In vitro stimulation of glia cells by a lymphocyteproduced factor

Journal o['the Neurological Sciences, 1980, 46:55-62

55

© Elsevier/North-Holland Biomedical Press

IN VITRO STIMULATION OF G L I A CELLS BY A LYMPHOCYTEPRODUCED FACTOR

A. F O N T A N A j, A. G R I E D E R 2, ST. A R R E N B R E C H T I and P. G R O B l

1Department q[" Internal Medicine, Section of Clinical Immunology, University Hospital, Zurich, and 2pharmaceutical Division, Preclinical Research, Sandoz L TD, Basle, (Switzerland) (Received 3 September, 1979) (Accepted 3 December, 1979)

SUMMARY

The factors responsible for the activation of astrocytes surrounding inflammatory brain tissues are unknown. The present study was designed to examine the ability of lymphocytes to produce astrocyte stimulating activity in cell culture. Normal rat lymphocytes stimulated with Concanavalin A, or sensitized lymphocytes, challenged with antigen in vitro, activate cultured rat glia cells by a soluble mediator which we have termed glia stimulating factor (GSF). In undifferentiated glioblasts both RNA synthesis, as measured by [5-3H]uridine uptake, and DNA synthesis, as measured by [6-3H]thymidine uptake, were stimulated by the presence of GSF. Preliminary characterisation showed the GSF to be non-dialysable and heat stable at 56 °C for 30 min, but not stable at 80 °C for 30 min. To study the effect of this factor on differentiated glia cells, brain cell cultures were treated with dibutyryl cyclic AMP (db-cAMP) which induces a morphologic transformation of glioblasts to multipolar cells that have a characteristic astrocytic appearance. After addition of GSF to db-cAMP treated astrocytes only an increase in RNA synthesis was observed. The significance of this in vitro phenomenon, mediated by a glia stimulating factor, to activation of astrocytes and astrocytic gliosis in human brain diseases is discussed.

INTRODUCTION

An early reaction of the central nervous system (CNS) to inflammatory processes is the perivenular accumulation of lymphocytes or "perivascular cuffing". Surrounding this inflammatory cuff there is an astrocytic reaction consisting of enlargment ofastrocytes with formation of large fibril-filled processes. The mechanisms

56 underlying this glial response are poorly understood. Local oedema and/or neuronal destruction are thought to induce astrocyte activation. Lymphocytes and monocytes play an important role in the tissue response to injury. In addition, when lymphocytes are stimulated by nonspecific mitogens or specific antigens in vitro, they produce a variety of soluble effector substances or lymphokines. Lymphokines play a prominent role in the development of inflammatory response and include factors which damage target cells (lymphotoxins), which cause cell proliferation (mitogenic factors) and which modulate the inflammatory response by mobilization and/or activation oflymphocytes, macrophages and granulocytes (for review see Cohen 1977). We have investigated whether lymphocytes might also serve as an inducer of astrocytic respofise to inflammation in the CNS. We report here that activated lymphocytes elaborate soluble substances in vitro which stimulate cultured glia cells and which we have given the preliminary term glial stimulating factor (GSF). MATERIAL A N D METHODS

Brain cell cultures Whole brains from 19-day rat fetuses (Lewis) were dissociated according to the procedure of Shapiro and Schier (1973). After removal of the meninges the brain tissue of 10-15 fetuses was disaggregated in solution D1GS (137 mM NaC1, 5.4 mM KC1, 0.17 mM Na2HPO 4, 5.5 mM glucose, 0.22 mM KHzPO 4, 59 mM sucrose, pH 7.4, 340 mOsm) by passage through nylon mesh of 210 #m and 132/~m pore diameter (Nitex, Schweizerische Seidengazefabrik AG, Thal, Switzerland). The dissociated cells were seeded in 60-mm plastic petri dishes (Falcon Plastics, Los Angeles, CA) in Dulbecco's modified Eagle's Medium (DMEM-H21, Gibco, Grand Island, NY) containing 20~0 fetal calf serum, 10 units/ml Na-penicillin G and 10 #g/ml streptomycin sulfate. The inoculation volume was 4 ml per petri dish containing 60 mg suspended cells. The cultures were grown at 37 °C in a humidified 5'J/o C O 2 / 9 5 ~ oo air atmosphere. Growth media were changed on the 3rd and the 6th day with DMEM supplemented with 5~o horse serum and 5~o fetal calf serum and on the 10th day of culture with DMEM containing 5~o horse serum. On the 22nd day of culture, lymphocyte supernatants as described below were added in various proportions. In some experiments, N 6, O2-dibutyryl adenosin 3': 5'-cyclic monophosphoric acid (db-cAMP, Sigma, St. Louis, MO) (final concentration 1 mM) was added on thb 19th day of culture in order to induce glia cell differentiation. These cultures also received lymphocyte supernatants on the 22nd day of culture. Lymphocyte culture supernatants Spleens from 12-month-old Lewis rats were first minced with scissors in cold Hank's balanced salt solution (HBSS) and fragments then squashed with a loose fitting teflon pestle against the walls of a glass tube. Tissue debris was left to settle and the supernatant cell suspension layered on Ficoll-Hypaque, d = 1.077 (Phar-

57 macia). After centrifugation at 900 x g for 30 min at room temperature, the lymphocyte layer was harvested and washed 3 times in HBSS. 25 × 106 lymphocytes were cultured in 15 ml of supplemented minimal Eagle's medium containing 2mercaptoethanol (3 x 10-5) and 0.5~o normal rat plasma, in Falcon plastic tissue culture flasks (75 cm 2 growth area) kept upright. For stimulation of lymphocytes, 5/~g/ml Concanavalin A (Con A, Miles Yeda) was added. After 24 hr, cultures were resuspended, centrifuged at 300 x g and supernatants (SN) harvested. SNs from ConA containing cultures are designated as active SNs (AS), those from cultures not containing ConA as control SN. These control SNs (CS), however, were supplemented with ConA to a final concentration of 5/~g/ml before assay on brain cell cultures. SNs were kept frozen at -20 °C until assayed. Furthermore, ct-methyl-Dmannoside (20 mg/ml) (Sigma, St. Louis, MO) was added in two experiments to both the AS and CS. AS and CS were also obtained by immunisation of Lewis rats with 2 mg of bovine 7-globulin (BGG) in Freund's complete adjuvant (CFA). Lymphocytes obtained from regional lymph nodes 7 days after immunisation were suspended in medium and cultured for 24 hr with or without the presence of 100 #g/ml BGG. Culture SNs were collected by centrifugation at 300 × g for 30 min. CS obtained from cultures in the absence of antigen were reconstituted with BGG after the harvest of such cell SNs (Yoshida et al. 1973).

Glia cell activating assay On the 22nd day of culture 200/~1 aliquots of AS or CS were added to the brain cells. After 72 hr 0.5 uCi/ml [5-3H]uridine (spec. act. 5 Ci/mM) and 0.25 uCi/ml [6-3H]thymidine (spec. act. 5 Ci/mM, Radiochemical Centre, Amersham, Great Britain) were added. Incorporation of the labeled nucleotides into RNA and DNA over 1-24 hr was used as a measure of glia cell activation. Extraction of the nucleic acids and scintillation counting was carried out, as previously described (Grieder et al. 1977). Briefly, at different time intervals after the incubation with AS or CS and labeled precursors the medium was removed and cells were washed thoroughly with cold saline followed by 3 washes with ice-cold 5 ~ trichloroacetic acid (TCA) in order to remove non-incorporated, acid-soluble label. The remaining TCAinsoluble components were dissolved overnight in 0.2 N NaOH at room temperature. Aliquots of this alkaline hydrolysate were used for protein determination according to the method of Lowry using bovine serum albumin as a standard. The remainder of the extract was neutralized with 1/10 volume of 2 N HCI at 4 °C and the DNA and proteins were precipitated with 2 volumes of ice-cold I 0 ~ TCA. After centrifugation, the radioactivity in the supernatant, corresponding to the incorporated [5-3H]uridine into RNA, was determined in a Packard Tri-Carb Model 332o liquid scintillation counter by measuring 1-ml aliquots dissolved in 10 ml Lumagel. The precipitate was lysed overnight in 0.5 ml 0.2 N NaOH, mixed with I0 ml Lumagel and used for determination of radioactivity incorporated by [6-3H]thymidine into DNA. A stimulation ratio (SR) was calculated by dividing the mean value of cpm per mg protein incorporated in cultures containing AS with the mean value of cpm

58 per mg protein of CS. In all experimental groups the range of the cpm observed in quadruplicate samples was within 10~,,, of the mean value.

Immunofluorescence with anti-GFA antisera For demonstration of the astrocyte-specific gliai fibrillary acid protein (GFA), cultures were established on glass cover slips. After 22 days in culture, cover slips were fixed in 3.7~o formaldehyde in phospate buffered saline (PBS) for 15 rain and in acetone at 4°C for 6 min. For immunofluorescence a sandwich method was employed. Fixed cells were overlayered with a 1 : 40 dilution of a rabbit antiserum against bovine G F A prepared according to Dahl and Bignami (1976) for 30 min. After two rinses with PBS, they were incubated for another 30 min with a fluoresceinated goat-anti-rabbit-immunoglobulin (Microbiological Associates, MD), also diluted ! : 40. After another two rinses in PBS, they were mounted in Permount.

RESULTS

With the culture conditions described, the cells grew to confluence and maintained an epithelial-like morphology. In the presence of db-cAMP many of the flat polygonal cells devoid of cell processes were transformed into multipolar cells resembling mature astrocytes. In these cultures most of the cells possessed intensely fluorescent processes or a finely reticulated perinuclear immunofluorescence after staining with anti-GFA antisera, whereas only a delicate glial framework was visible in cultures not treated with db-cAMP.

Effect of GSF on DNA and RNA synthesis Lymphocytes SNs were added to glia cell cultures on the 22nd day. The effect of glia stimulating factor (GSF) released by lymphocytes pulsed with ConA for 24 hr was evaluated during different labeling intervals of the labeled nucleotides. The highest SR were noted 3 hr after addition of the radioactive precursors, longer labeling periods were less effective. After 24 hr no difference was detectable in cultures being supplemented either with AS or CS (Fig. 1). Consequently, a 3-hr labeling period was used for subsequent experiments. Various concentrations of lymphocyte SN were assayed as above. A representative experiment is shown in Fig. 2. SNs of ConA activated lymphocytes stimulate glia cells to incorporate [6-3H]thymidine and [5-3H]uridine well above control levels. Maximum stimulation occurred following addition of 10'~"0 AS to the glia cell culture medium. Less stimulation was seen at higher or lower concentrations. CS had no stimulatory capacity as compared to cultures not receiving any SNs. The following tests were performed with a 10°/oconcentration of SN, and a 3-hr labeling period. Addition of 7-methyl-D-mannoside to CS did not alter the incorporation of the nucleotides into cultured glia cells. Also, it did not abrogate the glia cell stimulating activity of AS (Table 1). As also shown in Table 1~ the active material was retained by the dialysis membrane upon dialysis against distilled water for 40 hr.

59

•//"

Fig. 2

Fig. 1 2

SR 9 SR

8

1

8

7

7

6

6 5 5 4

3 2 1

°~°-o-~~_____~ I

I

I

I

1

2

3

4

TIME OF INCUBATION

4

3 2 1

I

24 h

O.1 0.5

{HOURS)

RELATIVE

1

5

10 9 0

CONCENTRATION

LYMPHOCYTE

30 OF

SUPERNATANT$

(%)

Effect of glia stimulating factor on RNA and DNA synthesis of cultured gila cells. Fig. 1. Stimulation ratio (SR) vs length of incubation time with [6-3H]thymidine (O) and [5JH]uridinc (O).

Fig. 2. Stimulation ratio (SR) vs GSF concentration in cultures labeled 3 hr with [6-3H]thymidine (O) and [5-3H]uridine (0).

TABLE 1 EFFECT OF GSF AFTER HEATING, DIALYSIS AND TREATMENT WITH -METHYL-D-MANNOSIDE

[6-3H]Thymidine

[5-3H]Uridine

Nontreated

56 °C for 30 min

80 °C for 30 min

After dialysis

a-mM c

AS a CS

56321 20850

60492 20850

18765 20850

30848 10763

35945 11438

SR b

2.7

2.9

0.9

2.8

3.1

AS a CS

28003 5283

29961 5283

3457 5283

26041 5078

23696 5198

SR b

5.3

5.6

0.65

5.1

4.5

a Geometric mean cpm of [6-3H]thymidine or [5-3H]uridine incorporation "of quadruplicate cultures. Standard errors are not indicated since they were less than 10% of the mean values. b Stimulation ratio calculated from (cpm/mg protein of As : cpm/mg protein of CS). c ~.mM = ~-methyl-o-mannoside (20 mg/ml). T h e a c t i v i t y c o u l d n o t b e d e s t r o y e d b y h e a t i n g t o 56 °C f o r 30 m i n . H o w e v e r , u p o n h e a t i n g t o 80 °C f o r 30 m i n a c t i v i t y w a s c o m p l e t e l y a b o l i s h e d . In order to study the effect of antigen-activated

lymphocytes

as opposed

to

lymphocytes stimulated with ConA, SNs of lymphocytes from animals immunized

60 with B G G a n d challenged in vitro with this antigen were a s s a y e d for G S F activity. These SNs gave identical results as C o n A i n d u c e d A S ( d a t a not shown).

GSF-activity in db-cA MP treated glia cell cultures .In o r d e r to test the G S F on differentiated astrocytes, d b - c A M P was a d d e d to the glia cell cultures. The effect o f C o n A - s t i m u l a t e d l y m p h o c y t e SN on d b - c A M P t r e a t e d glia cells was different c o m p a r e d to that o b s e r v e d in u n t r e a t e d glia cells. W h i l e the [5-3H]uridine u p t a k e o f the glia cells was similar, no a u g m e n t a t i o n o f [6-3H]thymidine i n c o r p o r a t i o n was o b s e r v e d with the A S (Table 2). A s was the case for the C o n A - a c t i v a t e d l y m p h o c y t e SNs, the a n t i g e n - i n d u c e d A S did n o t increase the [6-3H]thymidine u p t a k e c o m p a r e d to control. H o w e v e r , they h a d a significant s t i m u l a t o r y activity on R N A synthesis (Table 2). DISCUSSION The o b s e r v a t i o n s m a d e in this study m a y p r o v i d e insight into the p a t h o p h y s i o logic m e c h a n i s m s l e a d i n g to a s t r o c y t e a c t i v a t i o n d u r i n g i n f l a m m a t o r y processes o f the brain. Both C o n A - a c t i v a t e d l y m p h o c y t e s a n d a n t i g e n - s t i m u l a t e d l y m p h o c y tes release a factor o r factors that stimulates rat glioblasts in culture. N o glia stim u l a t i n g factor ( G S F ) was p r o d u c e d by u n s t i m u l a t e d l y m p h o c y t e s or C o n A itself. F u r t h e r m o r e , i n a c t i v a t i o n o f C o n A by ~ - m e t h y l - D - m a n n o s i d e did not reverse the glia s t i m u l a t i n g activity o f AS, which m a k e s a direct C o n A - m e d i a t e d effect on c u l t u r e d glia cells unlikely. T h e G S F is n o n d i a l y z a b l e , a n d h e a t - s t a b l e at 56 °C for 30 min, but not at 80 °C

TABLE 2 EFFECT OF GSF ON db-cAMP-TREATED GLIA CELLS Not treated

[6-3H]Thymidine

Supernatant

mean cpm per mg proteina

ConA

61775 17153

3.6

983 842

1.1

BGG - ASc CS

42350 15685

2.7

1872 1586

1.1

ConA -

78320 12632

6.2

49399 9649

5.1

54430 10269

5.3

37113 7731

4.8

- AS b

CS

[5-3H]Uridine

db-cAMP-treated

AS b

CS BGG -AS ~ CS

SR a

cpm per mg proteina mean

a Calculated as in Table 1. b SNs of lymphocytes activated in vitro with ConA. c SNs of sensitized lymphocytes challenged in vitro with bovine ),-globulin.

SR a

61 for 30 min. These properties are similar to those described for other lymphokines (Rocklin et al. 1972; Yoshida et al. 1976; Larsson et al. 1979). Many of the flat epitheloid cells we have cultured can be interpreted as being astrocyte precursors since direct immunofluorescence performed with an antiserum against the glial fibrillary acid protein (GFA), a marker for astroglia (Bock et al. 1975; Bignami and Dahl 1977), revealed that the cells form a delicate glial framework. A similar staining has been described in cultures of fetal mouse cerebral hemispheres by Kozak et al. in 1978. In addition, db-cAMP induced marked morphological alterations, converting the undifferentiated glia cells towards a more differentiated state. This observation is similar to those of Lim et al. (1973) and Moonen and Sensenbrenner (1975). The differentiated cells resemble mature astrocytes with numerous processes. The investigation with the anti-GFA antisera made the additional important observation that the cells in rib-cAMP-treated cultures showed intensely perinuclear immunofluorescence and long fluorescent cell processes. Together with a growing list of biochemical (Kimelberg et al. 1978), immunohistological (Bock et al. 1977; Kozak et al. 1978) and electron-microscopic studies (Haugen and Laerum 1978), these experiments support the concept that primary cultures derived from 19-day-old rat embryos are composed predominantly of astroglial cells and their precursors. These cells are likely to be the target of the GSF elaborated by activated lymphocytes. Glia stimulating factor exerts a blastogenic response in astrocyte precursor cells as evidenced by an increase in [6-3H]thymidine uptake. In more differentiated astrocytes, produced by the addition of db-cAMP to the glia cell cultures, the GSF stimulates only the RNA synthesis, but not further DNA synthesis. The question of whether mature astrocytes are able to resume replicative activities in response to various stimuli is still unsettled. Our observation supports the most widely held view that replication of mature astrocytes is not common (Varon and Somjen 1979). GSF could be generated in vivo, whenever specific immunologic mechanisms take place in the brain, triggered by infections or autoimmune reactions. Candidates for such disease states, in which GSF might play a role, are e.g. viral encephalitis, multiple sclerosis and focal epilepsy, all disorders, for which in recent years either immune mechanisms or astrocyte activation or both phenomena have been observed (Ettlinger and Lowrie 1976; Brotchi et al. 1978; Hirsch and Parks 1979). It remains to be elucidated which functional properties of astrocytes could be altered by GSF. 'The only biologic factor so far found to influence cultured astrocytes was detected by Lim et al. in 1972. They observed that glioblasts grown in a monolayer culture can be stimulated by a factor to differentiate into multipolar interconnected astrocyte-like cells (Lim et al. 1972, 1973). Whether there are any connections between GSF and the factor described by Lim et al. remains to be studied.

62 ACKNOWLEDGEMENT T h e a u t h o r s t h a n k R o g e r Vuille f o r e x p e r t t e c h n i c a l a s s i s t a n c e .

REFERENCES Allen, J.V., G. Glover, St. R. McKeown and D. McCormick (1979) The cellular origin of lysosomal enzymes in the plaque in multiple sclerosis, Neuropath. appl. Neurobiol., 5:197 210. Bignami, A. and D. Dahl (19771 Specificity of glial fibrillary acid protein for astroglia, J. Histochem+ Cytochem., 25: 466~,69. Bock+ E., O.S. Jorgensen, L. Dittman and L.F. Eng (19751 Determination of brain-specific antigens in short-term cultivated rat astroglial cells and in rat synaptosomes, J. Neurochem., 25 : 867 874. Bock, E., M. Moller, C. Nissen and M. Sensenbrenner (1977) Glial fibrillary acidic protein in primary astroglial cell cultures derived from newborn rat brain+ FEBS Lett.. 83:207 211. Brotchi, J., A. Dresse, J. Scuv6e-Moreau and M.A. Gerebtzoff (1978) Histochemical study of cobalt induced focal epilepsy, Exp. Brain Res., 32: 459469. Cohen, St. (1977) The role of cell-mediated immunity in the induction of inflammatory responses, Amer. J. Path.+ 88: 502-528. Dahl, D. and A. Bignami (1976) Immunogenic properties of the glial fibrillary acid protein. Brain Res., 116: 150-157. Ettlinger, G. and M.B. Lowrie (1976) An immunological factor in epilepsy+ Lancet, 1 : 1386. Grieder, A., R. Maurer and H. Staehelin (1977) Comparative study of early effects of epipodophyllotoxin derivatives and other cytostatic agents on mastocytoma cultures, Cancer Res., 37 : 2998 3005. Haugen, A. and O.D. Laerum (1978) Induced glial differentiation of fetal rat brain cells in culture An ultrastructural study, Brain Res., 150:225-238. Hirsch, H.E. and M.E. Parks (1979) A thiol proteinase highly elevated in and around the plaques of multiple sclerosis, J. Neurochem., 32 : 505 513. Kimelberg, H.K., S. Narumi and R.S. Bourke (1978) Enzymatic and morphological properties of primary rat brain astrocyte culture and enzyme development in vivo+ Brain Res., 153 : 55 77. Kozak, L.P., D. Dahl and A. Bignami (19781 Glial fibrillary acidic protein in reaggregating and monolayer cultures of fetal mouse cerebral hemispheres, Brain Res., 150:631 637. Larsson, E.L., Mannervik B. and H. Blomgren (1979) Characterisation of a human mitogenic factor released by phytohemagglutinin-activated lymphocytes, Cell Immunol.+ 44: 424-432. Lim, R., W.K.P. Li and K. Mitsunobu (19721 Maturation-stimulating effect of brain extracts on dissociated embryonic brain cells in culture, Abstr. of 2nd Ann. Meeting Soc. Neurosci., Houston, TX, p. 181. Lim, R., K. Mitsunobu and W.K.P. Li (1973) Maturation-stimulating effect of brain extracts and dibutyryl cyclic AMP on dissociated embryonic brain cells in cultures, Exp. Cell. Res., 79: 243-246. Moonen, G. and M. Sensenbrenner (19751 Effect of dibutyryl cyclic AMP on cultured brain cells from chick embryos of different ages, Experientia (Basel), 32: 40-42. Rocklin, R.E., H.G. Remold and J.R. David (1972) Characterisation of human lnigration inhibitory factor from antigen-stimulated lymphocytes. Cell. lmmunol., 5: 436-445. Shapiro, D.L. and B.K. Schrier (19731 Cell cultures of fetal rat brain, Exp. Cell. Res., 77: 239-247. Varon, S.S. and G.G. Somjen (1979) Glial typology and gliogenesis, Neurosci. Re~'. Pro.~r. Bull. : NeuronGlia Interaction, 17:19 41. Yoshida, T., H. Sonozaki and St. Cohen (1973) The production of migration inhibitory factor by B and T cells of the guinea pig, J. exp. Med., 138 : 784-793. Yoshida, T., T. Kuratsuji, A. Takada, Y. Takada, J. Minowada and St. Cohen (19761 Lymphokine-like factors produced by human lymphoid cell lines with B or T cell surface markers, J. lmmunol., ! 17: 548-554.