Oligodendrocyte killing by quinolinic acid in vitro

Brain Research 896 (2001) 157–160 www.elsevier.com / locate / bres

Short communication

Oligodendrocyte killing by quinolinic acid in vitro Wendy Cammer* Department of Neurology, Albert Einstein College of Medicine, 1300 Morris Park Avenue, F-140, Bronx, NY 10461, USA Accepted 19 December 2000

Abstract Quinolinic acid, which is produced by macrophages and microglia, can kill neurons in vivo and in vitro. To test whether quinolinic acid is toxic to oligodendrocytes, glial cells cultured from the brains of 2-day-old rats were incubated with quinolinic acid at concentrations known to kill neurons. The cells were then fixed and immunostained with MAbO4 to mark immature and mature oligodendrocytes and anti-myelin basic protein (MBP) to mark mature oligodendrocytes. The data indicated up to 54% reductions in the numbers of O4-positive cells in cultures after incubation with quinolinic acid. Apoptosis of O4-positive cells began during the first 6 h, and some of the apoptotic cells became fragmented. Further apoptosis, and clumping of dead MBP-positive oligodendrocytes, occurred during longer incubation with quinolinic acid. Thus, quinolinic acid arising from macrophages and microglia during autoimmune disease may take part in a mechanism of oligodendrocyte injury and killing.  2001 Elsevier Science B.V. All rights reserved. Theme: Disorders of the nervous system Topic: Neurotoxicity Keywords: Oligodendrocyte; Glial cell; Microglia; Quinolinic acid

Quinolinic acid, which is generated from tryptophan via the kynurenic acid pathway, can kill neurons in vivo and in vitro [24,25]. Activated monocytes and macrophages produce quinolinic acid in vitro (e.g. Ref. [22]), and human microglia are known to convert L-tryptophan to quinolinic acid [7,12]. Quinolinic acid is elevated in the CNS of patients with inflammatory conditions resulting from physical trauma or autoimmune neurological disease [13] and in the spinal cords of rats with experimental autoimmune encephalomyelinitis (EAE), as compared to normal spinal cords [8]. The concentration of quinolinic acid in human brain during inflammatory neurological disease is |5 mM [13], which could be many-fold higher locally in specific lesions [25]. To our knowledge the effects of quinolinic acid on oligodendrocytes have not been reported, and we suggest that high concentrations could be generated locally by microglia and macrophages in close proximity to oligodendrocytes in vivo, particularly after the breakdown of the blood–brain-barrier that takes place in multiple sclerosis (MS) and EAE. To test whether quinolinic acid is toxic to *Tel.: 11-718-430-2013; fax: 11-718-430-8790. E-mail address: [email protected] (W. Cammer).

oligodendrocytes in vitro, glial cells cultured from the brains dissected from anesthetized 2-day-old Sprague– Dawley rats [4–6] were incubated with quinolinic acid for 6, 24 or 48 h at various days in vitro (DIV) and then fixed with 4% paraformaldehyde and immunostained with MAbO4 to mark immature and mature oligodendrocytes and anti-myelin basic protein (MBP) to mark mature oligodendrocytes [1,4]. The concentrations and incubation times were consistent with the range of concentrations and incubation times used previously to damage neurons in slices or tissue cultures, e.g. perturbation of gene expression by 1 mM quinolinic acid for 6 h [24]; boosting of lipid peroxidation by 0.08 mM for 30 min [20]; depression of the neuronal population spike after 0.5–0.8 mM for 2 min [26]; neurotoxicity of cortical cell cultures, ED 50 of 2.0 mM after 2 min [15]; necrosis of granule cells at 10 mM [10]; and neurotoxicity at ED 50 of 0.25–0.40 mM for 24 h [23]. O4-positive cells and MBP-positive cells were counted in eight fields on three coverslips under each experimental condition. The data in Table 1 show up to 54% reductions in the numbers of O4-positive cells in cultures at $7 DIV after 48 h of treatment with 100 mM–1 mM quinolinic acid.

0006-8993 / 01 / $ – see front matter  2001 Elsevier Science B.V. All rights reserved. PII: S0006-8993( 01 )02017-0

W. Cammer / Brain Research 896 (2001) 157 – 160

158

Table 1 Effects of quinolinic acid on numbers of O4-positive cells a DIV

6 7 7 8 13

Time (h)

Control

48 24 48 48 48

3968.3 5367.1 58612 6868.0 74613

10 times cells per field 0.1 mM quin. 3667.4 (n.s.) 4565.3 (n.s.) 4665.2 (,0.05) 4968.4 (,0.01) 5166.4 (,0.002)

% Reduction vs. control

21 28 31

1.0 mM quin. 3567.6 (n.s.) 3867.1 (,0.01) 2968.3 (,0.002) 3166.4 (,0.002) 4067.6 (,0.002)

% Reduction vs. control 28 50 54 46

a

Cultures were incubated with quinolinic acid for 24 or 48 h (second column) and then fixed with 4% paraformaldehyde. ‘DIV’ indicates the day of fixation. The data represent three separate culture preparations, each yielding three coverslips under each condition, and eight fields were counted under each condition. Each value was compared to the control value at the top of the respective column, and the differences and % changes were calculated. P values (in parentheses) were calculated (by the two-tailed t-test), to represent probabilities that the differences from control values could be due to chance. n.s., indicates that the difference from the control value is not statistically significant.

Apoptosis of O4-positive cells began during the first 6 h of incubation with quinolinic acid (Fig. 1, panels A–D), and some of the apoptotic cells became fragmented (panels E and F) [18]. Further apoptosis, and clumping of dead

cells, occurred during longer incubation with quinolinic acid (Fig. 2). The toxicity of quinolinic acid toward oligodendrocytes is surprising only if one accepts two assumptions, which

Fig. 1. Apoptotic oligodendrocytes after 6-h treatment with quinolinic acid. Quinolinic acid (1 mM) was added to glial-cell cultures at 7 DIV or 14 DIV, and the cells were permitted to grow for 6 h, after which they were fixed with 4% paraformaldehyde and refrigerated for 0 to 3 days. The cells were immunostained with MAbO4 or anti-MBP, as described [4], with rhodamine-conjugated goat anti-mouse IgM or rhodamine-conjugated goat anti-mouse IgG1 (Southern Biotechnology, Birmingham, AL), respectively, as second antibodies, and double-stained by the tunel method to detect apoptotic cells, using the method specified with the In Situ Cell Death Detection Kit with fluorescein (from Enzo / Boehringer-Mannheim, Indianapolis, IN). Panel A, 7 DIV, control, O4; panel B, same field as A, tunel; panel C, 7 DIV, 1 mM quinolinic acid, O4; panel D, same field as C, tunel; panel E, 14 DIV, 1 mM quinolinic acid, MBP, one fragmented and one healthy MBP1 cell; panel F, same field as E, tunel. Arrows point to apoptotic oligodendrocytes. Scale bars550 mm.

W. Cammer / Brain Research 896 (2001) 157 – 160

159

interact with NMDA receptors did inhibit the neurotoxic effects of quinolinic acid and concluded that quinolinic acid could injure cells without interacting with a specific receptor [9]. The data in Table 1 suggest that oligodendrocytes are reproducibly susceptible to quinolinic acid, and Figs. 1 and 2, that cell death occurs via apoptosis. Thus, quinolinic acid may take part in a mechanism of oligodendrocyte injury during autoimmune disease of the CNS. TNF-a appears to be another such endogenous factor (reviewed in Ref. [3]). In view of the heterogeneity in the pathology of MS [17,19], there are probably a number of pathological mechanisms. Fig. 2. Apoptotic oligodendrocytes after 48-h treatment with 1 mM quinolinic acid. Quinolinic acid was added to glial-cell cultures at 5 DIV and the cells were allowed to grow for 2 additional days, after which they were fixed and stained as described in the legend to Fig. 1. Red shows O4 immunostaining, and green shows tunel staining. Panel A, control, O4positive cells. Panel B, quinolinic acid treatment, clump of O4-positive cells with tunel-positive nuclei. Scale bar represents 25 mm.

are: (1) that NMDA-type glutamate receptors are absent from oligodendrocytes, and (2) that quinolinic acid acts exclusively via NMDA receptors. However, there are wellestablished observations that provide exceptions to each of the two assumptions.

Assumption [1. Firstly, an early report claiming NMDA receptors to be absent from oligodendrocytes was based on data obtained with the CG4 cell line [27], as distinguished from primary, secondary or bulk-isolated oligodendrocytes. Glutamate agonists had been shown to inhibit proliferation of glial precursors [16] and to trigger desensitizing currents in CG4 cells [10], but only non-NMDA receptors appeared to be involved in those phenomena [10,16]. Moreover, the activities of NMDA receptors can be difficult to detect in the absence of optimal concentrations of glycine [25,26,29], and oligodendrocyte progenitors may have an atypical receptor for glycine [2], which may confound the binding of conventional NMDA agonists. Moreover, after 1995 some evidence appeared suggesting the presence of NMDA-type receptors on oligodendrocyte progenitors [2,28] and possibly on mature oligodendrocytes as well [14]. These receptors may account for the cell death that we observe with quinolinic acid, if this toxin is, indeed, acting through NMDA receptors in our system. Assumption [2. Secondly, reports from several laboratories indicate that the toxicity of quinolinic acid may occur through alternative mechanisms that do not require NMDA receptors [21]. For example, Schwarcz et al. [23] observed that some agents that did not

Acknowledgements We thank the staff of the Analytical Imaging Facility at the Albert Einstein College of Medicine for assistance in obtaining Fig. 2. This work was supported by grant RG2791-A-1 from the National Multiple Sclerosis Society.

References [1] R. Bansal, A.E. Warrington, A.L. Gard, B. Ranscht, S.E. Pfeiffer, Multiple and novel specificities of monoclonal antibodies O1, O4 and RMAb used in the analysis of oligodendrocyte development, J. Neurosci. Res. 24 (1989) 548–557. [2] S. Belachew, B. Malgrange, J.M. Rigo, B. Rogister, P. Couke, C. Mazy-Servais, G. Moonen, Developmental regulation of neuroligand-induced responses in cultured oligodendroglia, Neuroreport 9 (1998) 973–980. [3] W. Cammer, Effects of TNF-a on immature and mature oligodendrocytes and their progenitors in vitro, Brain Res. 864 (2000) 213–219. [4] W. Cammer, H. Zhang, Maturation of oligodendrocytes is more sensitive to TNF-a than is survival of precursors and immature oligodendrocytes, J. Neuroimmunol. 97 (1999) 37–42. [5] F.-C. Chiu, J.E. Goldman, Synthesis and turnover of cytoskeletal proteins in cultured astrocytes, J. Neurochem. 42 (1984) 166–174. [6] R. Cole, J. DeVellis, Preparation of astrocyte and oligodendrocyte cultures from primary rat glial cultures, in: A. Shahar, J. DeVellis, A. Vernadakis, B. Haber (Eds.), A Dissection and Tissue Culture Manual of the Nervous System, Alan R. Liss, New York, 1989, pp. 121–133. [7] M.C. Espey, O.N. Chernyshev, J.F. Reinhard, M.A. Namboodiri, C.A. Colton, Activated human microglia produce the excitotoxin quinolinic acid, Neuroreport 8 (1997) 431–434. [8] E.M. Flanagan, J.B. Erikson, O.H. Viveros, S.Y. Chang, J. Reinhard, Neurotoxin quinolinic acid is selectively elevated in spinal cords of rats with experimental allergic encephalomyelitis, J. Neurochem. 64 (1995) 1192–1196. [9] A.C. Foster, J.F. Collins, R. Schwarcz, On the excitotoxic properties of quinolinic acid, 2,3-piperidine dicarboxylic acids and structurally related compounds, Neuropharmacology 22 (1983) 1331–1342. [10] V. Gallo, D.K. Patneau, M.L. Mayer, F.M. Vaccarino, Excitatory amino acid receptors in glial progenitor cells: molecular and functional properties, Glia 11 (1994) 94–101. [12] M.P. Heyes, C.L. Achim, C.A. Wiley, E.O. Major, K. Saito, S.P. Markey, Human microglia convert L-tryptophan into the neurotoxin quinolinic acid, Biochem. J. 320 (1996) 595–597.

160

W. Cammer / Brain Research 896 (2001) 157 – 160

[13] M.P. Heyes, K. Saito, J.S. Crowley et al., Quinolinic acid and kynurenine pathway metabolism in inflammatory and non-inflammatory neurological disease, Brain 115 (1992) 1249–1273. [14] M.V. Johnston, Neurotransmitters and vulnerability of the developing brain, Brain Dev. 17 (1995) 301–306. [15] J.P. Kim, D.W. Choi, Quinolinate neurotoxicity in cortical cell culture, Neuroscience 23 (1987) 423–432. [16] H.-N. Liu, G. Almazan, Glutamate induces c-fos proto-oncogene expression and inhibits proliferation in oligodendrocyte progenitors: receptor characterization, Eur. J. Neurosci. 7 (1995) 2355–2363. [17] C.F. Lucchinetti, W. Bruck, M. Rodriguez, H. Lassmann, Distinct patterns of multiple sclerosis pathology indicate heterogeneity of pathogenesis, Brain Pathol. 6 (1996) 259–274. [18] M.P. Mattson, Apoptotic and anti-apoptotic synaptic signaling mechanisms, Brain Pathol. 10 (2000) 300–312. [19] K. Ozawa, G. Suchanek, H. Breitschopf, W. Bruck, H. Budka, K. Jellinger, H. Lassmann, Patterns of oligodendrocyte pathology in multiple sclerosis, Brain 117 (1994) 1311–1322. [20] C. Rios, A. Santamaria, Quinolinic acid is a potent lipid peroxidant in rat brain homogenates, Neurochem. Res. 16 (1991) 1139–1143. [21] E. Rodriquez-Martinez, A. Camancho, P.D. Maldonado, J. PedrazaChaverri, D. Santamaria, S. Galvan-Arzate, A. Santamaria, Effect of quinolinic acid on endogenous antioxidants in rat corpus striatum, Brain Res. 858 (2000) 436–439. [22] K. Saito, C.Y. Chen, M. Masana, J.S. Crowley, S.P. Markey, M.P. Heyes, 4-chloro-3-hydroxyanthranilate, 6-chlorotryptophan and nor-

[23] [24]

[25] [26]

[27]

[28]

[29]

harmane attenuate quinolinic acid formation by interferon-g-stimulated monocytes (THP-1 cells), Biochem. J. 291 (1993) 11–14. R. Schwarcz, A.C. Foster, E.D. French, W.O. Whetsell, Excitotoxic models for neurodegenerative disorders, Life Sci. 35 (1987) 19–32. B. Seidel, G. Keilhoff, T. Reinheckel, G. Wolf, Differentially expressed genes in hippocampal cell cultures in response to an excitotoxic insult by quinolinic acid, Brain Res. Mol. Brain Res. 60 (1998) 296–300. T.W. Stone, Neuropharmacology of quinolinic and kynurenic acids, Pharmacol. Rev. 45 (1993) 309–379. T.W. Stone, Differences of neuronal sensitivity to amino acids and related compounds in the rat hippocampal slice, Neurosci. Lett. 59 (1985) 313–317. A. Yoshioka, M. Hardy, D.P. Younkin, J.B. Grinspan, J.L. Stern, D. Pleasure, Alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionate (AMPA) receptors mediate excitotoxicity in the oligodendroglial lineage, J. Neurochem. 64 (1995) 2442–2448. C. Wang, W.F. Pralong, M.F. Schulz, G. Rougan, J.M. Aubry, S. Pagliusi, J.Z. Kiss, Functional N-methyl-D-aspartate receptors in O-2A glial precursor cells: critical role in regulating polysialic acid-neural cell adhesion molecule expression and cell migration, J. Cell Biol. 135 (1996) 1565–1581. S.G. Zhu, E.G. McGeer, P.L. McGeer, Effect of MK-801, kynurenate, glycine, dextrorphan and 4-acetylpyridine on striatal toxicity of quinolinate, Brain Res. 481 (1989) 356–360.