in its synthetic pathway. Folic acid is involved in two steps of this pathway; other cofactors involved are glutamine, glycine, aspartic acid and Mg z + ions. When these compounds were tested on the cells (Table I), none were found to affect neuroblast proliferation under our conditions. Thus deficiency in purine biosynthesis does not seem to be due to a defect of cofactor(s) in the synthetic pathway of inosinic acid. The insufficient purine synthesis in cultured neuroblasts could be due to the fact that in vivo purines are also provided by the blood. The liver is known to synthesize purines which are transferred into the blood as metabolites for other organs. If this is true, other cell types may need purines for maximal growth in culture. Indeed, it was found that guanine and adenine stimulated the proliferation of other brain cell types, like glial and meningeal cells (Table II). Folic acid elicited some effect, but not as high as purines. In fibroblasts from 8-day-old whole chick embryos purines still stimulated thymidine incorporation, but folic acid was inactive. On

329 TABLE ! E F F E C T O F C O M P O U N D S INVOLVED IN T H E SYNTHETIC P A T H W A Y O F INOSINIC A C I D (IMP) ON NEUROBLAST P R O L I F E R A T I O N IN C U L T U R E Details as for Fig. 1. Results are the m e a n o f at least 3 different dishes with ± ~.D. Additives

[3H]Thymidine incorporation, % o f control

Final concentration

Folic acid Folinic acid Glutamine Glutamic acid Aspartic acid Glycine Mg 2 +

100 93 90 82 85 106 98

0.2 0.3 0.2 0.2 0.2 0.3 0.2

+ _+ + + + ± +

6 8 4 0.5 I 15 9

mM mM mM mM mM mM mM

heart fibroblasts from 15-day-old chick embryos purines and folic acid exerted no effects. These results indicate that neuroblasts are not the only cell type which needs exogenous purines for maximal proliferation in culture; other cerebral or noncerebral cells have the same requirement. On the other hand, heart fibroblasts do not show this requirement. This result is consistent with the observation that heart has the capacity to recover functionally after lack of oxygen while brain does not [6, 10]. In these papers the authors concluded that it seemed very likely that organs capable of post-anoxic metabolic and functional restoration were characterized by their ability to enhance nucieotide synthesis de novo after lack of oxygen. Most classical culture media, such as Eagle's and Ham's media, RPMI medium and medium 199, are not able to sustain maximal growth of various cells in primary culture. The results reported here provide evidence that folic acid and purine bases must be added to satisfy the nutritional requirements of various cell types. The question of purine biosynthesis in cultured cells was thought to be solved by the addition of folic acid. It is not, and has to be re-examined, at least for primary cultures. TABLE ll EFFECT OF PURINES A N D FOLIC ACID ON PROLIFERATION OF VARIOUS CELL TYPES IN PRIMARY CULTURE Glial cells were derived from brain hemispheres of 15-day-old chick-embryos. Meningeal cells were prepared from cerebral hemispheres of 8-day-old' chick embryos. Fibroblasts were derived from whole 8-day-old chick embryos after removal of brain and spinal cord. Heart fibroblasts were isolated from 15-day-old chick embryos. All cultures were grown in M E M supplemented with 20% FCS. They were treated on day 4, since the response was maximal at that time. [~H]Thymidine incorporation was started on day 5 and lasted, for 5 h. Results are expressed as percent of incorporation in control untreated cultures. They are means of at least 5 dishes, from two independent experiments with +_ S.D. Additives

Glial cells

Meningeal cells

Fibroblasts

Heart fibroblasts

Guanine Adenine Folicacid

144 _+ 1 150 + 13 129 + 5

188 + 15 149 + 10 132 + 9

151 + 18 153 + 17 85 + 2

104 + 9 95 =t: 5 93 + 5

We thank Mme Marie-France Knoetgen for technical assistance. We are grateful to Dr. E. Bock for the chicken anti-D2 serum. This work was supported by grants from the Institut National de la Sant6 et de la Recherche M6dicale (81.79.113) and from D.G.R.S.T. (80.7.0349). G.L. is Charg6 de Recherche at the INSERM; M.S. is Maitre 6e Recherche at the CNRS. I Barakat, I. and Sensenbrenner, M., Brain extracts that promote the proliferation of neuroblasts from chick embryo in culture, Develop. Brain Res., I (1981) 355-368. 2 Bock, E., Nervous system specific proteins, J. Neurochem., 30 (1978) 7-14. 3 Bourget, P.A. and Tremblay, G.C., Pyrimidine biosynthesis in rat brain, J. Neurochem., 19 (1972) 1617-1624. 4 Clarke, G.D. and Smith, C., The response of normal and polyoma virus-transformed BHK/21 cells to exogenous purines, J. Cell Physiol., 81 (1973) 125-132. 5 Colby, C. and Edlin, G., Nucleotide pool levels in growing, inhibited and transformed chick fibroblasl cells, Biochemistry, 9 (1970) 9 ! 7-920. 6 Gerlach, E., Marko, P., Zimmer, H.-G., Pechan, I. and Trendelenburg, Ch., Different response of adenine nucleotide synthesis de novo in kidney and brain during aerobic recovery from anoxia and ischemia, Experientia, 27 (1971) 876-878. 7 Gregory, S. and Kern, M., Adenosine and adenine nucleotides are mitogenic for mouse thymocytes, Biochem. biophys. Res. Commun., 83 (1978) I 1i 1-1116. 8 Hakala, M.T. and Taylor, E., The ability of purine and thymidine derivatives and of glycine to support the growth of mammalian cells in culture, J. biol. Chem., 234 (1958) 126-128. 9 Held, !. and Wells, W., Observation on purine metabolism :n rat brain, J. Neurochem., 16 (1969) 529-536. 10 Kleihues, P., Kobayashi, K. and Hossmann, K.-A., Purine nucleotide metabolism in the cat brain after one hour of complete ischemia, J. Neurochem., 23 (1974) 417-425. I I Korr, H., Light microscopical autoradiography of nervous tissue. In Ch. Heym and W.G. Forssmann (Eds.), Techniques in Neuroanatomical Research, Springer-Verlag, Berlin, 1981, pp. 218-244. 12 Lieberman, !. and Ore, P., Control of growth of mammalian cells in culture with folic acid, thymidine, and purines, J. biol. Chem., 235 (1960) II 19-1123. 13 McAuslan, B.R., Reilly, W. and Hannan, G.N., Stimulation of endothelial cell proliferation by precursors of thymidylate, J. Cell Physiol., 100 (1979) 87-94. 14 McKinnon, A.M. and Deller, D.J., Purine nucleotide synthesis in gastrointestinal mucosa, Biochim. biophys. Acta, 319 (1973) I-4. 15 Marz, R., Wohlhueter, R.M. and Piagemann, P.G.W., Growth rate of cultured Novikoff rat hepatoma cells as a function of the rate of thymidine and hypoxanthine transport, J. Membrane Biol., 34 (1977) 277-288. 16 Phillips, E. and Newsholme, E.A., Maximum activities, properties and distribution of 5'-nucleotidase, adenosine kinase and adenosine deaminase in rat and human brain, J. Neurochem., 33 (1979) 553-558. 17 Santos, J.N., Hempstead, K.W., Kopp, L.E. and Miech, R.P., Nucleotide metabolism in rat brain, J. Neurochem., 15 (1968) 367-376. 18 Scott, J.k., Human leukocyte metabolism in vitro. I. Incorporation of adenine-8-t4C and formateC ~4 into the nucleic acids of leukemic ieukocytes, J. clin. Invest., 41 (1962) 67-79. 19 Sensenbrenner, M., Wittendorp, E., Barakat, i. and Rechenmann, R.V., Autoradiographic study of proliferating brain cells in culture, Develop. Biol., 751 (1980) 268-277. 20 Shaw, G. and Thomas, S.E., Purine nucleotide de novo biosynthesis in the brain, J. Neurochem., 27 (1976) 637-639,

21 Taylor-Papadimitriou and Rozengurt, E., The role of thymidine uptake in the control of cell proliferation, Exp. Cell Res., 119 (1979) 393-396. 22 Warnick, C.T., Muzik, H. and Paterson, A.R.P., Interference with nucleoside transport in mouse lynaphoma cells proliferating in culture, Cancer Res., 32 (1972) 2017-2022.