- Email: [email protected]
Calmodulin is expressed by reactive microglia in the hippocampus of kainic acid-treated mice
Pergamon
PII:
Neuroscience Vol. 81, No. 3, pp. 699–705, 1997 Copyright ? 1997 IBRO. Published by Elsevier Science Ltd Printed in Great Britain. All rights reserved 0306–4522/97 $17.00+0.00 S0306-4522(97)00245-5
CALMODULIN IS EXPRESSED BY REACTIVE MICROGLIA IN THE HIPPOCAMPUS OF KAINIC ACID-TREATED MICE C. SOLA v ,*‡ J. M. TUSELL† and J. SERRATOSA* *Department of Pharmacology and Toxicology, Institut d’Investigacions Biome`diques de Barcelona, CSIC, c/Jordi Girona 18-26, 08034 Barcelona, Spain †Department of Neurochemistry, Institut d’Investigacions Biome`diques de Barcelona, CSIC, c/Jordi Girona 18-26, 08034 Barcelona, Spain Abstract––Calmodulin is a calcium-binding protein that is highly abundant in the brain, where it is involved in many essential functions. The protein is mainly expressed by neuronal cells. Calmodulin is encoded by three different genes in mammals, all of them producing an identical protein. Alterations in the expression of either calmodulin genes or protein have been reported in the rodent brain by several authors in different experimental situations. However, no mention has been made to date of possible alterations in calmodulin expression in glial cells in response to certain stimuli. In the present study, we found an increase in the expression of calmodulin in reactive microglial cells in the mouse hippocampus 24 h after an intraperitoneal administration of a convulsant dose of kainic acid. The results show that a high expression of calmodulin can be added to the list of changes described to occur in microglial cells when they become reactive microglia in response to certain kinds of stimuli, in contrast to the non-detectable level of expression of this protein observed in the resting microglial cells. It is difficult to explain such an increase due to the great number of processes in which calmodulin is involved, but the great level of calmodulin observed in the reactive microglial cells shows that calmodulin immunolabelling can be used to reveal these kinds of cells. ? 1997 IBRO. Published by Elsevier Science Ltd. Key words: calmodulin, microglia, excitotoxicity, convulsant, immunohistochemistry.
Calmodulin is a major calcium-binding protein in eukaryotic non-muscle and smooth muscle cells. The activity of the protein is regulated by the intracellular levels of Ca2+, and many of the effects of Ca2+ appear to be mediated by calmodulin-regulated enzymes. Calmodulin has a fundamental role in cellular function6 and it is a key regulator of cell proliferation.23,28 This protein is very abundant in the central nervous system, where it is involved in neurotransmission.8,9,13 Calmodulin is widely expressed in the rodent brain,5,12,19,25,30,32 where it is encoded by three different genes, which produce an identical protein.27 It appears to be expressed mainly by neuronal cells, but it is also expressed by glial cells. We found detectable levels of calmodulin in the nuclei of both neuronal and glial cells from the rat brain in a previous study.38 In a recent study, we describe alterations in the expression of calmodulin genes and protein in the brain of mice injected intraperitoneally with a convulsant dose of kainic acid, a potent neurotoxic agent.33 Alterations in the expression of one calmodulin gene or another in neurons has also been
reported in several experimental situations by other authors.5,14,21,26 However, no interest has been focused until now on determining possible alterations in the expression of calmodulin in the different types of glial cells in response to certain stimuli. In our above mentioned work,33 we additionally observed the presence of calmodulin immunolabelling in reactive glial cells in the hippocampus of kainic acid-treated mice. In the present study, we characterize the glial reaction occurring in the mouse hippocampus following the administration of a convulsant dose of kainic acid to identify the type of glial cell (astrocytes and/or microglial cells) responsible for calmodulin immunoreactivity. In addition to calmodulin immunoreactivity, glial fibrillary acidic protein (GFAP) immunoreactivity was used to identify astroglial cells and lectin reaction was considered to detect microglial cells in the hippocampus of control and kainic acid-treated mice.
‡To whom correspondence should be addressed. Abbreviations: ABC, avidin–biotin–horseradish peroxidase complex; GFAP, glial fibrillary acidic protein; PB, phosphate buffer; PBS, phosphate-buffered saline. 699
EXPERIMENTAL PROCEDURES
Experimental animals and tissue processing Animal care followed the Spanish Legislation on Protection of Animals Used for Experimental and Other Scientific Purposes, in agreement with EC regulations (O.J. of EC no. L 358/1, 18/12/1986). Male OF1 mice (Iffa Credo, 30 g body weight) were used. Mice were anaesthetized with a mixture of ketamine and xylacine and killed by
700
C. Sola` et al.
transcardiac perfusion. Each animal was perfused with a short initial flush of 0.1 M phosphate buffer (PB), pH 7.4, and it was fixed with 100 ml of a freshly prepared 4% paraformaldehyde solution in the same PB. For postfixation, brains were removed and immersed in the same fixative for 12 h, and then washed in the PB. The brains were kept at 4)C in PB until coronal sections (50 µm thick) were obtained using a Vibratome. These sections were then stored at "20)C in a cryoprotective solution (30% glycerol, 30% ethyleneglycol, 40% phosphate-buffered saline (PBS: 2.7 mM KCl/1.5 mM KH2PO4/136 mM NaCl/8 mM Na2HPO4)). Kainic acid treatment Mice were injected i.p. with a dose of 55 mg/kg of kainic acid (0.2 ml kainic acid dissolved in saline/30 g body weight) and the animals that showed convulsions during the first hour following the administration were selected. Four convulsant mice were killed 24 h after treatment for immunohistochemical studies. Three control animals were injected with saline alone and processed in parallel to kainic acid-treated animals. Histological examination Coronal brain sections of the kainic acid-treated mice were stained with Cresyl Violet in order to determine the possible presence of areas of neuronal loss as a consequence of kainic acid treatment. Antibodies and immunocytochemistry reagents Monoclonal mouse anti-calmodulin antibody was obtained from Zymed Laboratories. This antibody is against a synthetic peptide corresponding to amino acids 128–148 of bovine brain calmodulin. Polyclonal rabbit anti-cow GFAP was obtained from DAKO. Biotinylated lectin from Lycopersicon esculentum (tomato) was obtained from Sigma to visualize microglial cells.1,37 Horse serum, goat serum, blocking kit to avoid non-specific binding of biotin–avidin system reagents, biotinylated secondary antibodies and avidin–biotin–horseradish peroxidase (ABC kits) were obtained from Vector. Immunocytochemistry Free-floating 50 µm coronal brain sections, which had been obtained using a Vibratome and stored at "20)C with a cryoprotective solution as described above, were washed twice in PBS for 10 min; they were then incubated in 3% H2O2 and 10% methanol in PBS for 25 min and washed in PBS and PBS–0.5% Triton. The sections processed for calmodulin and GFAP immunoreactivity were incubated with PBS–0.5% Triton/0.2% gelatin/3% normal horse serum for 2 h (in the case of the polyclonal antibody to GFAP, the sections were also incubated with 0.2% glycine and 0.2% lysine) and then washed again in PBS–0.5% Triton. All the sections were incubated with the blocking kit reagents to block non-specific binding of biotin–avidin system reagents and then with the primary antibody or the biotinylated lectin (calmodulin antibody 1:200; GFAP antibody 1:400; biotinylated lectin 2.5 µg/ml) in PBS–0.2% Triton/2% gelatin/1% normal horse serum (without the serum in the case of lectin sections) overnight a 4)C. The sections processed for calmodulin and GFAP immunoreactivity were then incubated with the biotinylated second antibodies. The sections processed to detect calmodulin immunoreactivity were incubated with biotinylated horse anti-mouse IgG as second antibody. To detect GFAP immunoreactivity the sections were incubated with biotinylated goat anti-rabbit
IgG. In both cases dilutions and incubation times followed the ABC kit manufacturer’s recommendations. In all cases (biotinylated lectin and biotinylated second antibodies), the sections were processed according to the conventional ABC protocol to detect the colorimetric reaction. Each step was followed by washing in PBS–0.5% Triton and PBS. The staining was developed with 0.025% diaminobenzidine in PBS containing 0.03% H2O2. Finally, sections were washed in PBS, mounted in gelatin-coated slides, air dried, dehydrated in alcohol, cleared in xylene and coverslipped in DPX. Controls: (1) The omission of the primary antibody abolished the immunostaining. (2) As we used a monoclonal antibody to localize calmodulin in mouse sections, we performed an additional control to check the specificity of the signal obtained. Since the calmodulin antibody was produced in isotype IgG1, we used another monoclonal antibody (anti-calcineurin) produced in the same isotype and we performed the immunocytochemistry in sections from the same animals (control and kainic acid-treated mice) considering the same isotype concentration. We obtained a different pattern of immunolabelling for each antibody, showing the specificity of the response of the tissue to each antibody. Quantification Several counts of the number of glial-immunoreactive cells per unit area (0.25 mm2) were performed to evaluate possible differences between control and kainic acid-treated mice. Three different areas were considered per animal for this purpose.
RESULTS
We did not detect areas of neuronal loss in kainic acid-treated mice after Cresyl Violet staining of the brain sections (Fig. 1A and B). In the control mouse hippocampus, calmodulin immunoreactivity was mainly localized in the pyramidal cells of Ammon’s horn and the granule cells of the dentate gyrus (Fig. 1C). Nevertheless, a large number of calmodulin-immunoreactive glial cells were observed in the hippocampus of kainic acid-treated mice (Figs 1D and 2B). No glial cells of this type were detected in the control hippocampus by calmodulin immunostaining (Figs 1C and 2A). In order to identify the glial cells responsible for calmodulin immunoreactivity, we used lectin as a marker of microglial cells and GFAP antibody as a marker of astrocytes. Lectin-reactive cells (microglial cells) were detected throughout the hippocampus of control mice (Fig. 2C). These cells were constituted by small cell bodies and numerous long, thin, branched processes, all of them characteristic of resting microglial cells. Lectinreactive cells were also observed in the hippocampus of kainic acid-treated mice (Fig. 2D). Nevertheless, the morphology of these cells was substantially different from that observed in the control mice: their processes were much thicker and shorter than in the control microglial cells, characteristic features of activated microglial cells. On the other hand, we observed an increase in the density of microglial cells in the hippocampus of kainic acid-treated mice when compared to the control animals (Table 1). In
Fig. 1. Glial reaction in the hippocampus of kainic acid-treated mice. (A and B) Cresyl Violet staining of the hippocampus in coronal brain sections of a control (A) and a kainic acid-treated mouse (B) showing the absence of areas of neuronal death. (C and D) Calmodulin immunoreactivity in the hippocampus of a control (C) and a kainic acid-treated mouse (D) showing the presence of immunolabelled glial cells throughout the hippocampus of the treated mouse (arrowheads). Scale bar=1 mm. Abbreviations: CA1–CA3, CA1–CA3 fields of Ammon’s horn; DG, dentate gyrus; pyr, pyramidal cell layer of the hippocampus; g, granulle cell layer of the dentate gyrus.
702
C. Sola` et al.
Fig. 2. Immunolocalization of calmodulin in reactive microglial cells of the hippocampus of kainic acid-treated mice. Calmodulin immunoreactivity (A and B), lectin reaction (C and D) and GFAP immunoreactivity (E and F) in the hippocampus of control (left column) and kainic acid-treated mice (right column). Notice the presence of numerous calmodulin-immunoreactive glial cells in the hippocampus of kainic acid-treated mice (B). These immunoreactive cells are shown in more detail in the inset (small arrows). Notice also the clear microglial reaction in the hippocampus of the treated animals (D): the processes of microglial cells (lectin-positive cells) become thicker and shorter than in the control hippocampus (insets in C and D). All these changes occur in the absence of changes in the astroglia (GFAP-positive cells) (E and F). Arrows point to latent microglial cells (LM), small arrows show reactive microglial cells (RM) and arrowheads show astrocytes (A). Scale bars=300 µm (A to F) and 50 µm (A to F, insets).
addition, there were no differences between the number of reactive microglial cells and the number of calmodulin-immunoreactive glial cells in the hippocampus of the treated animals (Table 1).
GFAP-immunoreactive cells (astrocytes) were evenly distributed in the hippocampus of control mice. No differences were observed between control and kainic acid-treated mice (Fig. 2E, F).
Calmodulin expression in reactive microglia Table 1. Density of lectin-labelled and calmodulinimmunoreactive microglial cells (number of microglial cells/ 0.25 mm2) in the hippocampus of control and kainic acidtreated mice Experimental situation Lectin reaction Calmodulin immunoreactivity
Control
Kainic acid
10.83&0.17 Non-detected
57.87&2.38a 64.00&2.14b
Data represent means&S.E.M. (n=3 mice/group; three areas of 0.25 mm2 each were measured per animal). a Statistically significantly different from the control value (Student’s t-test, P<0.001). b No differences were observed between the densities of calmodulin-immunoreactive glial cells and lectin-reactive cells in the hippocampus of kainic acid-treated mice.
DISCUSSION
The systemic administration of a convulsant dose of kainic acid to mice induced a glial reaction in the brain. This glial reaction was mainly constituted by reactive microglial cells, which were mostly detected in the hippocampus of kainic acid-treated mice. Strong immunoreactivity for calmodulin was observed in these reactive microglial cells, while control microglial cells did not show significant calmodulin immunolabelling. It has been widely documented that non-neuronal cells can respond to neuronal cell injury, constituting what is known as glial activation. Astroglial or microglial activation has been observed in the CNS as a response to several experimental situations in which neuronal injury is induced, such as ischaemia,20,29,31 axotomy,4,15,17 electrical stimulation22 or neurotoxic insult.2,3,20,36 Glial activation has also been observed in the human brain in neurodegenerative diseases.7,16,18 We observed a glial reaction in the hippocampus of kainic acid-treated mice 24 h after administration, which appeared to represent the activation of microglial cells, while no astroglial activation was detected. Both astroglial and microglial activations have been reported in the hippocampus of kainic acid-treated rats in the presence of neuronal degeneration.3,20 These authors show that the microglial reaction is an early response to injury while the astroglial activation appears later. Accordingly, the absence of astroglial reaction in our experimental conditions may be because it occurs later. On the other hand, differences in the response of the activated glia have also been observed, depending on the degree of injury attained.17,20,35 Thus, the glial reaction induced in the presence of neuronal degeneration is very different from the glial reaction in cases in which the injury does not result in neuronal death. Indeed, we did not detect areas of neuronal loss either at 24 h of kainic acid administration or at the different times following kainic acid administration considered in a previous study (two, four and eight days).33 Although the alterations
703
in neuronal function that we induced with kainic acid produced microglial activation, they may not have been strong enough to result in astroglial reaction. We also observed an increase in the number of microglial cells in the hippocampus following kainic acid administration. This increase could be due to microglial cell proliferation and/or to the migration of microglial cells from other brain areas. An extracerebral origin of these cells could be ruled out for three reasons: (i) the short delay in the appearance of these cells after kainic acid administration; (ii) the morphology of the reactive microglial cells, which does not resemble that of macrophages; and (iii) the absence of areas of neuronal death that could explain the presence of macrophages arising from the blood vessels. Glial activation is associated with both morphological and phenotypic changes, which reflects either properties of the cells before the injury, or the expression of new properties. Thus, activated microglia have been reported to increase the expression or to express de novo several molecules, such as surface antigens (type 3 complement receptors or CR3, major histocompatibility complex), heat-shock protein 72, cytokines and basic fibroblast growth factor, among others.10,11,24,34,39 This is the first report of an increase in the expression of calmodulin in activated microglia. As calmodulin is involved in many cellular functions it is difficult to establish the reason for this increase. For instance, as calmodulin is involved in the regulation of cytoeskeletal proteins, the increase in calmodulin expression could be related to the morphological changes observed in the activated microglia. On the other hand, if the observed increase in the density of the microglial cells in the hippocampus after kainic acid is due to the proliferation of the microglial cells, the role of calmodulin as key regulator of cell proliferation could also account for the increased calmodulin expression in these cells. Probably more than one factor is responsible for the great increase in the density of microglial cells observed. Additional experiments should be performed to elucidate the possible role of the increased calmodulin expression in the reactive microglial cells. Irrespective of its function, it is clear that calmodulin is strongly expressed in reactive microglia and that calmodulin immunocytochemistry can be used to show the presence of reactive microglial cells in brain sections. Acknowledgements—This work was partially supported by grants SAF 95-0041-CO2-01 from CICYT (Comisio´n de Investigacio´n Cientı´fica y Te´cnica), 95-0206 from FIS (Fondo de Investigaciones Sanitarias) and GR93-8.012 from CIRIT (Comissio´ Interdepartamental de Recerca i Tecnologia). The authors thank Dr Marc Soriano for supplying the Tomato lectin and for thoughtful comments during the realization of the present work. The authors also thank Eduard Bustamante for technical assistance with photographic material.
704
C. Sola` et al. REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37.
Acarı´n L., Vela J. M., Gonza´lez B. and Castellano B. (1994) Demonstration of poly-N-acetyl-lactosamine residues in ameboid and ramified microglial cells in rat brain by tomato lectin binding. J. Histochem. Cytochem. 42, 1033–1041. Acarı´n L., Gonza´lez B., Castellano B. and Castro A. J. (1996) Microglial response to N-methyl--aspartate-mediated excitotoxicity in the immature rat brain. J. comp. Neurol. 367, 361–374. Akiyama H., Tooyama I., Kondo H., Ikeda K., Kimura H., McGeer E. G. and McGeer P. L. (1994) Early response of brain resident microglia to kainic acid-induced hippocampal lesions. Brain Res. 635, 257–268. Barron K. D., Marciano F. F., Amundson R. and Mankes R. (1990) Perineuronal glial responses after axotomy of central and peripheral axons. A comparison. Brain Res. 523, 219–229. Barro´n S., Tusell J. M. and Serratosa J. (1995) Effect of hexachlorocyclohexane isomers on calmodulin mRNA expression in the central nervous system. Molec. Brain Res. 30, 279–286. Cheung W. Y. (1980) Calmodulin plays a pivotal role in cellular regulation. Science 207, 19–27. Delacourte A. (1990) General and dramatic glial reaction in Alzheimer brains. Neurology 40, 33–37. DeLorenzo R. J. (1981) The calmodulin hypothesis of neurotransmission. Cell Calcium 2, 365–385. DeLorenzo R. J. (1982) Calmodulin in neurotransmitter release and synaptic function. Fedn Proc. 41, 2265–2272. Finsen B. R., Jorgensen M. B., Diemer N. H. and Zimmer J. (1993) Microglial MHC antigen expression after ischemic and kainic acid lesions of the adult hippocampus. Glia 7, 41–49. Frautschy S. A., Walicke P. A. and Baird A. (1991) Localization of basic fibroblast growth factor and its mRNA after CNS injury. Brain Res. 553, 291–299. Gannon M. N. and McEwen B. S. (1994) Distribution and regulation of calmodulin mRNAs in rat brain. Molec. Brain Res. 22, 186–192. Gnegy M. E. (1993) Calmodulin in neurotransmitter and hormone action. A. Rev. Pharmac. Toxicol. 32, 45–70. Gnegy M. E., Agrawal A., Hewlett K., Yeung E. and Yee S. (1994) Repeated haloperidol increases both calmodulin and a calmodulin-binding protein in rat striatum. Molec. Brain Res. 27, 195–204. Graeber M. B. and Kreutzberg G. W. (1986) Astrocytes increase in glial fibrillary acidic protein during retrograde changes of facial motor neurons. J. Neurocytol. 15, 363–373. Graeber M. B. and Streit W. J. (1990) Microglia: immune network in the CNS. Brain Path. 1, 2–5. Graeber M. B., Streit W. J. and Kreutzberg G. W. (1990) The glial response to motor neuron axotomy: role of microglia and astrocytes. Proceedings of the XIth International Congress of Neuropathology, pp. 282–287. Haga S., Akai K. and Ishii T. (1989) Demonstration of microglial cells in and around senile (neuritic) plaques in the Alzheimer brain. Acta neuropath. 77, 569–575. Ikeshima H., Yuasa S., Matsuo K., Kawamura K., Hata J. and Takano T. (1993) Expression of three nonallelic genes coding calmodulin exhibits similar localization on the central nervous system of adult rats. J. Neurosci. Res. 36, 111–119. Jorgensen M. B., Finsen B. R., Jensen M. B., Castellano B., Diemer N. H. and Zimmer J. (1993) Microglial and astroglial reactions to ischemic and kainic acid-induced lesions to the adult rat hippocampus. Expl Neurol. 120, 70–88. Landry C. F., Ivy G. O. and Brown I. R. (1992) Effect of a discrete dorsal forebrain lesion in the rat on the expression of neuronal and glial-specific genes: induction of calmodulin, NF-L, SC1, and GFPA mRNA. J. Neurosci. Res. 32, 280–289. Lee-Hall L., Borke R. C. and Anders J. J. (1989) Transection or electrical stimulation of the hypoglossal nerve increases glial fibrillary acidic protein immunoreactivity in the hypoglossal nucleus. Brain Res. 490, 157–161. Lu K. P. and Means A. R. (1993) Regulation of the cell cycle by calcium and calmodulin. Endocr. Rev. 14, 40–58. Morioka T., Kalehua A. N. and Streit W. J. (1992) Progressive expression of immunomolecules on microglial cells in rat dorsal hippocampus following transient forebrain ischemia. Acta neuropath. 83, 149–157. Ni B., Rush S., Gurd J. W. and Brown I. R. (1992) Molecular cloning of calmodulin mRNA species which are preferentially expressed in neurons in the rat brain. Molec. Brain Res. 13, 7–17. Ni B. and Brown I. R. (1993) Modulation of a neuronal calmodulin mRNA species in the rat brain stem by reserpine. Neurochem. Res. 18, 185–192. Nojima H. (1989) Structural organisation of multiple rat calmodulin genes. J. molec. Biol. 208, 269–282. Rasmussen C. D. and Means A. R. (1989) Calmodulin, cell growth and gene expression. Trends Neurosci. 12, 433–438. Rischke R. and Krieglstein J. (1991) Postischemic neuronal damage causes astroglial activation and increase in local cerebral glucose utilization of rat hippocampus. J. cerebr. Blood Flow Metab. 11, 106–113. Roberts-Lewis J. M., Cimino M., Krause R. G. II, Tyrrell D. F. Jr., Davis L. G., Weiss B. and Lewis M. E. (1990) Anatomical localization of calmodulin mRNA in the rat brain with cloned cDNA and synthetic oligonucleotide probes. Synapse 5, 247–254. Sharp F. R., Lowenstein D., Simon R. and Hisanaga K. (1991) Heat shock protein hsp 72 induction in cortical and striatal astrocytes and neurons following infarction. J. cerebr. Blood Flow Metab. 11, 621–627. Sola` C., Tusell J. M. and Serratosa J. (1996) Comparative study of the pattern of expression of calmodulin messenger RNAs in the mouse brain. Neuroscience 75, 245–256. Sola` C., Tusell J. M. and Serratosa J. (1997) Differential response of calmodulin genes in the mouse brain after systemic kainic acid administration. Neuroscience 78, 155–164. Soriano M., Planas A. M., Rodrı´guez-Farre´ E. and Ferrer I. (1994) Early 72-kDa heat shock protein induction in microglial cells following focal ischemia in the rat brain. Neurosci. Lett. 182, 205–207. Streit W. J. (1993) Microglial–neuronal interactions. J. chem. Neuroanat. 6, 261–266. van den Berg K. J. and Gramsbergen J. B. P. (1993) Long-term changes in glial fibrillary acidic protein and calcium levels in rat hippocampus after a single systemic dose of kainic acid. Ann. N. Y. Acad. Sci. 679, 394–401. Vela J. M., Dalmau I., Gonza´lez B. and Castellano B. (1996) The microglial reaction in spinal cords of jimpy mice is related to apoptotic oligodendrocytes. Brain Res. 712, 134–142.
Calmodulin expression in reactive microglia 38.
705
Vendrell M., Aligue´ R., Bachs O. and Serratosa J. (1991) Presence of calmodulin and calmodulin-binding proteins in the nuclei of brain cells. J. Neurochem. 57, 622–628. 39. Wießner C., Gehrmann J., Lindholm D., To¨pper R., Kreutzberg G. W. and Hossman K. A. (1993) Expression of transforming growth factor-â1 and interleukin-1â mRNA in rat brain following transient forebrain ischemia. Acta neuropath. 86, 439–446. (Accepted 6 May 1997)
PII:
Neuroscience Vol. 81, No. 3, pp. 699–705, 1997 Copyright ? 1997 IBRO. Published by Elsevier Science Ltd Printed in Great Britain. All rights reserved 0306–4522/97 $17.00+0.00 S0306-4522(97)00245-5
CALMODULIN IS EXPRESSED BY REACTIVE MICROGLIA IN THE HIPPOCAMPUS OF KAINIC ACID-TREATED MICE C. SOLA v ,*‡ J. M. TUSELL† and J. SERRATOSA* *Department of Pharmacology and Toxicology, Institut d’Investigacions Biome`diques de Barcelona, CSIC, c/Jordi Girona 18-26, 08034 Barcelona, Spain †Department of Neurochemistry, Institut d’Investigacions Biome`diques de Barcelona, CSIC, c/Jordi Girona 18-26, 08034 Barcelona, Spain Abstract––Calmodulin is a calcium-binding protein that is highly abundant in the brain, where it is involved in many essential functions. The protein is mainly expressed by neuronal cells. Calmodulin is encoded by three different genes in mammals, all of them producing an identical protein. Alterations in the expression of either calmodulin genes or protein have been reported in the rodent brain by several authors in different experimental situations. However, no mention has been made to date of possible alterations in calmodulin expression in glial cells in response to certain stimuli. In the present study, we found an increase in the expression of calmodulin in reactive microglial cells in the mouse hippocampus 24 h after an intraperitoneal administration of a convulsant dose of kainic acid. The results show that a high expression of calmodulin can be added to the list of changes described to occur in microglial cells when they become reactive microglia in response to certain kinds of stimuli, in contrast to the non-detectable level of expression of this protein observed in the resting microglial cells. It is difficult to explain such an increase due to the great number of processes in which calmodulin is involved, but the great level of calmodulin observed in the reactive microglial cells shows that calmodulin immunolabelling can be used to reveal these kinds of cells. ? 1997 IBRO. Published by Elsevier Science Ltd. Key words: calmodulin, microglia, excitotoxicity, convulsant, immunohistochemistry.
Calmodulin is a major calcium-binding protein in eukaryotic non-muscle and smooth muscle cells. The activity of the protein is regulated by the intracellular levels of Ca2+, and many of the effects of Ca2+ appear to be mediated by calmodulin-regulated enzymes. Calmodulin has a fundamental role in cellular function6 and it is a key regulator of cell proliferation.23,28 This protein is very abundant in the central nervous system, where it is involved in neurotransmission.8,9,13 Calmodulin is widely expressed in the rodent brain,5,12,19,25,30,32 where it is encoded by three different genes, which produce an identical protein.27 It appears to be expressed mainly by neuronal cells, but it is also expressed by glial cells. We found detectable levels of calmodulin in the nuclei of both neuronal and glial cells from the rat brain in a previous study.38 In a recent study, we describe alterations in the expression of calmodulin genes and protein in the brain of mice injected intraperitoneally with a convulsant dose of kainic acid, a potent neurotoxic agent.33 Alterations in the expression of one calmodulin gene or another in neurons has also been
reported in several experimental situations by other authors.5,14,21,26 However, no interest has been focused until now on determining possible alterations in the expression of calmodulin in the different types of glial cells in response to certain stimuli. In our above mentioned work,33 we additionally observed the presence of calmodulin immunolabelling in reactive glial cells in the hippocampus of kainic acid-treated mice. In the present study, we characterize the glial reaction occurring in the mouse hippocampus following the administration of a convulsant dose of kainic acid to identify the type of glial cell (astrocytes and/or microglial cells) responsible for calmodulin immunoreactivity. In addition to calmodulin immunoreactivity, glial fibrillary acidic protein (GFAP) immunoreactivity was used to identify astroglial cells and lectin reaction was considered to detect microglial cells in the hippocampus of control and kainic acid-treated mice.
‡To whom correspondence should be addressed. Abbreviations: ABC, avidin–biotin–horseradish peroxidase complex; GFAP, glial fibrillary acidic protein; PB, phosphate buffer; PBS, phosphate-buffered saline. 699
EXPERIMENTAL PROCEDURES
Experimental animals and tissue processing Animal care followed the Spanish Legislation on Protection of Animals Used for Experimental and Other Scientific Purposes, in agreement with EC regulations (O.J. of EC no. L 358/1, 18/12/1986). Male OF1 mice (Iffa Credo, 30 g body weight) were used. Mice were anaesthetized with a mixture of ketamine and xylacine and killed by
700
C. Sola` et al.
transcardiac perfusion. Each animal was perfused with a short initial flush of 0.1 M phosphate buffer (PB), pH 7.4, and it was fixed with 100 ml of a freshly prepared 4% paraformaldehyde solution in the same PB. For postfixation, brains were removed and immersed in the same fixative for 12 h, and then washed in the PB. The brains were kept at 4)C in PB until coronal sections (50 µm thick) were obtained using a Vibratome. These sections were then stored at "20)C in a cryoprotective solution (30% glycerol, 30% ethyleneglycol, 40% phosphate-buffered saline (PBS: 2.7 mM KCl/1.5 mM KH2PO4/136 mM NaCl/8 mM Na2HPO4)). Kainic acid treatment Mice were injected i.p. with a dose of 55 mg/kg of kainic acid (0.2 ml kainic acid dissolved in saline/30 g body weight) and the animals that showed convulsions during the first hour following the administration were selected. Four convulsant mice were killed 24 h after treatment for immunohistochemical studies. Three control animals were injected with saline alone and processed in parallel to kainic acid-treated animals. Histological examination Coronal brain sections of the kainic acid-treated mice were stained with Cresyl Violet in order to determine the possible presence of areas of neuronal loss as a consequence of kainic acid treatment. Antibodies and immunocytochemistry reagents Monoclonal mouse anti-calmodulin antibody was obtained from Zymed Laboratories. This antibody is against a synthetic peptide corresponding to amino acids 128–148 of bovine brain calmodulin. Polyclonal rabbit anti-cow GFAP was obtained from DAKO. Biotinylated lectin from Lycopersicon esculentum (tomato) was obtained from Sigma to visualize microglial cells.1,37 Horse serum, goat serum, blocking kit to avoid non-specific binding of biotin–avidin system reagents, biotinylated secondary antibodies and avidin–biotin–horseradish peroxidase (ABC kits) were obtained from Vector. Immunocytochemistry Free-floating 50 µm coronal brain sections, which had been obtained using a Vibratome and stored at "20)C with a cryoprotective solution as described above, were washed twice in PBS for 10 min; they were then incubated in 3% H2O2 and 10% methanol in PBS for 25 min and washed in PBS and PBS–0.5% Triton. The sections processed for calmodulin and GFAP immunoreactivity were incubated with PBS–0.5% Triton/0.2% gelatin/3% normal horse serum for 2 h (in the case of the polyclonal antibody to GFAP, the sections were also incubated with 0.2% glycine and 0.2% lysine) and then washed again in PBS–0.5% Triton. All the sections were incubated with the blocking kit reagents to block non-specific binding of biotin–avidin system reagents and then with the primary antibody or the biotinylated lectin (calmodulin antibody 1:200; GFAP antibody 1:400; biotinylated lectin 2.5 µg/ml) in PBS–0.2% Triton/2% gelatin/1% normal horse serum (without the serum in the case of lectin sections) overnight a 4)C. The sections processed for calmodulin and GFAP immunoreactivity were then incubated with the biotinylated second antibodies. The sections processed to detect calmodulin immunoreactivity were incubated with biotinylated horse anti-mouse IgG as second antibody. To detect GFAP immunoreactivity the sections were incubated with biotinylated goat anti-rabbit
IgG. In both cases dilutions and incubation times followed the ABC kit manufacturer’s recommendations. In all cases (biotinylated lectin and biotinylated second antibodies), the sections were processed according to the conventional ABC protocol to detect the colorimetric reaction. Each step was followed by washing in PBS–0.5% Triton and PBS. The staining was developed with 0.025% diaminobenzidine in PBS containing 0.03% H2O2. Finally, sections were washed in PBS, mounted in gelatin-coated slides, air dried, dehydrated in alcohol, cleared in xylene and coverslipped in DPX. Controls: (1) The omission of the primary antibody abolished the immunostaining. (2) As we used a monoclonal antibody to localize calmodulin in mouse sections, we performed an additional control to check the specificity of the signal obtained. Since the calmodulin antibody was produced in isotype IgG1, we used another monoclonal antibody (anti-calcineurin) produced in the same isotype and we performed the immunocytochemistry in sections from the same animals (control and kainic acid-treated mice) considering the same isotype concentration. We obtained a different pattern of immunolabelling for each antibody, showing the specificity of the response of the tissue to each antibody. Quantification Several counts of the number of glial-immunoreactive cells per unit area (0.25 mm2) were performed to evaluate possible differences between control and kainic acid-treated mice. Three different areas were considered per animal for this purpose.
RESULTS
We did not detect areas of neuronal loss in kainic acid-treated mice after Cresyl Violet staining of the brain sections (Fig. 1A and B). In the control mouse hippocampus, calmodulin immunoreactivity was mainly localized in the pyramidal cells of Ammon’s horn and the granule cells of the dentate gyrus (Fig. 1C). Nevertheless, a large number of calmodulin-immunoreactive glial cells were observed in the hippocampus of kainic acid-treated mice (Figs 1D and 2B). No glial cells of this type were detected in the control hippocampus by calmodulin immunostaining (Figs 1C and 2A). In order to identify the glial cells responsible for calmodulin immunoreactivity, we used lectin as a marker of microglial cells and GFAP antibody as a marker of astrocytes. Lectin-reactive cells (microglial cells) were detected throughout the hippocampus of control mice (Fig. 2C). These cells were constituted by small cell bodies and numerous long, thin, branched processes, all of them characteristic of resting microglial cells. Lectinreactive cells were also observed in the hippocampus of kainic acid-treated mice (Fig. 2D). Nevertheless, the morphology of these cells was substantially different from that observed in the control mice: their processes were much thicker and shorter than in the control microglial cells, characteristic features of activated microglial cells. On the other hand, we observed an increase in the density of microglial cells in the hippocampus of kainic acid-treated mice when compared to the control animals (Table 1). In
Fig. 1. Glial reaction in the hippocampus of kainic acid-treated mice. (A and B) Cresyl Violet staining of the hippocampus in coronal brain sections of a control (A) and a kainic acid-treated mouse (B) showing the absence of areas of neuronal death. (C and D) Calmodulin immunoreactivity in the hippocampus of a control (C) and a kainic acid-treated mouse (D) showing the presence of immunolabelled glial cells throughout the hippocampus of the treated mouse (arrowheads). Scale bar=1 mm. Abbreviations: CA1–CA3, CA1–CA3 fields of Ammon’s horn; DG, dentate gyrus; pyr, pyramidal cell layer of the hippocampus; g, granulle cell layer of the dentate gyrus.
702
C. Sola` et al.
Fig. 2. Immunolocalization of calmodulin in reactive microglial cells of the hippocampus of kainic acid-treated mice. Calmodulin immunoreactivity (A and B), lectin reaction (C and D) and GFAP immunoreactivity (E and F) in the hippocampus of control (left column) and kainic acid-treated mice (right column). Notice the presence of numerous calmodulin-immunoreactive glial cells in the hippocampus of kainic acid-treated mice (B). These immunoreactive cells are shown in more detail in the inset (small arrows). Notice also the clear microglial reaction in the hippocampus of the treated animals (D): the processes of microglial cells (lectin-positive cells) become thicker and shorter than in the control hippocampus (insets in C and D). All these changes occur in the absence of changes in the astroglia (GFAP-positive cells) (E and F). Arrows point to latent microglial cells (LM), small arrows show reactive microglial cells (RM) and arrowheads show astrocytes (A). Scale bars=300 µm (A to F) and 50 µm (A to F, insets).
addition, there were no differences between the number of reactive microglial cells and the number of calmodulin-immunoreactive glial cells in the hippocampus of the treated animals (Table 1).
GFAP-immunoreactive cells (astrocytes) were evenly distributed in the hippocampus of control mice. No differences were observed between control and kainic acid-treated mice (Fig. 2E, F).
Calmodulin expression in reactive microglia Table 1. Density of lectin-labelled and calmodulinimmunoreactive microglial cells (number of microglial cells/ 0.25 mm2) in the hippocampus of control and kainic acidtreated mice Experimental situation Lectin reaction Calmodulin immunoreactivity
Control
Kainic acid
10.83&0.17 Non-detected
57.87&2.38a 64.00&2.14b
Data represent means&S.E.M. (n=3 mice/group; three areas of 0.25 mm2 each were measured per animal). a Statistically significantly different from the control value (Student’s t-test, P<0.001). b No differences were observed between the densities of calmodulin-immunoreactive glial cells and lectin-reactive cells in the hippocampus of kainic acid-treated mice.
DISCUSSION
The systemic administration of a convulsant dose of kainic acid to mice induced a glial reaction in the brain. This glial reaction was mainly constituted by reactive microglial cells, which were mostly detected in the hippocampus of kainic acid-treated mice. Strong immunoreactivity for calmodulin was observed in these reactive microglial cells, while control microglial cells did not show significant calmodulin immunolabelling. It has been widely documented that non-neuronal cells can respond to neuronal cell injury, constituting what is known as glial activation. Astroglial or microglial activation has been observed in the CNS as a response to several experimental situations in which neuronal injury is induced, such as ischaemia,20,29,31 axotomy,4,15,17 electrical stimulation22 or neurotoxic insult.2,3,20,36 Glial activation has also been observed in the human brain in neurodegenerative diseases.7,16,18 We observed a glial reaction in the hippocampus of kainic acid-treated mice 24 h after administration, which appeared to represent the activation of microglial cells, while no astroglial activation was detected. Both astroglial and microglial activations have been reported in the hippocampus of kainic acid-treated rats in the presence of neuronal degeneration.3,20 These authors show that the microglial reaction is an early response to injury while the astroglial activation appears later. Accordingly, the absence of astroglial reaction in our experimental conditions may be because it occurs later. On the other hand, differences in the response of the activated glia have also been observed, depending on the degree of injury attained.17,20,35 Thus, the glial reaction induced in the presence of neuronal degeneration is very different from the glial reaction in cases in which the injury does not result in neuronal death. Indeed, we did not detect areas of neuronal loss either at 24 h of kainic acid administration or at the different times following kainic acid administration considered in a previous study (two, four and eight days).33 Although the alterations
703
in neuronal function that we induced with kainic acid produced microglial activation, they may not have been strong enough to result in astroglial reaction. We also observed an increase in the number of microglial cells in the hippocampus following kainic acid administration. This increase could be due to microglial cell proliferation and/or to the migration of microglial cells from other brain areas. An extracerebral origin of these cells could be ruled out for three reasons: (i) the short delay in the appearance of these cells after kainic acid administration; (ii) the morphology of the reactive microglial cells, which does not resemble that of macrophages; and (iii) the absence of areas of neuronal death that could explain the presence of macrophages arising from the blood vessels. Glial activation is associated with both morphological and phenotypic changes, which reflects either properties of the cells before the injury, or the expression of new properties. Thus, activated microglia have been reported to increase the expression or to express de novo several molecules, such as surface antigens (type 3 complement receptors or CR3, major histocompatibility complex), heat-shock protein 72, cytokines and basic fibroblast growth factor, among others.10,11,24,34,39 This is the first report of an increase in the expression of calmodulin in activated microglia. As calmodulin is involved in many cellular functions it is difficult to establish the reason for this increase. For instance, as calmodulin is involved in the regulation of cytoeskeletal proteins, the increase in calmodulin expression could be related to the morphological changes observed in the activated microglia. On the other hand, if the observed increase in the density of the microglial cells in the hippocampus after kainic acid is due to the proliferation of the microglial cells, the role of calmodulin as key regulator of cell proliferation could also account for the increased calmodulin expression in these cells. Probably more than one factor is responsible for the great increase in the density of microglial cells observed. Additional experiments should be performed to elucidate the possible role of the increased calmodulin expression in the reactive microglial cells. Irrespective of its function, it is clear that calmodulin is strongly expressed in reactive microglia and that calmodulin immunocytochemistry can be used to show the presence of reactive microglial cells in brain sections. Acknowledgements—This work was partially supported by grants SAF 95-0041-CO2-01 from CICYT (Comisio´n de Investigacio´n Cientı´fica y Te´cnica), 95-0206 from FIS (Fondo de Investigaciones Sanitarias) and GR93-8.012 from CIRIT (Comissio´ Interdepartamental de Recerca i Tecnologia). The authors thank Dr Marc Soriano for supplying the Tomato lectin and for thoughtful comments during the realization of the present work. The authors also thank Eduard Bustamante for technical assistance with photographic material.
704
C. Sola` et al. REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37.
Acarı´n L., Vela J. M., Gonza´lez B. and Castellano B. (1994) Demonstration of poly-N-acetyl-lactosamine residues in ameboid and ramified microglial cells in rat brain by tomato lectin binding. J. Histochem. Cytochem. 42, 1033–1041. Acarı´n L., Gonza´lez B., Castellano B. and Castro A. J. (1996) Microglial response to N-methyl--aspartate-mediated excitotoxicity in the immature rat brain. J. comp. Neurol. 367, 361–374. Akiyama H., Tooyama I., Kondo H., Ikeda K., Kimura H., McGeer E. G. and McGeer P. L. (1994) Early response of brain resident microglia to kainic acid-induced hippocampal lesions. Brain Res. 635, 257–268. Barron K. D., Marciano F. F., Amundson R. and Mankes R. (1990) Perineuronal glial responses after axotomy of central and peripheral axons. A comparison. Brain Res. 523, 219–229. Barro´n S., Tusell J. M. and Serratosa J. (1995) Effect of hexachlorocyclohexane isomers on calmodulin mRNA expression in the central nervous system. Molec. Brain Res. 30, 279–286. Cheung W. Y. (1980) Calmodulin plays a pivotal role in cellular regulation. Science 207, 19–27. Delacourte A. (1990) General and dramatic glial reaction in Alzheimer brains. Neurology 40, 33–37. DeLorenzo R. J. (1981) The calmodulin hypothesis of neurotransmission. Cell Calcium 2, 365–385. DeLorenzo R. J. (1982) Calmodulin in neurotransmitter release and synaptic function. Fedn Proc. 41, 2265–2272. Finsen B. R., Jorgensen M. B., Diemer N. H. and Zimmer J. (1993) Microglial MHC antigen expression after ischemic and kainic acid lesions of the adult hippocampus. Glia 7, 41–49. Frautschy S. A., Walicke P. A. and Baird A. (1991) Localization of basic fibroblast growth factor and its mRNA after CNS injury. Brain Res. 553, 291–299. Gannon M. N. and McEwen B. S. (1994) Distribution and regulation of calmodulin mRNAs in rat brain. Molec. Brain Res. 22, 186–192. Gnegy M. E. (1993) Calmodulin in neurotransmitter and hormone action. A. Rev. Pharmac. Toxicol. 32, 45–70. Gnegy M. E., Agrawal A., Hewlett K., Yeung E. and Yee S. (1994) Repeated haloperidol increases both calmodulin and a calmodulin-binding protein in rat striatum. Molec. Brain Res. 27, 195–204. Graeber M. B. and Kreutzberg G. W. (1986) Astrocytes increase in glial fibrillary acidic protein during retrograde changes of facial motor neurons. J. Neurocytol. 15, 363–373. Graeber M. B. and Streit W. J. (1990) Microglia: immune network in the CNS. Brain Path. 1, 2–5. Graeber M. B., Streit W. J. and Kreutzberg G. W. (1990) The glial response to motor neuron axotomy: role of microglia and astrocytes. Proceedings of the XIth International Congress of Neuropathology, pp. 282–287. Haga S., Akai K. and Ishii T. (1989) Demonstration of microglial cells in and around senile (neuritic) plaques in the Alzheimer brain. Acta neuropath. 77, 569–575. Ikeshima H., Yuasa S., Matsuo K., Kawamura K., Hata J. and Takano T. (1993) Expression of three nonallelic genes coding calmodulin exhibits similar localization on the central nervous system of adult rats. J. Neurosci. Res. 36, 111–119. Jorgensen M. B., Finsen B. R., Jensen M. B., Castellano B., Diemer N. H. and Zimmer J. (1993) Microglial and astroglial reactions to ischemic and kainic acid-induced lesions to the adult rat hippocampus. Expl Neurol. 120, 70–88. Landry C. F., Ivy G. O. and Brown I. R. (1992) Effect of a discrete dorsal forebrain lesion in the rat on the expression of neuronal and glial-specific genes: induction of calmodulin, NF-L, SC1, and GFPA mRNA. J. Neurosci. Res. 32, 280–289. Lee-Hall L., Borke R. C. and Anders J. J. (1989) Transection or electrical stimulation of the hypoglossal nerve increases glial fibrillary acidic protein immunoreactivity in the hypoglossal nucleus. Brain Res. 490, 157–161. Lu K. P. and Means A. R. (1993) Regulation of the cell cycle by calcium and calmodulin. Endocr. Rev. 14, 40–58. Morioka T., Kalehua A. N. and Streit W. J. (1992) Progressive expression of immunomolecules on microglial cells in rat dorsal hippocampus following transient forebrain ischemia. Acta neuropath. 83, 149–157. Ni B., Rush S., Gurd J. W. and Brown I. R. (1992) Molecular cloning of calmodulin mRNA species which are preferentially expressed in neurons in the rat brain. Molec. Brain Res. 13, 7–17. Ni B. and Brown I. R. (1993) Modulation of a neuronal calmodulin mRNA species in the rat brain stem by reserpine. Neurochem. Res. 18, 185–192. Nojima H. (1989) Structural organisation of multiple rat calmodulin genes. J. molec. Biol. 208, 269–282. Rasmussen C. D. and Means A. R. (1989) Calmodulin, cell growth and gene expression. Trends Neurosci. 12, 433–438. Rischke R. and Krieglstein J. (1991) Postischemic neuronal damage causes astroglial activation and increase in local cerebral glucose utilization of rat hippocampus. J. cerebr. Blood Flow Metab. 11, 106–113. Roberts-Lewis J. M., Cimino M., Krause R. G. II, Tyrrell D. F. Jr., Davis L. G., Weiss B. and Lewis M. E. (1990) Anatomical localization of calmodulin mRNA in the rat brain with cloned cDNA and synthetic oligonucleotide probes. Synapse 5, 247–254. Sharp F. R., Lowenstein D., Simon R. and Hisanaga K. (1991) Heat shock protein hsp 72 induction in cortical and striatal astrocytes and neurons following infarction. J. cerebr. Blood Flow Metab. 11, 621–627. Sola` C., Tusell J. M. and Serratosa J. (1996) Comparative study of the pattern of expression of calmodulin messenger RNAs in the mouse brain. Neuroscience 75, 245–256. Sola` C., Tusell J. M. and Serratosa J. (1997) Differential response of calmodulin genes in the mouse brain after systemic kainic acid administration. Neuroscience 78, 155–164. Soriano M., Planas A. M., Rodrı´guez-Farre´ E. and Ferrer I. (1994) Early 72-kDa heat shock protein induction in microglial cells following focal ischemia in the rat brain. Neurosci. Lett. 182, 205–207. Streit W. J. (1993) Microglial–neuronal interactions. J. chem. Neuroanat. 6, 261–266. van den Berg K. J. and Gramsbergen J. B. P. (1993) Long-term changes in glial fibrillary acidic protein and calcium levels in rat hippocampus after a single systemic dose of kainic acid. Ann. N. Y. Acad. Sci. 679, 394–401. Vela J. M., Dalmau I., Gonza´lez B. and Castellano B. (1996) The microglial reaction in spinal cords of jimpy mice is related to apoptotic oligodendrocytes. Brain Res. 712, 134–142.
Calmodulin expression in reactive microglia 38.
705
Vendrell M., Aligue´ R., Bachs O. and Serratosa J. (1991) Presence of calmodulin and calmodulin-binding proteins in the nuclei of brain cells. J. Neurochem. 57, 622–628. 39. Wießner C., Gehrmann J., Lindholm D., To¨pper R., Kreutzberg G. W. and Hossman K. A. (1993) Expression of transforming growth factor-â1 and interleukin-1â mRNA in rat brain following transient forebrain ischemia. Acta neuropath. 86, 439–446. (Accepted 6 May 1997)