Cytokines in the central nervous system: Expression of macrophage colony stimulating factor and its receptor during development

Journal of Neuroimmunology

ELSEVIER

Journal of Neuroimmunology 52 (1994) 9-17

Cytokines in the central nervous system: Expression of macrophage colony stimulating factor and its receptor during development Yuan Chang ,,a, Shane Albright b, Frank Lee h " Division of Neuropathology, Department of Pathology, Columbia University College of Physicians & Surgeons, 630 West 168th St., New York, N Y 10032, USA b Department of Molecular Biology, DNAX Research Institute, 901 California Avenue, Palo Alto, CA 94304, USA Received 19 July 1993; revision received and accepted 10 February 1994

Abstract

To investigate the potential role of cytokines in the development of the central nervous system, we analyzed the production of cytokine mRNA transcripts by Sl-nuclease protection analysis in the brains of Swiss-Webster mice during fetal development and after birth. Cytokines studied were interleukin (IL)-la, IL-2, IL-3, IL-4, IL-5, IL-6, IL-7, macrophage-colony stimulating factor (M-CSF), granulocyte-colony stimulating factor (G-CSF), granulocyte macrophage-colony stimulating factor (GM-CSF), and leukemia inhibitory factor (LIF). Only mRNA transcripts for M-CSF were found to be produced constitutively in normal brain tissue. These transcripts were detected continuously from embryonic day (ED) 13 through adulthood. Transcripts encoding the M-CSF receptor (c-fms) were also detected at all of these time points. Despite identification of M-CSF transcripts in the brains of normal mice during development and M-CSF protein in cell cultures, neuropathological examination of the brains of op lop mice, a naturally occurring mouse mutant defective in the production of functional M-CSF, showed no cytoarchitectural abnormalities. Key words: Cytokines; Microglia; Macrophage colony stimulating factor (M-CSF); Osteopetrotic (op/op) mouse; Development

1. Introduction

Cytokines are hormone-like polypeptide factors that mediate growth or differentiation of various cells. Many cytokines were first described in the hemopoietic system where they have been called interleukins or colony stimulating factors. These factors are involved in controlling the growth of blood cell lineages as well as regulating various aspects of the immune response. In addition, some cytokines have been found to affect non-hemopoietic cells, including parenchymal cells of the central nervous system (CNS). Not only have cells of the CNS been found to respond to cytokines, they can also be stimulated to produce a variety of cytokines or polypeptides with cytokine-like properties. For example, cultured astrocytes can be induced to produce IL-1 (Nieto-Sampedro and Berman, 1987), IL-6 (Frei et al., 1989), GM-CSF (Malipiero et al., 1990) and

* Corresponding author. Phone (212) 305 3426; Fax (212) 305 1533. 0165-5728/94/$07.00 © 1994 Elsevier Science B.V. All rights reserved SSDI 0 1 6 5 - 5 7 2 8 ( 9 4 ) 0 0 0 2 5 - J

TNFo~ (Lieberman et al., 1989; Chung and Benveniste, 1990). Possible roles for cytokines include controlling or augmenting immune responses in the brain which is otherwise 'immunologically privileged' due to the blood-brain barrier (BBB) and lack of native lymphoid cells. In support of this hypothesis, similarities between these two cellular systems extend beyond soluble growth factors to the cells themselves. For example, astrocytes can function as antigen-presenting cells similar to macrophages and dendritic cells in the other tissues (Fontana et al., 1984; Fierz et al., 1985). Potentially, cytokines and their neural target cells may initiate and amplify a timely immunological response in the CNS prior to breaching of the BBB by recruited lymphocytes, the usual mediators of the immune response. An alternative hypothesis to explain these shared factors arises from a comparison of the hemopoietic and nervous systems in development. In both cases, multipotent progenitor or stem cells give rise to a variety of differentiated cell lineages. Underscoring the

Id Chang et al. /Journal ~)( Neuroimmunology 52 (1994) 9-17

10

similarities to hemopoiesis, the term 'neuropoiesis' has been used to describe the analogous manner in which CNS multipotent progenitor cells of the periventricular germinal tissue develop into a variety of morphologically and functionally distinct parenchymal cells (Anderson, 1989). Like hemopoiesis, it is believed that environmental signals are required in the CNS to direct the orderly appearance of cell lineages during development, and to direct the migration, organization and terminal differentiation of parenchymal cells. Since cytokines have both an effector and affector function for neural ceils in models of immunological injury, we investigated the expression of various cytokine mRNAs in the normal developing mouse brain to explore a possible role for these regulators in 'neuropoiesis.'

removed, rinsed in PBS, and then divided for use in RNA extraction or for histological examination.

2. 2. Preparation of RNA Extraction of total RNA was based on the method of Chirgwin et al. (1979). Briefly, brain tissue was homogenized in guanidinium-thiocyanate solution. Samples were then layered over cesium chloride and centrifuged for 20-24 h at 25K rpm in a SW28 or SW41 rotor. The protein and DNA layers were then removed, and the RNA pellet recovered by ethanol precipitation. Poly A + RNA was selected using oligo(dT)-cellulose chromatography (Aviv and Leder, 1972).

2.3. S1 nuclease analysis 2. Materials and methods

2.1. Animals Embryos were obtained from timed, pregnant Swiss-Webster mice (Simonsen Laboratories, Gilroy, CA) at embryonic day (ED) 13 and ED17. Pregnant females were allowed to carry their litters to term, and pups were then killed at various postnatal (PN) days: PNld, PN5d, PN10d, PN15d, PN20d. These embryonic and postnatal dates were chosen to coincide with the appearance of various cell lineages in the mouse CNS: ED13 embryos for neurons, ED17 embryos for astrocytes, and newborn and post-natal mice for oligodendrocytes. Adult brains were obtained from 10-12week-old female mice. Brains were removed from embryos, newborn mice and adults after decapitation, and rinsed in phosphate-buffered saline (PBS). These brains were then either frozen on dry ice and stored at - 8 0 ° C or immediately subjected to R N A extraction. Osteopetrotic (op/op) mice and control littermates (Jackson Laboratories, Bar Harbor, ME) were killed by decapitation at 26, 29 and 35 days. The brains were

Approximately 25 ~g of Poly A + RNA samples were hybridized with 5'[32P] end-labeled D N A probes for each of the cytokines (50 ng DNA) in a total volume of 15 p,1 of 80% formamide, 40 mM PIPES (pH 6.4), 0.4 M NaC1, and 1 mM EDTA, for a minimum of 24 h at 50°C. The resulting hybrids were treated with S1 nuclease (300 U / m l ) in 150/xl of buffer containing 250 mM NaC1, 30 mM sodium acetate (pH 4.5), and 1 mM ZnC12 (45 min. 37°C). The $1 digestions were terminated by transferring samples to an ice bath for 10 min; the hybrids were then ethanol-precipitated with 10-20 tLg of carrier tRNA and analyzed by electrophoresis on 7M u r e a / 5 % polyacrylamide gels. The gels were dried under vacuum and exposed to Kodak X A R film with intensifying screens at -80°C. The derivation of the murine cDNA probes used in these studies is summarized in Table 1. The cDNA probes were typically labelled to a specific activity of 0.5-1.0 × 106 cpm//xg. A minimum of 50 ng of probe was added to hybridizations to assure that the probe was in excess. With exposure times approximately 3-5 days used for the autoradiographs, we estimate the sensitiv-

Table 1 Hybridization probes for Sl-nuclease protection assay Gene

Plasmid

enzyme

Fragment size

Reference

IL-1 Ig-2 IL-3 Ik-4 IE-5 1L-6 IL-7 LIF G-CSF GM-CSF M-CSF c-fms

GEM-3-mlL1 GEM-I-mlL2 pUC-mIL3 genomic pcD-mlL4 pcD-mIL5 pcD-mlL6 pcD-SRa-mlL7 pcD-SRa-mLIF GEM-3-mG-CSF pcD-mGM-CSF pcD-SRA-mM-CSF p755

NdeI (linear) BglII (linear) BglII (linear) SacI (linear) NcoI (fragment) BglII (linear) ClaI (fragment) Ncol (fragment) Dra I (fragment) EcoRV (linear) BstEII (linear) ApaI (linear)

780 bp 300 bp 400 bp 256 bp 103 bp 650 bp 190 bp 295 bp 850 bp 434 bp 550 bp 13l bp

Lomedico et al. (1984) Yokata et al. (1985) Yokata et al. (1984) Lee et al. (1986) Yokata et al. (1986) Chiu et al. (1988) Namen et al. (1988) Gearing et al. (1988) Tsuchiya et al. (1986) Miyatake et al. (1985) De Lamarter et al. (1987) Rothwell et al. (1987)

Y Changet al./Journal of Neuroimmunology 52 (1994) 9-17

11

ity of the hybridization technique to be about 10 pg of specific m R N A in the 25/~g of poly A + m R N A .

2.4. Light microscopic and immunohistochemical studies Brains removed from 26, 29 and 35-day-old op l o p mice and their control littermates were immediately placed in formalin and fixed for at least 4 days. Brains were then coronally cut, paraffin-embedded, and sectioned at 5 ~m. Hematoxylin and eosin stains were applied to some sections for routine evaluation of cytological and architectural features. Microglial cell bodies and processes were demonstrated with lectin Ricinus communis agglutinin-1 (RCA-1) using an avidin-biotin peroxidase method (Mannoji et al., 1986). To determine if there were significant morphological differences between microglia in o p / o p mice and controls, lectin-stained microglia were viewed on a Laborlux S microscope with an M T I CCD72 videocamera and digitized with a data translation frame-grabber card in an N I H Image software. Pixels contained in each cell body and its visible processes were measured. To determine numerical variations, microglial from ten scattered 4 0 x power fields from o p / o p mice and controls were counted and compared. Cells were analyzed only if both cell nucleus and associated processes were identifiable. Sections were also examined for glial fibrillary acidic protein (GFAP) (DAKO, Carpenteria, CA), using a peroxidase-antiperoxidase method.

3. Results

3.1. Cytokine mRNA expression during mouse CNS development These studies were initiated to survey the expression of various cytokines in the CNS. We used S-1 nuclease analysis to detect cytokine m R N A transcripts because the method is sensitive and because specific probes can be easily prepared for each cytokine. Although other detection methods such as polymerase chain reaction (PCR) are more sensitive, the S-1 nuclease method is more readily correlated with biologically relevant levels of m R N A production. The following cytokines were analyzed in this study: IL-la, IL-3, IL-4, IL-5, IL-6, IL-7, M-CSF, GM-CSF, G-CSF, and LIF. Results of the $1 nuclease analysis are shown in Figs. 1 - 4 and summarized in Table 2. Of the cytokines that were analyzed, only M-CSF transcripts could be detected. These transcripts were present starting at ED13 and remained detectable throughout fetal and post-natal life (Fig. 3). m R N A transcripts for the M-CSF receptor which is encoded by the c-fins gene were also analyzed in the brains of fetal and neonatal mice using the $1 nuclease method.

1070872603-

IL-1 (780 bp)

118-

IL-5 (103 bp)

872 603IL-6 (650 bp) Fig. l. S-I nuclease analysis of mRNAs for the cytokines IL-1, IL-5 and IL-6 using poly A + RNA from the indicated days of embryonic (E) or post-natal (PN) mouse brains. The arrow indicates the position of the expected S-1 nuclease product based on positive controls. The positive control RNA samples were from P388D1 macrophage cells (IL-1); ConA-induced D10 T cells (IL-5); IL-l-induced 30E stromal cells (IL-6). Negative controls contained 25 /xg of yeast tRNA. IL-5 is produced at very low levels in the positive control.

Like M-CSF, c-fms transcripts were found to be present starting at ED13 and continuing throughout development and post-natal life (Fig. 4), suggesting a functional role for M-CSF in normal mouse brain development.

3.2. Characterization of op / op mice M-CSF, a hemopoietic growth factor, functions in the survival, proliferation, differentiation, and activation of cells of the m a c r o p h a g e / m o 0 o c y t e lineage. Since brain microglia have certain functions analogous to those of macrophages found in other tissues, and have been postulated to be derived from bone marrow m a c r o p h a g e / m o n o c y t e progenitors, the M-CSF in the brains of mice during pre- and post-natal life may play an essential role in the generation and maintenance of such cells. To study the possible role of M-CSF in the development of microglia in vivo, we examined mice carrying

12

E ('hang et al. /.Iournal ~I' Neuroimmunolo~,3' 52 (1~)94) ~ 17

oo

603 . . . . .

603 . . . .

A

J

w

310 281-

310221 -

:

J

0

e

D

~ '

~.~,

G M - C S F (434 bp)

m - C S F (550 bp)

oo 310_~ ~ 281- : ~ 234-

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

::

:

:

LIF (295

:

:

bp) 256IL-3 (400 bp)

310 281-

234-': IL-4 (256 bp) Fig. 2. S-1 nuclease analysis of mRNAs for the cytokines GM-CSF, LIF and 1L-4 using poly A + RNA from the indicated days of embryonic (E) or post-natal (PN) mouse brains. The arrow indicates the position of the expected S-I nuclease product based on positive controls. The positive control RNA samples were from ConA-induced D10 T-cells (GM-CSF and IL-4); |L-1 induced 30E stromal cells (LIF). Negative controls contained 25/~g of yeast tRNA.

the op m u t a t i o n , a recessive m u t a t i o n recently charact e r i z e d by m o l e c u l a r analysis to o c c u r in the c o d i n g r e g i o n of t h e M - C S F g e n e ( Y o s h i d a et al., 1990). Homozygous op/op mice p r o d u c e no f u n c t i o n a l MC S F and r e p r e s e n t a u n i q u e e x p e r i m e n t a l m o d e l for assessing how the a b s e n c e of M - C S F affects cells of the CNS. Phenotypically, op l o p mice are s m a l le r b o t h in w e i g h t and l e n g t h w h e n c o m p a r e d to th e i r c o n t r o l l i t t e r m a t e s . This is d u e to a d e f e c t in b o n e r e m o d e l i n g c a u s e d by a s e v e r e deficiency in M - C S F - d e p e n d e n t osteoclasts, cells d e r i v e d f r o m m o n o c y t e - m a c r o p h a g e pr e c ur s o rs . In addition, th e s e m i c e have a deficiency in the n u m b e r s of p e r i p h e r a l b l o o d m o n o c y t e s and perit o n e a l m a c r o p h a g e s , a l t h o u g h t h e ir p r o g e n i t o r cells are p r e s e n t in b o t h s p l e e n and b o n e m a r r o w (WiktorJ e d r z e j c z a k et al., 1982). Brains f r o m o p / o p mice w e r e d e c r e a s e d in weight

Fig. 3. S-I nuclease analysis of mRNAs for the cytokines M-CSF and IL-3 using poly A + RNA from the indicated days of embryonic (E) or post-natal (PN) mouse brains. The arrow indicates the position of the expected S-1 nuclease product based on positive controls. The positive control RNA samples were from IL-l-induced 30E stromal cells (M-CSF); ConA-induced D10 T-cells (IL-3). Negative controls contained 25 /~g of yeast tRNA. In the IL-3-positive control, the strong band approximately 250 bp long is due to digestion at an internal S-1 nuclease-sensitive site and is observed occasionally.

c o m p a r e d to brains f r o m l i t t e r m a t e controls; however, h e m i s p h e r i c sh ap e and c o n t o u r w e r e similar. Microscopic e x a m i n a t i o n o f h e m a t o x y l i n and eosin stained sections of t h e b r ai n s h o w e d n o r m a l a r c h i t e c t u r a l features with w e l l - d e m a r c a t e d gray-white interface, basal ganglia structures, and cortical layers• l m m u n o p e r o x i dase studies r e v e a l e d no i n c r e a s e in G F A P staining

• .~

:~

118 C-fins (131 bp) Fig. 4. S-I nuclease analysis of the M-CSF receptor gene, c-fms using poly A + RNA from the indicated days of embryonic (E) or post-natal (PN) mouse brains. The arrow indicates the position of the expected S-I nuclease product based on positive controls. The positive control RNA sample was from P388D1 macrophage cells. The negative control contained 25 ~g of yeast tRNA.

Y. Chang et al. /Journal of Neuroimmunology 52 (1994) 9-17

which would indicate gliosis. To identify microglia, op/op brains were immunostained with the lectin, RCA-1 (Mannoji et al., 1986). The average number of microglia counted per ten 40 x power fields total led 25 in control brains and 27 in op/op brains. Op/op brains were also comparable to controls in terms of microglia morphology (Fig. 5A, B). Digitized images

13

showed the average density of control brain microglia to be 96.33 pixels as compared to 90.80 pixels in the op/op brain. No cytoarchitectural differences were found in the brains of op/op mice with the M-CSF deficiency as compared to control mice. There are several possible explanations for the normal development of microglia in op/op animals. The

Fig. 5. Photomicrographs of sections of brains from wild-type control mouse (A) and op/op mouse (B) stained with RCA-1 using avidin-biotin peroxidase development to reveal microglia (Magnification 72 x ).

14

E Chang et al. /,hmrnal of Neuroimmunolo~' 52 (1994) 9-17

Table 2 Summary of S1 nuclease mapping results

4. Discussion

tRNA

Poly(A) + RNA

Results

IL-4 IL-5 1L-6 II-7 LIF

~/C ~iff~/~/-

~/C ~/( ~ND ~-

-

~-CSF

C

7-

-

GM-CSF M-CSF

C ~

7(-

+ +

7-

7-

+

Cytokines IL-1 IL-3

-

Receptors

c-fins

f - , determined; ND, not determined.

redundancy of many cytokines is well documented and, potentially, other hemopoietic growth factors known to be active on macrophages could be substituting for the missing M-CSF. To test this possibility, we performed S-1 nuclease assays on Ol)/Ol) mRNA using probes for IL-3, GM-CSF and IL-6, three growth factors known to have potent macrophage-stimulating activities, looking for possible up-regulation of these trophic factors that could act in lieu of M-CSF to maintain the microglial population. Only mRNA for M-CSF, which is transcribed but cannot be translated as a result of the mutation, was detected at levels comparable to controls (data not shown). These results suggest that o p / op mice have no homeostatic mechanism involving other known cytokines that could compensate for the lack of M-CSF. We cannot rule out, however, the possibility that undetectable levels of one of these other factors are responsible for supporting microglia. Another possible explanation comes from previous studies using Northern blotting that revealed multiple size classes of M-CSF mRNA, probably resulting from alternative splicing or multiple polyadenylation sites (Ladner et al., 1987; Cerretti et al., 1988). We therefore considered the possibility of tissue-specific alternative splicing of the M-CSF transcript in the mRNA from op/op mice brains leading to deletion of the mutation and production of an alternative, functional form of M-CSF. To determine whether alternative splicing might produce functional M-CSF in Ol)/OlO mice, PCR was performed on cDNA reverse transcribed from both wild-type and Ol)/Ol) mRNA using sets of primers spanning the site of the op mutation. The sizes of the resulting PCR products were consistent with predictions based on the M-CSF cDNAs which have been sequenced, and no differences were observed between wild-type and op/Ol) samples (data not shown).

The finding that hemopoietic cytokines can also be produced by neural tissues has generated a great deal of interest and conjecture. The function of these cytokines as mediators between the immune and nervous systems have been studied in various diseases including multiple sclerosis and immune-mediated encephalitis (Leist et al., 1988; Hofman et al., 1989). However, their putative role in the development of the CNS during embryogenesis has only begun to be explored. Although cells of the CNS have been shown to produce a variety of cytokines, most of these results are derived from experimental models involving immunological, chemical, or mechanical injury. But since the response of cellular systems during both development and injury is similar and involves growth, differentiation and maturation, we examined the potential role of these mediators during development. In the studies presented here, we used S1 nuclease analysis to determine in vivo cytokine mRNA expression in normal mouse brain tissue during development and after birth. We examined mRNA transcripts of cytokines with known functions in the hemopoietic system. Some of these of the cytokines may also play a role in the CNS. IL-1, for example, is produced by macrophages, but is also released by cultured microglia on stimulation (Giulian and Baker, 1986; Hetier et al., 1988) and is mitogenic for astrocytes in vitro (Giulian and Lachman, 1985). IL-2, a T-cell cytokine, has been found to influence the proliferation and differentiation of oligodendrocytes (Benveniste and Merril, 1986). IL-3, another T-cell cytokine that is a multi-lineage hemopoietic growth factor, appears to be a trophic factor for central cholinergic neurons in vitro and in vivo (Kamegai et al., 1990). LIF is a protein that has diverse effects that include the regulation of the growth and differentiation of myeloid cells, the stimulating of bone remodeling, as well as the promotion of cholinergic differentiation in cultured rat sympathetic neurons (Yamamori et al., 1989). IL-4 and IL-5, other T-cell cytokines which act on B-cells and other cells, have not been reported to have a functional role in the central nervous system. IL-6, a B-cell-stimulating factor produced by macrophages, was found to be produced by cultured microglial cells immortalized with oncogenic retroviruses (Righi et al., 1989), and is capable of supporting neuronal survival in cholinergic neuron culture (Hama et al., 1989). IL-7, a growth factor for lymphocytes also produced by stromal cells, has not been detected in the CNS. M-CSF and GM-CSF are produced by a variety of cells including stromal and endothelial cells. Their association in vivo and in vitro with CNS-derived cells has been well documented (Giulian and Ingeman, 1988; Thery et al., 1990). Of the cytokines we studied, only M-CSF was found

Y. Changet al. / Journal of Neuroimmunology52 (1994)9-17 throughout development and post-natal life. In addition, transcripts for c-fms, the cellular receptor for M-CSF were also detected throughout the same period. The detection of M-CSF receptor strongly suggests the presence of cells in the brain that are capable of responding to brain-derived M-CSF. Our results for the expression of M-CSF transcripts are consistent with those obtained in prior studies. For example, Troutt and Lee (1989), in a survey of the distribution of cytokines in adult mice, but not during development, found constitutive expression of M-CSF transcripts in all tissues including brain. Thery et al. (1990), using Northern blot analysis of mouse cerebral RNA, found M-CSF transcripts from ED14 until 2 weeks after birth. Wesselingh et al (1990), using PCR, found M-CSF as well as TGF-/3, GM-CSF, and IL-4 transcripts. The detection of GM-CSF and IL-4 m R N A by PCR and not by S1 nuclease suggests that their levels are lower than M-CSF m R N A levels and that their functional role is unclear. We did not test for TGF-/3. Microglia are thought to act primarily as phagocytic ceils of the brain. In the resting state, they are inconspicuous and smaller than macroglial ceils (astrocytes and oligodendroglia). During traumatic and immunological injury to the brain, microglia are activated and are widely believed to transform into ameboid microglia - alternatively referred to as rod-cells, gitter cells or brain macrophages. This change in morphology is accompanied by a change in the phenotypic expression of immunohistochemical marker profiles with the activated forms more closely resembling macrophages and blood-borne monocytes (Streit et al., 1988). Although microglia have been identified and described as early as 1932, controversy still exists with regard to the histogenesis of these cells. Found in the CNS throughout development, microglia are alternatively believed to be derived from the neuroectoderm arising from the periventricular germinal matrix tissue, or are thought to be of mesodermal origin arising from bone-marrow progenitor cells with subsequent migration into the brain after capillary infiltration. The results of previous experiments which indicate that M-CSF acts on microglia in vitro (Giulian and Ingeman, 1988; Alliot et al., 1991), and the results of our own work which show a similar temporal expression pattern for M-CSF m R N A and the presence of microglia in the CNS all suggest that M-CSF may be an essential factor for the growth and maintenance of the microglial population in the developing brain. This prompted us to examine the osteopetrotic ( o p / o p ) mouse which carries a recessive mutation in the M-CSF gene. This mutation consists of a single base pair insertion in the coding region of the M-CSF gene located on chromosome 3 (Yoshida et al., 1990). O p / o p mutants, although severely defective in bone remodeling in post-natal life, are relatively normal at birth -

15

most likely as a result of the presence of maternal M-CSF present in the uterine endometrium which is accessible to o p / o p embryos (Rengenstreif and Rossant, 1989). However, maintenance of various types of M-CSF-dependent cells should be affected post-natally, as is the case with osteoclasts. If one postulates that microglia are derived from mesodermal monoc y t e / m a c r o p h a g e progenitors and are M-CSF-dependent, one would expect o p / o p mice to exhibit some microglial defects. Examination of o p / o p mouse brains, however, revealed no discernible difference in microglial population compared to control littermates. These results rule out an exclusive role for M-CSF in microglial development. Because we could find no evidence for elevated levels of one of the other known macrophage-stimulating cytokines in o p / o p mice, we conclude that other undiscovered factors must be involved in maintaining the brain microglial population. Our results nevertheless do not rule out the possibility of a mesodermal origin for brain microglia. It is possible that microglial may not be terminally differentiated cells. Rather, they may be analogous to the m a c r o p h a g e / m o n o c y t e progenitor cells found in the spleen and bone marrow of normal mice and which are even present in the spleens of o p / o p mice. In o p / o p animals, these cells are fully capable of generating mature cells of the macrophage lineage in response to exogenously supplied M-CSF. Thus, brain microglia may represent a progenitor cell population that is maintained in the brain by as yet unidentified factors but which can act as the source of activated cells classically seen in the brain under conditions of injury. Further work is warranted to characterize the microglial population found in the o p / o p brain to elucidate the possible role of these cells as progenitor cells.

Acknowledgements We thank Pauline Chu for technical assistance, G. Jackson Snipes for useful discussions and Patrick S. Moore for reviewing the manuscript. DNAX Research Institute is supported by Schering Plough Corporation.

References Alliot, F., Lecain, E., Grima, B. and Pessac, B. (1991) Microglial progenitors with a high proliferative potential in the embryonic and adult mouse brain. Proc. Natl. Acad. Sci. USA 88, 1541-1545. Anderson, D.J. (1989) The neural crest cell lineage problem: Neuropoiesis? Neuron 3, 1-12. Aviv, H. and Leder, P. (1972) Purification of biologically active globin messenger RNA by chromatography on oligothymidylic acid-cellulose. Proc. Natl. Acad. Sci. USA 69, 1408-1412. Benveniste, E.N. and Merril, J.E. (1986) Stimulation of oligoden-

16

E Chang et al. /Journal of Neuroimmunolog3' 52 (1994) 9-17

droglial proliferation and maturation by interleukin-2. Nature 321, 610-613. Cerretti, D.P., Wignall, J., Anderson, D., Tushinski, R.J., Gallis, B.M., Stya, M., Gillis, S., Urdal, D.L. and Cosman, D. (1988) Human macrophage-colony stimulating factor: alternative RNA and protein processing from a single gene. Mol. Immunol. 25, 761-770. Chirgwin, J.M., Przybyla, A.E., MacDonald, R.J. and Rutter, W.J. (1979) Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochemistry 18, 5294-5299. Chiu, C., Moulds, C., Coffman, R., Rennick, D. and Lee, F. (1988) Multiple biological activities are expressed by a mouse interleukin-6 cDNA clone isolated from bone marrow stromal cells. Proc. Natl. Acad. Sci. USA 85, 7099-7103. Chung, I.Y. and Benveniste, E.N. (1990) Tumor necrosis factor-a production by astrocytes. Induction by lipopolysaccharide, IFN-y, and IL-1/3. J. Immunol. 144, 2999-3007. De Lamarter, J.F., Hession, C., Semon, D., Gough, N.M., Rothenbuhler, R. and Mermod, J.J. (1987) Nucleotide sequence of a cDNA encoding murine CSF-1 (Macrophage-CSF). Nucleic Acids Res. 15, 2389-2390. Fierz, W., Endler, B., Reske, K., Wekerle, H. and Fontana, A. (1985) Astrocytes as antigen-presenting cells. I. Induction of Ia antigen expression on astrocytes by T cells via immune interferon and its effects on antigen presentation. J. Immunol. 134, 3785-3793. Fontana, A., Fierz, W. and Wekerle, H. (1984) Astrocytes present myelin basic protein to encephalitogenic T-cell lines. Nature 307, 273-276. Frei, K., Malipiero, U.V., Leist, T.P., Zinkernagel, R.M., Schwab, M.E. and Fontana, A. (1989) On the cellular source and function of interleukin 6 produced in the central nervous system in viral diseases. Eur. J. Immunol. 19, 689-694. Gearing, D.P., King, J.A. and Gough, N.M. (1988) Complete sequence of murine myeloid leukaemia inhibitory factor. Nucleic Acids Res. 16, 9857. Giulian, D. and Baker, T.J. (1986) Characterization of ameboid microglia isolated from developing mammalian brain. J. Neurosci. 6, 2163-2178. Giulian, D. and Ingeman, J.E. (1988) Colony-stimulating factors as promoters of ameboid microglia. J. Neurosci. 8, 4704-4717. Giulian, D. and Lachman, LB. (1985) Interleukin-1 stimulation of astroglial proliferation after brain injury. Science 228, 497-499. Hama, T., Miyamoto, M., Tsukui, H., Nishio, C. and Hatanaka, H. (1989) Interleukin-6 as a neurotrophic factor for promoting the survival of cultured basal forebrain cholinergic neurons from postnatal rats. Neuroscience Lett. 104, 340-344. Hetier, E., Ayala, J., Denefle, P., Bousseau, A., Rouget, P., Mallat, M. and Prochiantz, A. (1988) Brain macrophages synthesize interleukin-1 and interleukin-1 mRNAs in vitro. J. Neurosci. Res. 21, 391-397. Hofman, F.M., Hinton, D.R., Johnson, K. and Merril, J.E. (1989) Tumor necrosis factor identified in multiple sclerosis brain. J. Exp. Med. 170, 607-612. Kamegai, M., Niijima, K., Kunishita, T., Masatoyo, N., Ogawa, M., Araki, M., Ueki, A., Konishi, Y. and Tabira, T. (1990) Interleukin 3 as atrophic factor for central cholinergic neurons in vitro and in vivo. Neuron 2, 429-436. Ladner, M.B., Martin, G.A., Noble, J.A., Nikoloff, D.M., Tal, R., Kawasaki, E.S. and White, T.J. (1987) Human CSF-I: gene structure and alternative splicing of mRNA precursors. EMBO J. 6, 2693-2698. Lee, F., Yokota, T., Otsuka, T., Meyerson, T., Vallaret, D., Coffman, R., Mosmann, T., Rennick, D., Roehm, N., Smith, C., Zlotnick, A. and Arai, K. (1986) Isolation and characterization of a mouse interleukin cDNA clone that expresses B-cell stimulatory factor-1 activities and T-cell and mast cell stimulating activities. Proc. Natl. Acad. Sci. USA 93, 2061-2065.

Leist, T.P., Frei, K., Kam-Hansen, S., Zinkernagel, R.M. and Fontana, A. (1988) Tumor necrosis factor-a in cerebrospinal fluid during bacterial, but not viral, meningitis. Evaluation in murine model infections and in patients. J. Exp. Med. 167,1743-1748. Lieberman, A.P., Pitha, P.M., Shin, H.S. and Shin, M.L. (1989) Production of tumor necrosis factor and other cytokines by astrocytes stimulated with lipopolysaccharide or a neurotropic virus. Proc. Natl. Acad. Sci. USA 86, 6348-6352. Lomedico, P.T., Gubler, U., Hellmann, C.P., Bukovich, M., Girl, J.G., Jan, Y.-C.E., Collier, K., Seminonow, R., Chua, A.O. and Mizel, S.G. (1984) Cloning and expression of murine interleukin-I cDNA in Escherichia coli. Nature 312, 458-462. Malipiero, U.V., Frei, K. and Fontana, A. (1990) Production of hemopoictic colony-stimulating factors by astrocytes. J. lmmunol. 144, 3816-3821. Mannoji, H., Yeger, H. and Becker, L.E. (1986) A specific histochemical marker (lectin Ricinus communis agglutinin-1) for normal human microglia, and application to routine histopathology. Acta. Neuropathol. (Berl). 71,341-343. Miyatake, S., Otsuka, T., Yokota, T., Lee, F. and Arai, K. (1985) Structure of the chromosomal gene for murine granulocyte-macrophage colony stimulating factor: comparison of the mouse and human genes. EMBO J. 4, 2561-2568. Namen, A.E., Lupton, S., Hjerrild, K., Wignall, J., Mochizuki, D.Y., Sehmierer, A., Mosley, B., March, D.J., Urdal, D. and Gillis, S. (1988) Stimulation of B-cell progenitors by cloned murine IL-7. Nature 333, 571--573. Nieto-Sampedro, M. and Berman, M.A. (1987) Interleukin-l-like activity in rat brain: sources, targets, and effect of injury. J. Neurosci. Res. 17, 214-219. Rengenstreif, L. and Rossant, J. (1989) Expression of the c-fms proto-oncogene and of the cytokine, CSF-1, during mouse embryogenesis. Dev. Biol. 133, 284-294. Righi, M., Mori, L., De Libero, G., Sironi, M., Biondi, A., Mantovani, A., Donini, S.D. and Ricciardi-Castagnoli, P. (1989) Monokine production by microglial cell clones. Eur. J. Immunol. 19, 1443-1448. Rothwell, V,M. and Rohrschneider, L.R. (1987) Murine c-fins cDNA: cloning, sequence analysis and retroviral expression. Oncogene Res. 1,311-324. Streit, W.J., Graeber, M.B. and Kreutzberg, G.W. (1988) Functional plasticity of microglia: a review. Glia 1, 301-307. Thery, C., Hetier, E., Evrard, D. and Mallat, M. (1990) Expression of macrophage colony-stimulating factor gene in the mouse brain during development. J. Neurosci. Res. 26, 129-133. Troutt, A.B. and Lee, F. (1989) Tissue distribution of murine hemopoietic growth factor mRNA production. J. Cell. Physicol. 138, 38-44. Tsuchiya, M., Asano, S., Kaziro, Y. and Nagata S. (1986) Isolation and characterization of the cDNA for murine granulocyte colony-stimulating factor. Proc. Acad. Sci. USA 83, 7633-7637. Wesselingh, S.L., Gough, N.M., Finlay-Jones, J.J. and McDonald, P.J. (1990) Detection of cytokine mRNA in astrocyte cultures using the polymerase chain reaction. Lymphokine Res. 9, 177-183. Wiktor-Jedrzejczak, W., Ahmed, A., Szczylik, C. and Skelly, R.R. (1982) Hematological characterization of congenital osteopetrosis in o p / o p mice. J. Exp. Med. 156, 1516-1527. Yamamori, T., Fukada, K., Aebersold, R., Korsching, S., Fann, M.-J. and Patterson, P.H. (1989) The cholinergic neuronal differentiation factor from heart cells is identical to leukemia inhibitory factor. Science 246, 1412-1416. Yokota, T., Lee, F., Rennick, D., Hall, C., Arai, N., Mosmann, T., Nabel, G., Cantor, H. and Arai, K. (1984) Isolation and characterization of mouse cDNA clone that expresses mast cell growth factor activity in monkey cells. Proc. Acad. Sci, USA. 81, 10701074. Yokota, T., Arai, N., Lee, F., Rennick, D., Mosmann, T. and Arai,

Y. Chang et al. / Journal of Neuroimmunology 52 (1994) 9-17 K. (1985) Use of a cDNA expression vector for isolation of mouse interleukin-2 cDNA clones: expression of T-cell growth factor activity after transefection of monkey cells. Proc. Acad. Sci. USA 82, 68-72. Yokota, T., Coffman, R.L., Hagiwara, H., Rennick, D.M., Takebe, Y., Yokota, K., Gemmel, L., Shrader, B., Yang, G., Meyerson, R., Hoy, R., Bancherear, J., de Vries, J., Lee, F., Arai, N. and Arai, K. (1987) Isolation and characterization of lymphokine

17

cDNA clones encoding mouse and human IgA-enhancing factor and eosinophil colony-stimulating gacotr activities: relationship to IL-5. Proc. Acad. Sci. USA 84, 7388-7392. Yoshida, H., Hayashi, S.-I., Kunisada, T., Ogawa, M., Nishikawa, S., Okamura, H., Sudo, T., Shiltz, L.D. and Nishikawa, S.-I. (1990) The murine mutation osteopetrosis is in the coding region of the macrophage colony stimulating factor gene. Nature 345, 442-444.