Brain Research, 516 (1990) 20-30 Elsevier
20 BRES 15431
Effects of methotrexate on astrocytes in primary culture: light and electron microscopic studies* Jocelyn B. Gregorios and David Soucy Department of Pathology (Neuropathology Section), University of Florida School of Medicine, Gainesville, FL 32605 (U.S.A.) (Accepted 10 October 1989)
Key words: Astrocyte; Cell culture; Encephalopathy; Methotrexate, Neurotoxicity
Recent studies on the animal model suggest that astrocytes may be a primary target for methotrexate (MTX) toxicity. To establish whether the astroglial alterations are due to a direct toxic effect of the drug, we studied the morphologic alterations, mitotic index, viability and growth rate of astrocytes in primary culture after exposure to varying concentrations of MTX in the absence or presence of dibutyryl cyclic AMP (dBcAMP). Dense bodies and cellular debris were noted by light and electron microscopy, and became more prominent with increasing doses and greater frequency of treatment. Degenerating cells and areas of necrosis were seen at higher concentrations. These changes became less conspicuous when MTX was given concurrently with dBcAMP. Large reactive-like astrocytes were also seen after MTX administration both in the absence or presence of dBcAMP. Mitotic rate inhibition was noted at all concentrations but was not dose-related. Cell viability was reduced and remained low up to 48 h after withdrawal of MTX and correlated well with drug concentration, although growth rate did not vary significantly from the control. Our findings show that pure populations of astrocytes can be adversely affected by MTX especially in the absence of bBcAMP, while also causing reactive-like changes in some cells. This report provides further evidence that astrocytes may be a primary target for MTX toxicity and suggests that the gliosis seen in MTX encephalopathy may in part be related to MTX-induced astrocytic injury. INTRODUCTION M e t h o t r e x a t e ( M T X ) , an antifolate drug, is widely used in the c h e m o t h e r a p y of metastatic and primary brain tumors. H o w e v e r , its efficacy is limited by a relatively high incidence of central nervous system (CNS) complications 1'22'23. A c u t e neurotoxicity generally occurs within hours or days after t r e a t m e n t and is characterized by lethargy, confusion, seizures or signs of increased intracranial pressure. Subacute neurotoxicity develops within days or weeks after therapy and manifests with somnolence and m o t o r dysfunction. D e l a y e d neurotoxicity occurs months o r years after M T X treatment, either alone or in combination with radiation, and may lead to c o m a and death. The pathologic correlates of acute m e t h o t r e x a t e e n c e p h a l o p a t h y are not clear since the symptoms are generally transient and most of the patients recover following withdrawal of the drug. In subacute and chronic cases, the pathologic changes are variable and may consist either of degenerative gray and white m a t t e r disease 8'33 or a disseminated necrotizing l e u k o e n c e p h a l o p a t h y 35. The pathogenetic mechanism and precise cellular target(s) of M T X toxicity are still not
established. A striking astroglial proliferation is generally seen even in the absence of vascular changes, neuron damage or myelin loss, suggesting that astrocytes may play a role in the CNS dysfunction 18. We have recently r e p o r t e d our morphological studies showing a dose-related selective involvement of astrocytes in rat brain following acute systemic and intraventricular injection of M T X 14. The oligodendrocytes, myelin, neuron and endothelial cells were relatively spared. White m a t t e r necrosis was also seen following intraventricular injection of high-dose MTX. T h e result of our in vivo studies suggested that, at least morphologically, the astrocytes might be a target for M T X toxicity. O n the other hand, we could not c o m p l e t e l y exclude the possibility that the astroglial alterations o b s e r v e d in the animal model might reflect a non-specific reaction to submicroscopic M T X - i n d u c e d injury to o t h e r c o m p o n e n t s of brain. We concluded that e x p e r i m e n t a l models using a relatively pure p o p u l a t i o n of astrocytes i n d e p e n d e n t of other CNS cellular c o m p o n e n t s still n e e d e d to be investigated in o r d e r to establish w h e t h e r the astrocytic changes o b s e r v e d in vivo were due to a direct toxic effect of MTX.
* Presented in part at the 2nd Annual Meeting of the Society for Experimental Neuropathology, New Orleans, September 23-24, 1989. Correspondence: J.B. Gregorios. Present address: University of Miami School of Medicine, Department of Pathology, P.O. Box 016960, Miami, FL 33101, U.S.A. 0006-8993/90/$03.50 t~) 1990 Elsevier Science Publishers B.V. (Biomedical Division)
21
Fig. 1. A: control culture without dBcAMP showing a monolayered sheet of polyhedral cells with regular nuclei, indistinct cytoplasmic borders and fine cytoplasmic processes. B: control culture with dBcAMP showing mostly stellate cells with well-developed processes, x900.
This report now describes the morphological correlates on a relatively pure population of astrocytes in primary culture after MTX administration. Additionally, the effects of MTX on mitotic activity, cell viability and proliferation of astrocytes were investigated. MATERIALS AND METHODS Primary astrocyte cultures were prepared from two- or threeday-old Sprague-Dawley rats using a modification of the method described by Booher and Seusenbrenner5. Briefly, the brains were stripped of leptomeninges and cerebral cortices were mechanically dissociated by mincing and trituration. The tissue was further disaggregated by vortexing and passed through 70 and 20/~m pore size sterile nylon sieves. Approximately 5 x l 0 s trypan blueexcluding cells were seeded onto polylysine-coated coverslips or directly into 60-ram Falcon tissue culture dishes with Dulbecco's modified Eagle's medium (DMEM) containing 10% fetal calf serum (FCS), penicillin (100 units/ml), streptomycin (100 /~g/ml) and fungizone (2.5/tg/ml). The cultures were incubated at 37 °C in a humidified chamber containing 5% CO 2 and 95% air. The cultures were fed every 2-3 days with DMEM containing 10% FCS and examined periodically by phase microscopy. Immunoperoxidase stains for glial fibrillary acidic protein (GFAP) were routinely performed and showed that up to 90% of the cell population consisted of astrocytes. The tests for mycoplasma contamination of cultures were negative. For light microscopic studies, two-week-old cultures grown on coverslips were given one, two or three daily doses of MTX at clinically relevant dosages (0.001-1.0 mM). Because MTX exerts its greatest effect on replicating cells, the drug was added to the cultures in the presence or absence of 0.5 mM dibutyryl cyclic AMP
(dBcAMP), an agent which is known to inhibit proliferation and induce differentiation of astrocytes in cultureTM. Twenty-four hours after the last dose, experimental and corresponding control cultures were fixed with 10% phosphate-buffered formalin and stained with May-Grunwald/Giemsa. For ultrastructural studies, the cells were fixed with phosphate-buffered 4% formaldehyde-l% glutaraidehyde (pH 7.4) for at least two hours, post-fixed in 1% osmium tetroxide for one hour, and dehydrated with increasing concentrations of alcohol. Gelatin capsules containing Epon 812 were inverted onto the plates and incubated at 37 °C for 24 h and 60 °C for another 48 h. The Epon-embedded cultures were then cut on edge using a Sorvall MT2-B ultra-microtome, stained with lead-uranyl acetate, and examined with a Zeiss 109 electron microscope. Mitotic index was calculated by counting the number of mitosis in 1000 high power fields (average 20 cells/lfigh power field), and expressing the value in experimental cultures as a percentage ratio compared to their corresponding controls. To determine the effect of MTX on viability and growth rate, two-week-old cultures grown in tissue culture dishes were treated with a single dose of MTX (0.1, 0.5 or 1.0 mM) which was added to DMEM containing 2% FCS. After 24 h of incubation at 37 °C in a humidified CO 2 chamber, single cell suspensions were obtained by further incubating the cultures in 0.05% trypsin-EDTA for about 15 rain. For viability studies, an aliquot of the cells in suspension was stained with 0.4% Trypan blue and counted in a hemocytometer. The percentage ratio of viable (trypan blue-excluding) cells compared to total cell count was calculated at 24 h after MTX exposure and at various times after withdrawal of the drug. For subculture and growth rate analysis, the remaining cell suspension was correspondingly diluted with DMEM plus 10% FCS in order to achieve a uniform cell count per sample. Approximately 104 viable cells were then seeded on to 24-well plates, and were incubated at 37 °C in a humidified CO 2 chamber for 1-12 days. The growth curve was obtained by plotting the total cell count on a log scale against time on a linear scale.
22
Fig. 2. Cultures treated with MTX in the absence of dBcAMP. A: cytoplasmic dense bodies and extracellular debris are noted with a single dose of 0.01 mM MTX. B: more prominent debris, numerous dense bodies and cytoplasmic vacuoles are seen with a single dose of 0.1 mM MTX. The nuclei are variable in size. C: after two daily doses of 0.1 mM MTX, several degenerating cells are present. D: in some areas, greater nuclear pleomorphism is noted and several cells appear to have more abundant cytoplasm (0.1 mM MTX, two doses), x900.
RESULTS
Light microscopy C o n t r o l cultures w i t h o u t d B c A M P consisted m o s t l y of
p o l y h e d r a l cells with indistinct c y t o p l a s m i c p r o c e s s e s and a b u n d a n t a m p h o p h i l i c c y t o p l a s m c o n t a i n i n g a few per i p h e r a l b a s o p h i l i c g r a n u l e s (Fig. 1A). Small c y t o p l a s m i c v a c u o l e s w e r e s o m e t i m e s n o t e d in p e r i n u c l e a r regions.
23
Fig. 3. Cultures exposed to MTX in the presence of dBcAMP. A: many cells appear enlarged with abundant cytoplasm and more prominent processes (0.1 mM MTX, single dose). B: steUate astrocytes contain thickened branching processes (0.5 mM MTX, single dose). C: a degenerating astrocyte and a hypertrophic cell with pleomorphic nucleus is present after exposure to a single dose of 1.0 mM MTX. x900.
Occasional scattered cellular debris were also present. T h e nuclei were relatively uniform in size and shape, had fine evenly distributed chromatin, and contained one or two p r o m i n e n t nucleoli. M a n y mitotic figures were present. A d d i t i o n of d B c A M P to control cultures caused
a significant decrease in the n u m b e r of mitoses and e n h a n c e d the formation of stellate astrocytes which comprised the m a i o r i t y of the cell p o p u l a t i o n (Fig. 1B). Cytoplasmic vacuoles were less n u m e r o u s c o m p a r e d to the n o n - d B c A M P controls.
24
Fig. 4. A: control cell without dBcAMP contains small bundles of intermediate filaments, mitochondria and small polyribosomal dusters. x12,000. B: control cell with dBcAMP contains moderate amount of intermediate filaments and rough endoplasmic reticulum, x 10,000. Cells that were exposed to 0.001 mM MTX in the absence of dBcAMP were essentially similar to nondBcAMP controls. Variably-sized spherical basophilic dense bodies were noted following the addition of a single dose of 0.01 mM MTX (Fig. 2A). These were seen in the perikaryon as well as in the cytoplasmic processes and became more prominent with increasing doses and greater frequency of treatment (Fig. 2B). The amount of extracellular debris also progressively increased in relation to drug dose. Randomly scattered degenerating cells with dense granular cytoplasm and pyknotic nuclei were seen after administration of 0.1 mM MTX (Fig. 2C) while occasional small zones of frank necrosis were noted with greater concentrations (0.5 and 1.0 mM). Mitotic figures became markedly reduced in number. The nuclei varied in size and sometimes were pleomorphic or lobulated. There was a focal increase in the size and number of cytoplasmic vacuoles which, however, did not correlate with dose and frequency of drug administration. Exposure of cultures to MTX at varying concentrations (0.01-1.0 mM) also resulted in the formation of numerous large polyhedral cells with abundant cytoplasm containing fine basophilic granules in the perinuclear region (Fig. 2D). Cultures exposed to 0.001 and 0.01 mM MTX concurrently with dBcAMP were difficult to distinguish from controls. Following addition of 0.1 mM MTX, many of the astrocytes appeared hypertrophic with abundant
cytoplasm containing perinuclear basophilic granulations (Fig. 3A). In other areas, stellate astrocytes with thick branching processes were present (Fig. 3B). The cytoplasmic dense bodies and cellular debris became more prominent only after addition of MTX in concentrations greater than 0.1 mM (Fig. 3C), and did not further increase with the duration and number of treatment. Degenerating cells were seen only after administration of 1.0 mM MTX, while areas of necrosis were not noted even with repeated dosages.
Electron microscopy Control cultures without dBcAMP consisted mostly of polyhedral cells with prominent cytoplasm containing a moderate amount of polyribosomes, small bundles of intermediate filaments, few mitochondria, Golgi bodies and short rough endoplasmic reticulum (RER) (Fig. 4A). The nuclei were round or oval in shape and had fine evenly dispersed chromatin. Several small electron-dense granular or lamellated bodies and occasional small membrane-bound vacuoles were seen within the cytoplasm of some cells. Control cultures with dBcAMP were similar to non-dBcAMP controls except for more prominent cytoplasmic processes and a greater amount of intermediate glial filaments (Fig. 4B). Cytoplasmic dense bodies and membrane-bound vacuoles were only occasionally noted. When a single dose of 0.01 mM MTX was added in the absence of dBcAMP, there was an apparent increase in number of dilated Golgi complexes, smooth endoplasmic
25
Fig. 5. Cultures treated with MTX in the absence of dBcAMP. A: several small lamellated electron dense bodies and prominent golgi bodies (arrow) are seen with single dose of 0.01 mM MTX. B: membrane-bound vacuoles, multivesicular and lamellated dense bodies are increased after 3 doses of 0.01 mM MTX. C: after a single dose of 0.5 mM MTX, the myelinoid dense bodies and cytoplasmic vacuoles are larger. D: large polyribosomal aggregates and prominent bundles of intermediate filaments are present within hypertrophic astrocyte (0.5 mM MTX, two doses), x 10,000. reticulum and small m e m b r a n e - b o u n d vacuoles (Fig. 5A). M a n y cytoplasmic dense bodies were present after two a n d t h r e e doses (Fig. 5B). S o m e cells contained thicker bundles of i n t e r m e d i a t e filaments which were
irregularly scattered within the cytoplasm. The mitochondria r e m a i n e d normal. T h e r e was no significant increase in n u m b e r of R E R although s m o o t h e n d o p l a s m i c reticulum b e c a m e m o r e p r o m i n e n t . Progressive accumulation
26
Fig. 6. Cultures treated with MTX in the presence of dBcAMP. A: distended Golgi bodies adjacent to a centriole and small vesicles are present after a single dose of 0.01 mM MTX. B: several membrane-bound lamellated and granular dense bodies are present in addition to dilated Golgi bodies (0.1 mM MTX, single dose). C: there is only a moderate increase in dense bodies after a single dose of 1.0 mM MTX. D: abundant intermediate filaments are present within the hypertrophic astrocyte (0.1 mM MTX, 3 doses), x 10,000.
of large lamellated and m e m b r a n e - b o u n d dense bodies was noted within the cytoplasm of astrocytes following 1-3 days' exposure to 0.1 m M MTX, and became more n u m e r o u s with higher doses (Fig. 5C). Dense bodies
were also seen within astroglial processes. Large memb r a n e - b o u n d cytoplasmic vacuoles and smooth endoplasmic reticulum were sometimes noted. P r o m i n e n t bundles of intermediate glial filaments were present in the
27 TABLE I
Mitotic index (expressed in per cent)
Control 0.01 mM MTX 0.05 mM MTX 0.1 mM MTX 0.5 mM MTX 1.0 mM MTX
1 Dose
2 Doses
3 Doses
100.0 52.0 73.9 56.5 76.1 41.3
100.0 9.3 14.4 39.2 48.5 34.0
100.0 27.9 29.1 50.0 77.3 55.2
cytoplasm and processes of the astrocytes. Polyribosomal aggregates also appeared to be focally increased, but did not correlate with MTX dose (Fig. 5D). Experimental cultures which were given 0.01 mM MTX concurrently with dBcAMP also showed many distended Golgi bodies usually surrounding a centriole (Fig. 6A). Addition of MTX at higher doses (0.1 and 0.5 mM) resulted in the formation of dense bodies and variably sized membrane-bound vacuoles, although these were not as numerous as in experimental cultures without dBcAMP (Fig. 6B,C). Large bundles of intermediate glial filaments filled the cytoplasm and processes of many cells
100
80
and were usually more abundant in the astrocytes that were less degenerated (Fig. 6D). Cultures that were given 1.0 mM MTX with dBcAMP showed degenerative changes in some cells while cellular hypertrophy with an apparent increase in intermediate filaments was noted in others.
Effects on mitosis, viability and growth Mitotic rate was reduced by up to 50% in the experimental cultures after 24 h of exposure to a single dose of MTX. The inhibitory effect of MTX on mitosis was most prominent after two doses and persisted after 3 doses. However, the extent of mitotic inhibition did not appear to directly correlate with drug concentration or frequency of drug treatment (Table I). Twenty-four hours after a single treatment of astrocytes with MTX, there was a 10-20% loss of cell viability in the experimental compared to the control cultures. The decrease in cell viability was dose-related and statistically significant (P < 0.01), persisting for 24 and 48 h after drug withdrawal and subculture. Viability was partially restored after 4 days and approximated that of the control cultures after 7 days (Fig. 7A). Despite this persistent loss of viability following administration of a single dose of MTX, there was no statistically significant difference between the growth rate of control and MTX-treated astrocytes at various time intervals (1-12 days) after subculture (Fig. 7B).
..I
|
60
> I,Z
DISCUSSION
40
uJ a.
20
0
I
I
0 hour
I
I
I
24 hours 48 hours 4 days
I
12 days
7 days
TIME AFTER WITHDRAWAL/SUBCULTURE •- I - - Control
--O- O,1 i11M MTX
,-o- 0 . 5 m M M T X
-o-
1.0mMM'rx
A
3211 Log 2
Increment
4.0 3.6
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
2.8 2.4 2,0
1.6 1.2
0.8 0.4 0.0 0 hour
24 hours
48, hours
4 days
7 days
12 days
Time after Plating B
- -
CONTROL
"--F--
0.I mM MTX
~
0.5 mM MTX
-"g'-
1.0 rnM MTX
Fig. 7. A: cell viability of astrocytes in culture at various time intervals after withdrawal from MTX (total of 12 determinations, no dBcAMP). B: growth curve of primary astrocyte culture~ after MTX withdrawal (total of 12 determinations, no dBcAMP).
Our studies show that MTX administration can cause morphologic changes in a relatively pure population of astrocytes in primary culture. The formation of cytoplasmic vacuoles, dense bodies, and extracellular debris that were noted by light and electron microscopy are similar to the changes seen in primary astrocyte cultures following administration of other toxic agents 15. The changes are most likely degenerative in nature, appear to be dose-related and are associated with small loci of necrosis at much higher concentrations of MTX. Although non-specific, these degenerative changes occurring in a relatively pure astroglial population provide supporting evidence that astrocytes may, indeed, be a target for MTX neurotoxicity. The mitotic inhibition and loss of viability that was observed in the astrocyte cultures are similar to the reported effects of MTX in other cell types6'11'24'28'39. The mechanism involved in MTX-induced toxicity to the astrocytes is not clear. MTX acts on replicating cells and is believed to exert its chemotherapeutic effect by competitive inhibition of dihydrofolate reductase (DHFR) 4'4°. DHFR catalyzes the formation of tetrahydrofolic acid which is important in DNA synthesisTM. MTX binding and DHFR activity have been demonstrated in primary CNS tumors,
28 chiefly astrocytomas and glioblastomas2. DHFR activity has also been reported in partially purified rabbit and mouse brain extracts, and is totally abolished by MTX 32'38. It is likely that astrocytes in culture also contain DHFR and that MTX may exert its adverse effect on these cells by inhibiting DHFR and DNA synthesis. Although a reduction in mitotic rate was persistently noted in MTX-treated astrocyte cultures after 1-3 administrations of MTX at various concentrations, the degree of mitotic inhibition did not correlate well with the drug dose and frequency of treatment. The reason for this apparent lack of correlation between drug dose and decline in mitotic activity is not clear. In other tissue culture systems, the rate of inhibition of DNA synthesis by MTX is found to be directly proportional to cellular growth rate and is greatest in cells that are in the S-phase of growth 19'2°. As in vivo, subpopulations of astroglial cells in different stages of maturation and growth have been documented to coexist in culture 1°'13'29'34. Since astrocyte cultures generally exhibit an asynchronous growth pattern, it is possible that the degree of MTXinduced mitotic inhibition observed in our cultures may be related to the number of sensitive cells that are in active growth (S) or the mitotic (G2/M) phase at the time of drug administration. On the other hand, subpopulations of cells that are in stationary (Go) phase of growth during the time of treatment may have been relatively resistant to the cytotoxic and/or mitotic-inhibitory effects of MTX. The differential sensitivity to MTX that was observed in our study may also be related to the inherent biological and biochemical properties of various subpopulations of astrocytes in culture 1°'29'3°'34. In addition to mitotic inhibition, a dose-related loss of cell viability was observed in astrocytes, suggesting that MTX may be acting on the cells through more than one mechanism. Studies in C6 rat glioma cells have shown that MTX can inhibit cell division while also altering morphology, stimulating production of S-100 protein and increasing steroid sulfatase activity26'27. MTX-induced DHFR inhibition causes a functional deficiency of reduced folate co-factors that are necessary for the metabolism of methionine, purine and histidine6"13'1'*'22.Likewise, DHFR inhibition may cause an intracellular build up of potentially toxic dihydrofolates1. Thus, MTX may affect morphological differentiation, biochemical functions and certain aspects of cellular metabolism other than DNA synthesis and cell division. It is possible that an MTX-induced decrease in folate co-factors and/or intracellular accumulation of oxidized folates may have caused metabolic perturbations in astrocytes leading to the degenerative changes and loss of viability that were seen in culture. Despite a persistent loss of cell viability, the growth rate in experimental cultures after MTX withdrawal did
not vary significantly from the control. This seeming inconsistency between cytotoxicity and cell growth after drug withdrawal is perhaps also related to the heterogeneity of cells in culture. The reactive-like astroglial hypertrophy and a relative increase in intermediate filaments that were noted in some cells in addition to morphologic evidence of cellular degeneration by light and electron microscopy after MTX treatment serve to support this assumption. To investigate the significance of the apparent astroglial hypertrophy and increased intermediate filaments, the cultures were stained with fluorescein-conjugated antibody to glial fibrillary acidic protein (GFAP) and examined by flow cytometry. Our preliminary findings showed increased GFAP content in 5-10% of the cell population in MTX-treated cultures compared to controls, both in the absence or presence of dBcAMP. This relative increase in GFAP content of treated cells suggests that at least a certain population of astrocytes may be undergoing an adaptive response to MTX since reactive astrocytes are generally characterized by the development of more abundant GFAP 29'31. Likewise, flow cytometric studies of the astrocyte cultures using fluorescein-labelled antibody to incorporated bromodeoxyuridine and propidium iodide showed a 4- to 5-fold increase in percentage of cells in S-phase, with a corresponding reduction in ratio of cells in Go/G1 and G2/M phases as well as a 3- to 4-fold increase in DNA content after 24 h of exposure to MTX (unpublished data). Thus, it is possible that while MTX may have exerted a cytotoxic effect on susceptible subpopulations of astrocytes (as evidenced by light and electron microscopic cellular degeneration as well as the decreased viability of treated cultures), it also allows certain adaptive responses in resistant cells which have not been injured by the drug. Apart from the degenerated cells, the large polygonal astrocytes that were seen following MTX administration in many ways resemble the reactive astrocytes found around sites of CNS injury. Similar changes have been described in culture and are believed to result from hypertrophy of activated astrocytes in response to environmental modification including addition of dBcAMP 3' 9,31. In our study, the hypertrophic astrocytes were found in MTX-treated cultures even in the absence of dBcAMP. Hence, they may represent a glial reaction to MTXinduced metabolic perturbations in culture, and may involve a certain subpopulation of astrocytes that are initially refractory to MTX. Cells that are grown in culture show an asynchronous pattern of growth and are distributed more or less randomly throughout all phases of the cell cycle. It is possible that astrocytes which are in quiescent state of the cell cycle during the period of treatment may be refractory to the deleterious effects of MTX, and may undergo reactive hypertrophy in response
29 tO the MTX-induced degeneration in the more sensitive cells (i.e. in growth phase of cell cycle). This may explain the reactive-like changes that were superimposed on astroglial degeneration following addition of MTX and may be relevant to the astrogliosis that is generally observed in clinical cases of M T X encephalopathy TM. The deleterious effects of M T X were consistently more prominent in the absence of d B c A M P and were less conspicuous when M T X was administered concurrently with dBcAMP. The cause for this apparent protective effect of d B c A M P on MTX-treated cultures can only be speculated on. Studies on Ehrlich ascites tumor cells in culture have shown that d B c A M P depresses net M T X uptake without affecting M T X efflux 4x. On the other hand, d B c A M P - i n d u c e d M T X efflux has also been reported !n isolated hepatocytes ~2. Recent literature also suggests that M T X may share a c o m m o n efflux route with cyclic nucleotides ~7. Cyclic A M P has been shown to affect D H F R gene expression and transcription 42. Likewise, d B c A M P and other agents that increase the intracellular levels of c A M P have been shown to block D H F R synthesis and entry of cells into S phase 16'25. Concurrent administration of dBcAMP to the astrocytes cultures may have inhibited the effect of MTX through any of the above mechanisms. In addition, dBcAMP inhibits proliferation of astrocytes in culture 7'36 and has been shown to arrest other cell systems in G1 phase of cell cycle27. Since REFERENCES 1 Abelson, H.T., Methotrexate and central nervous system toxicity, Cancer Treat. Rep., 62 (1978) 1999-2001. 2 Abelson, H.T., Corka, C., Fosburg, M. and Kornblith, P., Identification of dihydrofolate reductase in human nervous system tumors, Lancet, ii (1978) 184-185. 3 Basset, A.-E., Kalnins, V.I., Ahmed, I. and Fedoroff, S., A 48 kilodalton intermediate filament (IFAP) in reactive-like astrocytes induced by dibutyryl cyclic AMP in culture and in reactive astrocytes in situ, Z Neuropathol. Exp. Neurol., 48 (1989) 245-254. 4 Bertino, J.R., The mechanism of action of the folate antagonists in man, Cancer Res., 23 (1963) 1286-1306. 5 Booher, J. and Sensenhrenner, M., Growth and cultivation of dissociated neurons and glial cells from embryonic chick, rat and human brain in flask cultures, Neurobiology, 2 (1972) 97-105. 6 Borsa, J. and Whitmore, G.E, Cell-killing studies on the mode of action of methotrexate in L-cells in vitro, Cancer Res., 29 (1969) 737-744. 7 Coffino, E and Gray, J.W., Regulation of $49 lymphoma cell growth by cyclic adenosine 3"5"monophosphate, Cancer Res., 38 (1978) 4285-4288. 8 Crosley, C.J., Rorke, L.B., Evans, A. and Nigro, M., Central nervous system lesions in childhood leukemia, Neurology, 28 (1978) 678-675. 9 Fedoroff, S., McAuley, W.A., Houle, J.D. and Devon, R.M., Astrocyte cell lineage. V. Similarity of astrocytes that form in the presence of dBcAMP in cultures to reactive astrocytes in vivo, J. Neurosci. Res., 12 (1984) 14-27. 10 Fedoroff, S., Neal, J., Opus, M. and Kalnins, V.I., Astrocyte cell lineage. III. The morphology of differentiating mouse astrocytes in colony culture, J. Neurocytol., 13 (1984) 1-20.
MTX acts mainly on replicating cells, the dBcAMP-induced arrest in cell cycle may also render the astrocytes more refractory to the deleterious effects of MTX. In summary, our studies indicate that astrocytes in culture can be directly and adversely affected by MTX especially in the absence of dBcAMP, while at the same time also causing reactive-like changes in other cells. Reactive astroglial hyperplasia is a consistent finding in clinical cases of M T X encephalopathy, and is generally believed to be a reaction to MTX-induced injury to the CNS. While the precise cellular target(s) of MTXinduced encephalopathy are still not established, the results of this study, as well as the recently reported MTX-induced morphological alterations in astrocytes in vivo 7, provide supportive evidence that astrocytes may be a direct and primary target of M T X toxicity, and suggests that the astrogliosis seen in M T X encephalopathy may in part be related to MTX-induced astrocytic injury. The mechanism(s) involved in the observed MTX effects on astrocytes remain(s) to be clarified. The use of primary astrocyte cultures as described in this report may serve as a excellent model for further investigation of the pathogenetic mechanism(s) involved in acute M T X encephalopathy.
Acknowledgement. Supported by Research Grant NS 24853 from the National Institutes of Health.
11 Galivan, J., Evidence for cytotoxic activity of polyglutamate derivatives of methotrexate, Mol. Pharmacol., 17 (1980) 105110. 12 Gewirtz, D.A., Randolph, J.K. and Goldman, I.D., Effiux in isolated hepatocytes as a possible correlate of secretion in vivo: induced exit of folic acid analog methotrexate by dibutyryl cyclic AMP or isobutylmethylxanthine, Biochem. Biophys. Res. Commun., 101 (1981) 366-374. 13 Goldman, J.E. and Chiu, E-C., Growth kinetics, cell shape, and cytoskeleton of primary astrocyte cultures, J. Neurochem., 42 (1984) 175-184. 14 Gregorios, J.B., Gregorios, A.B., Mora, J., Marcillo, A., Fojaco, R.M. and Green, B., Morphologic alterations in rat brain following systemic and intraventricular methotrexate injection: light and electron microscopic studies, J. Neuropathol. Exp. Neurol., 48 (1989) 33-47. 15 Gregorios, J.B., Mozes, L.W., Norenberg, L.-O.B. and Norenberg, M.D., The morphologic effects of ammonia on primary astrocyte cultures. I. Light microscopic studies, J. Neuropathol. Exp. Neurol., 44 (1985) 397-403. 16 Gudewicz, T.M., Morhenn, V.B. and Kellems, R.E., The effect of polyoma virus, serum factors and dibutyryl cyclic AMP on dihydrofolate reductase synthesis, and the entry of quiescent cells into S phase, J. Cell Physiol., 108 (1981) 1-8. 17 Henderson, G.B. and Tsuji, J.M., Methotrexate efflux in L1210 cells. Kinetics and specificity properties of the efflux system sensitive to bromosulfophthalein and its possible identity with a system which mediates the efflux of 3"5"-cyclic AMP, J. Biol. Chem., 262 (1987) 13571-13578. 18 Hendin, R., DeVivo, D.S., Torack, R., Lell, M.E., Ragab, A.H. and Vietti, T.J., Parenchymatous degeneration of the central nervous system in childhood leukemia, Cancer, 33 (1974) 468-482.
30 19 Hryniuk, W.M., Fischer, G.A. and Bertino, J.R., S-phase cells of rapidly growing and resting populations. Differences in response to methotrexate, Mol. Pharmacol., 5 (1969) 557-564. 20 Johnson, L.E, Fisherman, C.L. and Abelson, H.T., Resistance of resting 3T6 mouse fibroblasts to methotrexate cytotoxicity, Cancer Res., 38 (1978) 2408-2412. 21 Kamen, B.A., Methotrexate, folate and the brain, Neurotoxicology, 7 (1986) 209-216. 22 Kamen, B.A., Nylen, P.A., Camitta, B.M. and Bertino, J.R., Methotrexate accumulation and folate depletion in cells as a possible mechanism of chronic toxicity to the drug, Br. J. Haematol., 49 (1981) 355-360. 23 Kaplan, R.S. and Wiernik, P.H., Neurotoxicity of antineoplastic drugs, Semin. Oncol., 9 (1982) 103-130. 24 Keefe, D.A., Capizzi, R.L. and Rudnick, S.A., Methotrexate cytotoxicity for L5178 Y/Asn-lymphoblasts: relationship of dose and duration of exposure to tumor cell viability, Cancer Res., 42 (1982) 1641-1645. 25 Kellems, R.E., Morhenn, V.B., Pfendt, E.A., Alt, EW. and Schimke, R.T., Polyoma virus and cyclic AMP-mediated control of dihydrofolate reductase mRNA abundance in methotrexateresistant mouse fibroblasts, J. Biol. Chem., 254 (1979) 309-318. 26 Kolber, A.R., Perumal, A.S., Goldstein, M.N. and Moore, B.W., Drug-induced differentiation of a rat glioma in vitro. II. The expression of S-100, a glial specific protein and steroid suifatase, Brain Research, 143 (1978) 513-520. 27 Leof, E.B., Wharton, W., Okeefe, E. and Pledger, W.J., Elevated intracellular concentrations of cyclic AMP inhibited serum-stimulated, density-arrested BALB/c-3T3 cells in mid-G1, J. Cell Biochem., 19 (1982) 93-103. 28 Li, J.C. and Kaminskas, E., Accumulation of DNA strand breaks and methotrexate cytotoxicity, Biochemistry, 81 (1984) 5694-5698. 29 Manthorpe, M., Adler, R. and Varon, S., Development, reactivity and GFA immunofluorescence of astroglia-containing monolayer cultures from rat brain, J. Neurocytol., 8 (1979) 605-621. 30 McCarthy, K., An autoradiographic analysis of fl-adrenergic receptors in immunocytochemically defined astroglia, J. Pharmacol. Exp. Ther., 226 (1983) 282-290.
31 Miller, R.H., Abney, E.R. and David, S., Is reactive gliosis a property of distinct subpopulation of astrocytes?, J. Neurosci., 6 (1986) 22-29. 32 Pollock, R.J. and Kaufman, S., Dihydrofolate reductase is present in brain, J. Neurochem., 30 (1978) 253-256. 33 Price, R.A. and Jamieson, P.A., The central nervous system in childhood leukemia. II. Subacute leukoencephalopathy, Cancer, 35 (1975) 306-318. 34 Raft, M.C., Abney, E.A., Cohen, J., Lindsay, R. and Noble, M., Two types of astrocytes in cultures of developing rat white matter: differences in morphology, surface gangliosides and growth characteristics, J. Neurosci., 3 (1983) 1289-1300. 35 Rubinstein, L.J., Herman, M.M., Long, T.F. and Wilbur, J.R., Disseminated necrotizing leukoencephalopathy: a complication of treated central nervous system leukemia, Cancer, 35 (1975) 291-305. 36 Sensenbrenner, M., Devilliers, G., Bock, E. and Porte, A., Biochemical and ultrastructural studies of cultures rat astroglial cells. Effect of brain extract and dibutyryl cyclic AMP on glial fibrillary acidic protein and glial filaments, Differentiation, 17 (1980) 51-61. 37 Silbert, S.W. and Goldstein, M.N., Drug-induced differentiation of a rat glioma in vitro, Cancer Res., 32 (1972) 1422-1427. 38 Spector, R., Levy, P. and Abelson, H.T., The development and regional distribution of dihydrofolate reductase in rabbit brain, J. Neurochem., 29 (1977) 919-921. 39 Taylor, I.W., Slowiaczek, P., Francis, P.R. and Tattersall, M.H.N., Biochemical and cell cycle perturbations in methotrexate-treated cells, Mol. Pharmacol., 21 (1982) 204-210. 40 Werkheiser, W.C., The biochemical, cellular and pharmacological action and effects of the folic acid antagonists, Cancer Res., 23 (1963) 1277-1285. 41 White, J.C., Carchman, R.A., Fry, D.-W. and Goldman, I.D., Relationship between membrane transport of methotrexate and endogenous cyclic adenosine 3"5" monophosphate in the Ehrlich ascites tumor, Cancer Res., 40 (1980) 2400-2404. 42 Wu, J.S., Wiedemann, L.M. and Johnson, L.F., Inhibition of dihydrofolate reductase gene expression following serum withdrawal or dBcAMP addition in methotrexate-resistant mouse fibroblasts, Exp. Cell Res., 141 (1982) 159-169.
20 BRES 15431
Effects of methotrexate on astrocytes in primary culture: light and electron microscopic studies* Jocelyn B. Gregorios and David Soucy Department of Pathology (Neuropathology Section), University of Florida School of Medicine, Gainesville, FL 32605 (U.S.A.) (Accepted 10 October 1989)
Key words: Astrocyte; Cell culture; Encephalopathy; Methotrexate, Neurotoxicity
Recent studies on the animal model suggest that astrocytes may be a primary target for methotrexate (MTX) toxicity. To establish whether the astroglial alterations are due to a direct toxic effect of the drug, we studied the morphologic alterations, mitotic index, viability and growth rate of astrocytes in primary culture after exposure to varying concentrations of MTX in the absence or presence of dibutyryl cyclic AMP (dBcAMP). Dense bodies and cellular debris were noted by light and electron microscopy, and became more prominent with increasing doses and greater frequency of treatment. Degenerating cells and areas of necrosis were seen at higher concentrations. These changes became less conspicuous when MTX was given concurrently with dBcAMP. Large reactive-like astrocytes were also seen after MTX administration both in the absence or presence of dBcAMP. Mitotic rate inhibition was noted at all concentrations but was not dose-related. Cell viability was reduced and remained low up to 48 h after withdrawal of MTX and correlated well with drug concentration, although growth rate did not vary significantly from the control. Our findings show that pure populations of astrocytes can be adversely affected by MTX especially in the absence of bBcAMP, while also causing reactive-like changes in some cells. This report provides further evidence that astrocytes may be a primary target for MTX toxicity and suggests that the gliosis seen in MTX encephalopathy may in part be related to MTX-induced astrocytic injury. INTRODUCTION M e t h o t r e x a t e ( M T X ) , an antifolate drug, is widely used in the c h e m o t h e r a p y of metastatic and primary brain tumors. H o w e v e r , its efficacy is limited by a relatively high incidence of central nervous system (CNS) complications 1'22'23. A c u t e neurotoxicity generally occurs within hours or days after t r e a t m e n t and is characterized by lethargy, confusion, seizures or signs of increased intracranial pressure. Subacute neurotoxicity develops within days or weeks after therapy and manifests with somnolence and m o t o r dysfunction. D e l a y e d neurotoxicity occurs months o r years after M T X treatment, either alone or in combination with radiation, and may lead to c o m a and death. The pathologic correlates of acute m e t h o t r e x a t e e n c e p h a l o p a t h y are not clear since the symptoms are generally transient and most of the patients recover following withdrawal of the drug. In subacute and chronic cases, the pathologic changes are variable and may consist either of degenerative gray and white m a t t e r disease 8'33 or a disseminated necrotizing l e u k o e n c e p h a l o p a t h y 35. The pathogenetic mechanism and precise cellular target(s) of M T X toxicity are still not
established. A striking astroglial proliferation is generally seen even in the absence of vascular changes, neuron damage or myelin loss, suggesting that astrocytes may play a role in the CNS dysfunction 18. We have recently r e p o r t e d our morphological studies showing a dose-related selective involvement of astrocytes in rat brain following acute systemic and intraventricular injection of M T X 14. The oligodendrocytes, myelin, neuron and endothelial cells were relatively spared. White m a t t e r necrosis was also seen following intraventricular injection of high-dose MTX. T h e result of our in vivo studies suggested that, at least morphologically, the astrocytes might be a target for M T X toxicity. O n the other hand, we could not c o m p l e t e l y exclude the possibility that the astroglial alterations o b s e r v e d in the animal model might reflect a non-specific reaction to submicroscopic M T X - i n d u c e d injury to o t h e r c o m p o n e n t s of brain. We concluded that e x p e r i m e n t a l models using a relatively pure p o p u l a t i o n of astrocytes i n d e p e n d e n t of other CNS cellular c o m p o n e n t s still n e e d e d to be investigated in o r d e r to establish w h e t h e r the astrocytic changes o b s e r v e d in vivo were due to a direct toxic effect of MTX.
* Presented in part at the 2nd Annual Meeting of the Society for Experimental Neuropathology, New Orleans, September 23-24, 1989. Correspondence: J.B. Gregorios. Present address: University of Miami School of Medicine, Department of Pathology, P.O. Box 016960, Miami, FL 33101, U.S.A. 0006-8993/90/$03.50 t~) 1990 Elsevier Science Publishers B.V. (Biomedical Division)
21
Fig. 1. A: control culture without dBcAMP showing a monolayered sheet of polyhedral cells with regular nuclei, indistinct cytoplasmic borders and fine cytoplasmic processes. B: control culture with dBcAMP showing mostly stellate cells with well-developed processes, x900.
This report now describes the morphological correlates on a relatively pure population of astrocytes in primary culture after MTX administration. Additionally, the effects of MTX on mitotic activity, cell viability and proliferation of astrocytes were investigated. MATERIALS AND METHODS Primary astrocyte cultures were prepared from two- or threeday-old Sprague-Dawley rats using a modification of the method described by Booher and Seusenbrenner5. Briefly, the brains were stripped of leptomeninges and cerebral cortices were mechanically dissociated by mincing and trituration. The tissue was further disaggregated by vortexing and passed through 70 and 20/~m pore size sterile nylon sieves. Approximately 5 x l 0 s trypan blueexcluding cells were seeded onto polylysine-coated coverslips or directly into 60-ram Falcon tissue culture dishes with Dulbecco's modified Eagle's medium (DMEM) containing 10% fetal calf serum (FCS), penicillin (100 units/ml), streptomycin (100 /~g/ml) and fungizone (2.5/tg/ml). The cultures were incubated at 37 °C in a humidified chamber containing 5% CO 2 and 95% air. The cultures were fed every 2-3 days with DMEM containing 10% FCS and examined periodically by phase microscopy. Immunoperoxidase stains for glial fibrillary acidic protein (GFAP) were routinely performed and showed that up to 90% of the cell population consisted of astrocytes. The tests for mycoplasma contamination of cultures were negative. For light microscopic studies, two-week-old cultures grown on coverslips were given one, two or three daily doses of MTX at clinically relevant dosages (0.001-1.0 mM). Because MTX exerts its greatest effect on replicating cells, the drug was added to the cultures in the presence or absence of 0.5 mM dibutyryl cyclic AMP
(dBcAMP), an agent which is known to inhibit proliferation and induce differentiation of astrocytes in cultureTM. Twenty-four hours after the last dose, experimental and corresponding control cultures were fixed with 10% phosphate-buffered formalin and stained with May-Grunwald/Giemsa. For ultrastructural studies, the cells were fixed with phosphate-buffered 4% formaldehyde-l% glutaraidehyde (pH 7.4) for at least two hours, post-fixed in 1% osmium tetroxide for one hour, and dehydrated with increasing concentrations of alcohol. Gelatin capsules containing Epon 812 were inverted onto the plates and incubated at 37 °C for 24 h and 60 °C for another 48 h. The Epon-embedded cultures were then cut on edge using a Sorvall MT2-B ultra-microtome, stained with lead-uranyl acetate, and examined with a Zeiss 109 electron microscope. Mitotic index was calculated by counting the number of mitosis in 1000 high power fields (average 20 cells/lfigh power field), and expressing the value in experimental cultures as a percentage ratio compared to their corresponding controls. To determine the effect of MTX on viability and growth rate, two-week-old cultures grown in tissue culture dishes were treated with a single dose of MTX (0.1, 0.5 or 1.0 mM) which was added to DMEM containing 2% FCS. After 24 h of incubation at 37 °C in a humidified CO 2 chamber, single cell suspensions were obtained by further incubating the cultures in 0.05% trypsin-EDTA for about 15 rain. For viability studies, an aliquot of the cells in suspension was stained with 0.4% Trypan blue and counted in a hemocytometer. The percentage ratio of viable (trypan blue-excluding) cells compared to total cell count was calculated at 24 h after MTX exposure and at various times after withdrawal of the drug. For subculture and growth rate analysis, the remaining cell suspension was correspondingly diluted with DMEM plus 10% FCS in order to achieve a uniform cell count per sample. Approximately 104 viable cells were then seeded on to 24-well plates, and were incubated at 37 °C in a humidified CO 2 chamber for 1-12 days. The growth curve was obtained by plotting the total cell count on a log scale against time on a linear scale.
22
Fig. 2. Cultures treated with MTX in the absence of dBcAMP. A: cytoplasmic dense bodies and extracellular debris are noted with a single dose of 0.01 mM MTX. B: more prominent debris, numerous dense bodies and cytoplasmic vacuoles are seen with a single dose of 0.1 mM MTX. The nuclei are variable in size. C: after two daily doses of 0.1 mM MTX, several degenerating cells are present. D: in some areas, greater nuclear pleomorphism is noted and several cells appear to have more abundant cytoplasm (0.1 mM MTX, two doses), x900.
RESULTS
Light microscopy C o n t r o l cultures w i t h o u t d B c A M P consisted m o s t l y of
p o l y h e d r a l cells with indistinct c y t o p l a s m i c p r o c e s s e s and a b u n d a n t a m p h o p h i l i c c y t o p l a s m c o n t a i n i n g a few per i p h e r a l b a s o p h i l i c g r a n u l e s (Fig. 1A). Small c y t o p l a s m i c v a c u o l e s w e r e s o m e t i m e s n o t e d in p e r i n u c l e a r regions.
23
Fig. 3. Cultures exposed to MTX in the presence of dBcAMP. A: many cells appear enlarged with abundant cytoplasm and more prominent processes (0.1 mM MTX, single dose). B: steUate astrocytes contain thickened branching processes (0.5 mM MTX, single dose). C: a degenerating astrocyte and a hypertrophic cell with pleomorphic nucleus is present after exposure to a single dose of 1.0 mM MTX. x900.
Occasional scattered cellular debris were also present. T h e nuclei were relatively uniform in size and shape, had fine evenly distributed chromatin, and contained one or two p r o m i n e n t nucleoli. M a n y mitotic figures were present. A d d i t i o n of d B c A M P to control cultures caused
a significant decrease in the n u m b e r of mitoses and e n h a n c e d the formation of stellate astrocytes which comprised the m a i o r i t y of the cell p o p u l a t i o n (Fig. 1B). Cytoplasmic vacuoles were less n u m e r o u s c o m p a r e d to the n o n - d B c A M P controls.
24
Fig. 4. A: control cell without dBcAMP contains small bundles of intermediate filaments, mitochondria and small polyribosomal dusters. x12,000. B: control cell with dBcAMP contains moderate amount of intermediate filaments and rough endoplasmic reticulum, x 10,000. Cells that were exposed to 0.001 mM MTX in the absence of dBcAMP were essentially similar to nondBcAMP controls. Variably-sized spherical basophilic dense bodies were noted following the addition of a single dose of 0.01 mM MTX (Fig. 2A). These were seen in the perikaryon as well as in the cytoplasmic processes and became more prominent with increasing doses and greater frequency of treatment (Fig. 2B). The amount of extracellular debris also progressively increased in relation to drug dose. Randomly scattered degenerating cells with dense granular cytoplasm and pyknotic nuclei were seen after administration of 0.1 mM MTX (Fig. 2C) while occasional small zones of frank necrosis were noted with greater concentrations (0.5 and 1.0 mM). Mitotic figures became markedly reduced in number. The nuclei varied in size and sometimes were pleomorphic or lobulated. There was a focal increase in the size and number of cytoplasmic vacuoles which, however, did not correlate with dose and frequency of drug administration. Exposure of cultures to MTX at varying concentrations (0.01-1.0 mM) also resulted in the formation of numerous large polyhedral cells with abundant cytoplasm containing fine basophilic granules in the perinuclear region (Fig. 2D). Cultures exposed to 0.001 and 0.01 mM MTX concurrently with dBcAMP were difficult to distinguish from controls. Following addition of 0.1 mM MTX, many of the astrocytes appeared hypertrophic with abundant
cytoplasm containing perinuclear basophilic granulations (Fig. 3A). In other areas, stellate astrocytes with thick branching processes were present (Fig. 3B). The cytoplasmic dense bodies and cellular debris became more prominent only after addition of MTX in concentrations greater than 0.1 mM (Fig. 3C), and did not further increase with the duration and number of treatment. Degenerating cells were seen only after administration of 1.0 mM MTX, while areas of necrosis were not noted even with repeated dosages.
Electron microscopy Control cultures without dBcAMP consisted mostly of polyhedral cells with prominent cytoplasm containing a moderate amount of polyribosomes, small bundles of intermediate filaments, few mitochondria, Golgi bodies and short rough endoplasmic reticulum (RER) (Fig. 4A). The nuclei were round or oval in shape and had fine evenly dispersed chromatin. Several small electron-dense granular or lamellated bodies and occasional small membrane-bound vacuoles were seen within the cytoplasm of some cells. Control cultures with dBcAMP were similar to non-dBcAMP controls except for more prominent cytoplasmic processes and a greater amount of intermediate glial filaments (Fig. 4B). Cytoplasmic dense bodies and membrane-bound vacuoles were only occasionally noted. When a single dose of 0.01 mM MTX was added in the absence of dBcAMP, there was an apparent increase in number of dilated Golgi complexes, smooth endoplasmic
25
Fig. 5. Cultures treated with MTX in the absence of dBcAMP. A: several small lamellated electron dense bodies and prominent golgi bodies (arrow) are seen with single dose of 0.01 mM MTX. B: membrane-bound vacuoles, multivesicular and lamellated dense bodies are increased after 3 doses of 0.01 mM MTX. C: after a single dose of 0.5 mM MTX, the myelinoid dense bodies and cytoplasmic vacuoles are larger. D: large polyribosomal aggregates and prominent bundles of intermediate filaments are present within hypertrophic astrocyte (0.5 mM MTX, two doses), x 10,000. reticulum and small m e m b r a n e - b o u n d vacuoles (Fig. 5A). M a n y cytoplasmic dense bodies were present after two a n d t h r e e doses (Fig. 5B). S o m e cells contained thicker bundles of i n t e r m e d i a t e filaments which were
irregularly scattered within the cytoplasm. The mitochondria r e m a i n e d normal. T h e r e was no significant increase in n u m b e r of R E R although s m o o t h e n d o p l a s m i c reticulum b e c a m e m o r e p r o m i n e n t . Progressive accumulation
26
Fig. 6. Cultures treated with MTX in the presence of dBcAMP. A: distended Golgi bodies adjacent to a centriole and small vesicles are present after a single dose of 0.01 mM MTX. B: several membrane-bound lamellated and granular dense bodies are present in addition to dilated Golgi bodies (0.1 mM MTX, single dose). C: there is only a moderate increase in dense bodies after a single dose of 1.0 mM MTX. D: abundant intermediate filaments are present within the hypertrophic astrocyte (0.1 mM MTX, 3 doses), x 10,000.
of large lamellated and m e m b r a n e - b o u n d dense bodies was noted within the cytoplasm of astrocytes following 1-3 days' exposure to 0.1 m M MTX, and became more n u m e r o u s with higher doses (Fig. 5C). Dense bodies
were also seen within astroglial processes. Large memb r a n e - b o u n d cytoplasmic vacuoles and smooth endoplasmic reticulum were sometimes noted. P r o m i n e n t bundles of intermediate glial filaments were present in the
27 TABLE I
Mitotic index (expressed in per cent)
Control 0.01 mM MTX 0.05 mM MTX 0.1 mM MTX 0.5 mM MTX 1.0 mM MTX
1 Dose
2 Doses
3 Doses
100.0 52.0 73.9 56.5 76.1 41.3
100.0 9.3 14.4 39.2 48.5 34.0
100.0 27.9 29.1 50.0 77.3 55.2
cytoplasm and processes of the astrocytes. Polyribosomal aggregates also appeared to be focally increased, but did not correlate with MTX dose (Fig. 5D). Experimental cultures which were given 0.01 mM MTX concurrently with dBcAMP also showed many distended Golgi bodies usually surrounding a centriole (Fig. 6A). Addition of MTX at higher doses (0.1 and 0.5 mM) resulted in the formation of dense bodies and variably sized membrane-bound vacuoles, although these were not as numerous as in experimental cultures without dBcAMP (Fig. 6B,C). Large bundles of intermediate glial filaments filled the cytoplasm and processes of many cells
100
80
and were usually more abundant in the astrocytes that were less degenerated (Fig. 6D). Cultures that were given 1.0 mM MTX with dBcAMP showed degenerative changes in some cells while cellular hypertrophy with an apparent increase in intermediate filaments was noted in others.
Effects on mitosis, viability and growth Mitotic rate was reduced by up to 50% in the experimental cultures after 24 h of exposure to a single dose of MTX. The inhibitory effect of MTX on mitosis was most prominent after two doses and persisted after 3 doses. However, the extent of mitotic inhibition did not appear to directly correlate with drug concentration or frequency of drug treatment (Table I). Twenty-four hours after a single treatment of astrocytes with MTX, there was a 10-20% loss of cell viability in the experimental compared to the control cultures. The decrease in cell viability was dose-related and statistically significant (P < 0.01), persisting for 24 and 48 h after drug withdrawal and subculture. Viability was partially restored after 4 days and approximated that of the control cultures after 7 days (Fig. 7A). Despite this persistent loss of viability following administration of a single dose of MTX, there was no statistically significant difference between the growth rate of control and MTX-treated astrocytes at various time intervals (1-12 days) after subculture (Fig. 7B).
..I
|
60
> I,Z
DISCUSSION
40
uJ a.
20
0
I
I
0 hour
I
I
I
24 hours 48 hours 4 days
I
12 days
7 days
TIME AFTER WITHDRAWAL/SUBCULTURE •- I - - Control
--O- O,1 i11M MTX
,-o- 0 . 5 m M M T X
-o-
1.0mMM'rx
A
3211 Log 2
Increment
4.0 3.6
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
2.8 2.4 2,0
1.6 1.2
0.8 0.4 0.0 0 hour
24 hours
48, hours
4 days
7 days
12 days
Time after Plating B
- -
CONTROL
"--F--
0.I mM MTX
~
0.5 mM MTX
-"g'-
1.0 rnM MTX
Fig. 7. A: cell viability of astrocytes in culture at various time intervals after withdrawal from MTX (total of 12 determinations, no dBcAMP). B: growth curve of primary astrocyte culture~ after MTX withdrawal (total of 12 determinations, no dBcAMP).
Our studies show that MTX administration can cause morphologic changes in a relatively pure population of astrocytes in primary culture. The formation of cytoplasmic vacuoles, dense bodies, and extracellular debris that were noted by light and electron microscopy are similar to the changes seen in primary astrocyte cultures following administration of other toxic agents 15. The changes are most likely degenerative in nature, appear to be dose-related and are associated with small loci of necrosis at much higher concentrations of MTX. Although non-specific, these degenerative changes occurring in a relatively pure astroglial population provide supporting evidence that astrocytes may, indeed, be a target for MTX neurotoxicity. The mitotic inhibition and loss of viability that was observed in the astrocyte cultures are similar to the reported effects of MTX in other cell types6'11'24'28'39. The mechanism involved in MTX-induced toxicity to the astrocytes is not clear. MTX acts on replicating cells and is believed to exert its chemotherapeutic effect by competitive inhibition of dihydrofolate reductase (DHFR) 4'4°. DHFR catalyzes the formation of tetrahydrofolic acid which is important in DNA synthesisTM. MTX binding and DHFR activity have been demonstrated in primary CNS tumors,
28 chiefly astrocytomas and glioblastomas2. DHFR activity has also been reported in partially purified rabbit and mouse brain extracts, and is totally abolished by MTX 32'38. It is likely that astrocytes in culture also contain DHFR and that MTX may exert its adverse effect on these cells by inhibiting DHFR and DNA synthesis. Although a reduction in mitotic rate was persistently noted in MTX-treated astrocyte cultures after 1-3 administrations of MTX at various concentrations, the degree of mitotic inhibition did not correlate well with the drug dose and frequency of treatment. The reason for this apparent lack of correlation between drug dose and decline in mitotic activity is not clear. In other tissue culture systems, the rate of inhibition of DNA synthesis by MTX is found to be directly proportional to cellular growth rate and is greatest in cells that are in the S-phase of growth 19'2°. As in vivo, subpopulations of astroglial cells in different stages of maturation and growth have been documented to coexist in culture 1°'13'29'34. Since astrocyte cultures generally exhibit an asynchronous growth pattern, it is possible that the degree of MTXinduced mitotic inhibition observed in our cultures may be related to the number of sensitive cells that are in active growth (S) or the mitotic (G2/M) phase at the time of drug administration. On the other hand, subpopulations of cells that are in stationary (Go) phase of growth during the time of treatment may have been relatively resistant to the cytotoxic and/or mitotic-inhibitory effects of MTX. The differential sensitivity to MTX that was observed in our study may also be related to the inherent biological and biochemical properties of various subpopulations of astrocytes in culture 1°'29'3°'34. In addition to mitotic inhibition, a dose-related loss of cell viability was observed in astrocytes, suggesting that MTX may be acting on the cells through more than one mechanism. Studies in C6 rat glioma cells have shown that MTX can inhibit cell division while also altering morphology, stimulating production of S-100 protein and increasing steroid sulfatase activity26'27. MTX-induced DHFR inhibition causes a functional deficiency of reduced folate co-factors that are necessary for the metabolism of methionine, purine and histidine6"13'1'*'22.Likewise, DHFR inhibition may cause an intracellular build up of potentially toxic dihydrofolates1. Thus, MTX may affect morphological differentiation, biochemical functions and certain aspects of cellular metabolism other than DNA synthesis and cell division. It is possible that an MTX-induced decrease in folate co-factors and/or intracellular accumulation of oxidized folates may have caused metabolic perturbations in astrocytes leading to the degenerative changes and loss of viability that were seen in culture. Despite a persistent loss of cell viability, the growth rate in experimental cultures after MTX withdrawal did
not vary significantly from the control. This seeming inconsistency between cytotoxicity and cell growth after drug withdrawal is perhaps also related to the heterogeneity of cells in culture. The reactive-like astroglial hypertrophy and a relative increase in intermediate filaments that were noted in some cells in addition to morphologic evidence of cellular degeneration by light and electron microscopy after MTX treatment serve to support this assumption. To investigate the significance of the apparent astroglial hypertrophy and increased intermediate filaments, the cultures were stained with fluorescein-conjugated antibody to glial fibrillary acidic protein (GFAP) and examined by flow cytometry. Our preliminary findings showed increased GFAP content in 5-10% of the cell population in MTX-treated cultures compared to controls, both in the absence or presence of dBcAMP. This relative increase in GFAP content of treated cells suggests that at least a certain population of astrocytes may be undergoing an adaptive response to MTX since reactive astrocytes are generally characterized by the development of more abundant GFAP 29'31. Likewise, flow cytometric studies of the astrocyte cultures using fluorescein-labelled antibody to incorporated bromodeoxyuridine and propidium iodide showed a 4- to 5-fold increase in percentage of cells in S-phase, with a corresponding reduction in ratio of cells in Go/G1 and G2/M phases as well as a 3- to 4-fold increase in DNA content after 24 h of exposure to MTX (unpublished data). Thus, it is possible that while MTX may have exerted a cytotoxic effect on susceptible subpopulations of astrocytes (as evidenced by light and electron microscopic cellular degeneration as well as the decreased viability of treated cultures), it also allows certain adaptive responses in resistant cells which have not been injured by the drug. Apart from the degenerated cells, the large polygonal astrocytes that were seen following MTX administration in many ways resemble the reactive astrocytes found around sites of CNS injury. Similar changes have been described in culture and are believed to result from hypertrophy of activated astrocytes in response to environmental modification including addition of dBcAMP 3' 9,31. In our study, the hypertrophic astrocytes were found in MTX-treated cultures even in the absence of dBcAMP. Hence, they may represent a glial reaction to MTXinduced metabolic perturbations in culture, and may involve a certain subpopulation of astrocytes that are initially refractory to MTX. Cells that are grown in culture show an asynchronous pattern of growth and are distributed more or less randomly throughout all phases of the cell cycle. It is possible that astrocytes which are in quiescent state of the cell cycle during the period of treatment may be refractory to the deleterious effects of MTX, and may undergo reactive hypertrophy in response
29 tO the MTX-induced degeneration in the more sensitive cells (i.e. in growth phase of cell cycle). This may explain the reactive-like changes that were superimposed on astroglial degeneration following addition of MTX and may be relevant to the astrogliosis that is generally observed in clinical cases of M T X encephalopathy TM. The deleterious effects of M T X were consistently more prominent in the absence of d B c A M P and were less conspicuous when M T X was administered concurrently with dBcAMP. The cause for this apparent protective effect of d B c A M P on MTX-treated cultures can only be speculated on. Studies on Ehrlich ascites tumor cells in culture have shown that d B c A M P depresses net M T X uptake without affecting M T X efflux 4x. On the other hand, d B c A M P - i n d u c e d M T X efflux has also been reported !n isolated hepatocytes ~2. Recent literature also suggests that M T X may share a c o m m o n efflux route with cyclic nucleotides ~7. Cyclic A M P has been shown to affect D H F R gene expression and transcription 42. Likewise, d B c A M P and other agents that increase the intracellular levels of c A M P have been shown to block D H F R synthesis and entry of cells into S phase 16'25. Concurrent administration of dBcAMP to the astrocytes cultures may have inhibited the effect of MTX through any of the above mechanisms. In addition, dBcAMP inhibits proliferation of astrocytes in culture 7'36 and has been shown to arrest other cell systems in G1 phase of cell cycle27. Since REFERENCES 1 Abelson, H.T., Methotrexate and central nervous system toxicity, Cancer Treat. Rep., 62 (1978) 1999-2001. 2 Abelson, H.T., Corka, C., Fosburg, M. and Kornblith, P., Identification of dihydrofolate reductase in human nervous system tumors, Lancet, ii (1978) 184-185. 3 Basset, A.-E., Kalnins, V.I., Ahmed, I. and Fedoroff, S., A 48 kilodalton intermediate filament (IFAP) in reactive-like astrocytes induced by dibutyryl cyclic AMP in culture and in reactive astrocytes in situ, Z Neuropathol. Exp. Neurol., 48 (1989) 245-254. 4 Bertino, J.R., The mechanism of action of the folate antagonists in man, Cancer Res., 23 (1963) 1286-1306. 5 Booher, J. and Sensenhrenner, M., Growth and cultivation of dissociated neurons and glial cells from embryonic chick, rat and human brain in flask cultures, Neurobiology, 2 (1972) 97-105. 6 Borsa, J. and Whitmore, G.E, Cell-killing studies on the mode of action of methotrexate in L-cells in vitro, Cancer Res., 29 (1969) 737-744. 7 Coffino, E and Gray, J.W., Regulation of $49 lymphoma cell growth by cyclic adenosine 3"5"monophosphate, Cancer Res., 38 (1978) 4285-4288. 8 Crosley, C.J., Rorke, L.B., Evans, A. and Nigro, M., Central nervous system lesions in childhood leukemia, Neurology, 28 (1978) 678-675. 9 Fedoroff, S., McAuley, W.A., Houle, J.D. and Devon, R.M., Astrocyte cell lineage. V. Similarity of astrocytes that form in the presence of dBcAMP in cultures to reactive astrocytes in vivo, J. Neurosci. Res., 12 (1984) 14-27. 10 Fedoroff, S., Neal, J., Opus, M. and Kalnins, V.I., Astrocyte cell lineage. III. The morphology of differentiating mouse astrocytes in colony culture, J. Neurocytol., 13 (1984) 1-20.
MTX acts mainly on replicating cells, the dBcAMP-induced arrest in cell cycle may also render the astrocytes more refractory to the deleterious effects of MTX. In summary, our studies indicate that astrocytes in culture can be directly and adversely affected by MTX especially in the absence of dBcAMP, while at the same time also causing reactive-like changes in other cells. Reactive astroglial hyperplasia is a consistent finding in clinical cases of M T X encephalopathy, and is generally believed to be a reaction to MTX-induced injury to the CNS. While the precise cellular target(s) of MTXinduced encephalopathy are still not established, the results of this study, as well as the recently reported MTX-induced morphological alterations in astrocytes in vivo 7, provide supportive evidence that astrocytes may be a direct and primary target of M T X toxicity, and suggests that the astrogliosis seen in M T X encephalopathy may in part be related to MTX-induced astrocytic injury. The mechanism(s) involved in the observed MTX effects on astrocytes remain(s) to be clarified. The use of primary astrocyte cultures as described in this report may serve as a excellent model for further investigation of the pathogenetic mechanism(s) involved in acute M T X encephalopathy.
Acknowledgement. Supported by Research Grant NS 24853 from the National Institutes of Health.
11 Galivan, J., Evidence for cytotoxic activity of polyglutamate derivatives of methotrexate, Mol. Pharmacol., 17 (1980) 105110. 12 Gewirtz, D.A., Randolph, J.K. and Goldman, I.D., Effiux in isolated hepatocytes as a possible correlate of secretion in vivo: induced exit of folic acid analog methotrexate by dibutyryl cyclic AMP or isobutylmethylxanthine, Biochem. Biophys. Res. Commun., 101 (1981) 366-374. 13 Goldman, J.E. and Chiu, E-C., Growth kinetics, cell shape, and cytoskeleton of primary astrocyte cultures, J. Neurochem., 42 (1984) 175-184. 14 Gregorios, J.B., Gregorios, A.B., Mora, J., Marcillo, A., Fojaco, R.M. and Green, B., Morphologic alterations in rat brain following systemic and intraventricular methotrexate injection: light and electron microscopic studies, J. Neuropathol. Exp. Neurol., 48 (1989) 33-47. 15 Gregorios, J.B., Mozes, L.W., Norenberg, L.-O.B. and Norenberg, M.D., The morphologic effects of ammonia on primary astrocyte cultures. I. Light microscopic studies, J. Neuropathol. Exp. Neurol., 44 (1985) 397-403. 16 Gudewicz, T.M., Morhenn, V.B. and Kellems, R.E., The effect of polyoma virus, serum factors and dibutyryl cyclic AMP on dihydrofolate reductase synthesis, and the entry of quiescent cells into S phase, J. Cell Physiol., 108 (1981) 1-8. 17 Henderson, G.B. and Tsuji, J.M., Methotrexate efflux in L1210 cells. Kinetics and specificity properties of the efflux system sensitive to bromosulfophthalein and its possible identity with a system which mediates the efflux of 3"5"-cyclic AMP, J. Biol. Chem., 262 (1987) 13571-13578. 18 Hendin, R., DeVivo, D.S., Torack, R., Lell, M.E., Ragab, A.H. and Vietti, T.J., Parenchymatous degeneration of the central nervous system in childhood leukemia, Cancer, 33 (1974) 468-482.
30 19 Hryniuk, W.M., Fischer, G.A. and Bertino, J.R., S-phase cells of rapidly growing and resting populations. Differences in response to methotrexate, Mol. Pharmacol., 5 (1969) 557-564. 20 Johnson, L.E, Fisherman, C.L. and Abelson, H.T., Resistance of resting 3T6 mouse fibroblasts to methotrexate cytotoxicity, Cancer Res., 38 (1978) 2408-2412. 21 Kamen, B.A., Methotrexate, folate and the brain, Neurotoxicology, 7 (1986) 209-216. 22 Kamen, B.A., Nylen, P.A., Camitta, B.M. and Bertino, J.R., Methotrexate accumulation and folate depletion in cells as a possible mechanism of chronic toxicity to the drug, Br. J. Haematol., 49 (1981) 355-360. 23 Kaplan, R.S. and Wiernik, P.H., Neurotoxicity of antineoplastic drugs, Semin. Oncol., 9 (1982) 103-130. 24 Keefe, D.A., Capizzi, R.L. and Rudnick, S.A., Methotrexate cytotoxicity for L5178 Y/Asn-lymphoblasts: relationship of dose and duration of exposure to tumor cell viability, Cancer Res., 42 (1982) 1641-1645. 25 Kellems, R.E., Morhenn, V.B., Pfendt, E.A., Alt, EW. and Schimke, R.T., Polyoma virus and cyclic AMP-mediated control of dihydrofolate reductase mRNA abundance in methotrexateresistant mouse fibroblasts, J. Biol. Chem., 254 (1979) 309-318. 26 Kolber, A.R., Perumal, A.S., Goldstein, M.N. and Moore, B.W., Drug-induced differentiation of a rat glioma in vitro. II. The expression of S-100, a glial specific protein and steroid suifatase, Brain Research, 143 (1978) 513-520. 27 Leof, E.B., Wharton, W., Okeefe, E. and Pledger, W.J., Elevated intracellular concentrations of cyclic AMP inhibited serum-stimulated, density-arrested BALB/c-3T3 cells in mid-G1, J. Cell Biochem., 19 (1982) 93-103. 28 Li, J.C. and Kaminskas, E., Accumulation of DNA strand breaks and methotrexate cytotoxicity, Biochemistry, 81 (1984) 5694-5698. 29 Manthorpe, M., Adler, R. and Varon, S., Development, reactivity and GFA immunofluorescence of astroglia-containing monolayer cultures from rat brain, J. Neurocytol., 8 (1979) 605-621. 30 McCarthy, K., An autoradiographic analysis of fl-adrenergic receptors in immunocytochemically defined astroglia, J. Pharmacol. Exp. Ther., 226 (1983) 282-290.
31 Miller, R.H., Abney, E.R. and David, S., Is reactive gliosis a property of distinct subpopulation of astrocytes?, J. Neurosci., 6 (1986) 22-29. 32 Pollock, R.J. and Kaufman, S., Dihydrofolate reductase is present in brain, J. Neurochem., 30 (1978) 253-256. 33 Price, R.A. and Jamieson, P.A., The central nervous system in childhood leukemia. II. Subacute leukoencephalopathy, Cancer, 35 (1975) 306-318. 34 Raft, M.C., Abney, E.A., Cohen, J., Lindsay, R. and Noble, M., Two types of astrocytes in cultures of developing rat white matter: differences in morphology, surface gangliosides and growth characteristics, J. Neurosci., 3 (1983) 1289-1300. 35 Rubinstein, L.J., Herman, M.M., Long, T.F. and Wilbur, J.R., Disseminated necrotizing leukoencephalopathy: a complication of treated central nervous system leukemia, Cancer, 35 (1975) 291-305. 36 Sensenbrenner, M., Devilliers, G., Bock, E. and Porte, A., Biochemical and ultrastructural studies of cultures rat astroglial cells. Effect of brain extract and dibutyryl cyclic AMP on glial fibrillary acidic protein and glial filaments, Differentiation, 17 (1980) 51-61. 37 Silbert, S.W. and Goldstein, M.N., Drug-induced differentiation of a rat glioma in vitro, Cancer Res., 32 (1972) 1422-1427. 38 Spector, R., Levy, P. and Abelson, H.T., The development and regional distribution of dihydrofolate reductase in rabbit brain, J. Neurochem., 29 (1977) 919-921. 39 Taylor, I.W., Slowiaczek, P., Francis, P.R. and Tattersall, M.H.N., Biochemical and cell cycle perturbations in methotrexate-treated cells, Mol. Pharmacol., 21 (1982) 204-210. 40 Werkheiser, W.C., The biochemical, cellular and pharmacological action and effects of the folic acid antagonists, Cancer Res., 23 (1963) 1277-1285. 41 White, J.C., Carchman, R.A., Fry, D.-W. and Goldman, I.D., Relationship between membrane transport of methotrexate and endogenous cyclic adenosine 3"5" monophosphate in the Ehrlich ascites tumor, Cancer Res., 40 (1980) 2400-2404. 42 Wu, J.S., Wiedemann, L.M. and Johnson, L.F., Inhibition of dihydrofolate reductase gene expression following serum withdrawal or dBcAMP addition in methotrexate-resistant mouse fibroblasts, Exp. Cell Res., 141 (1982) 159-169.
