Effects of neurotoxins on neurone-specific enolase and lactate dehydrogenase activity and leakage in neuroblastoma cells

Toxic. in Vitro Vol. 5, No. 5/6, pp. 439--442,1991 Printed in Great Britain.All rights reserved

0887-2333/91 $3.00+ 0.00 Copyright © 1991PergamonPresspie

EFFECTS OF NEUROTOXINS ON NEURONE-SPECIFIC ENOLASE A N D LACTATE DEHYDROGENASE ACTIVITY A N D LEAKAGE IN NEUROBLASTOMA CELLS S. M. THOMAS*,C. L. HARTLEYt and H. J. MASONt *Divisionof BiologicalSciences,School of BiologicalSciences,Hattield Polytechnic,College Lane, Hatfield, Herts. ALl0 9AB and i'Occupational Medicine and Hygiene Laboratories, Health & Safety Executive, 403 Edgware Road, London NW2 6LN, UK Abstract--Differentiation of the C 1300 N2A (mouse) and IMR 32 (human) neuroblastoma cell lines with bromodeoxyuridine induces the expression of neurone-specificenolase (NSE). The expression of NSE is increased approximately three-fold on differentiation of the IMR 32 cells and two-fold in the C1300 cells. The amount of intracellular NSE and lactate dehydrogenase (LDH), and the release of these cytosolic proteins, was followed after exposure of differentiated IMR 32 cells to the neurotoxins 1-methyl-4-phenyl pyridinium (MPP+) and kainic acid. After 48 hr of exposure to MPP + (10-8 M-10-6M) there was a concentration-dependent decrease in intracellular LDH and NSE. However, the amounts of these proteins measured in the medium suggested (i) that there was no concentration-related increase in cell death; and (ii) that the amounts in the medium reflectedintracellular levelsof these proteins. Data obtained previously showed that, after 24 hr exposure, these concentrations of neurotoxin caused changes in cellular protein degradation that were not accompanied by cell death. Several parameters of cellular protein metabolism show toxin-induced changes at low dose levelsin the absence of concomitant cell death. Therefore, indices of deranged protein metabolism may provide sensitive markers of neurotoxicity.

Introduction The aim of this study was to find an early indicator of cell damage by using in vitro tissue culture systems to investigate the effects of sublethal doses of known neurotoxins on protein homoeostasis. Neuroblastoma cell lines provide a convenient model for studying the molecular and biochemical mechanisms of neurotoxicity (Cole et al., 1985). The cell lines used in this study were the human neuroblastoma IMR 32 cell line and the mouse neuroblastoma cell line derived from the C1300 N2A clone. Various cell lines have been derived from clones of the C1300 mouse neuroblastoma and have been used extensively for in vitro studies of neurobiological function. Human cell lines have not been used widely in neurotoxicological studies, despite the lack of adequate animal models and the availability of human neuroblastoma cell lines. IMR 32 is a human neuroblastoma cell line established by Tumilowicz et al. (1970) and characterized morphologically and biochemically by a number of groups (Gupta et al., 1985; Prasad et al., 1973). We report results from experiments using the chemically dissimilar neurotoxins 1-methyl-4-phenylpyridinium (MPP+), the active metabolite of 1methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP), and kainic acid. The concentration-related effects on the intracellular levels and their extracellular leakage

of the two cytoplasmic proteins--lactate dehydrogenase (LDH) and neurone-specific enolase (NSE)--were compared. Sublethal concentrations at which protein degradation was affected by these neurotoxins were used (Thomas and Anderton, 1991). Materials and Methods

Tissue culture medium, (modified Eagle's medium; MEM), heat-inactivated foetal calf serum (HIFCS) and gentamicin were from Gibeo Ltd (Slough, Berkshire, UK). The human neuroblastoma (IMR 32) and mouse neuroblastoma (C1300 N2A) cell lines were obtained from Flow Laboratories (Irvine, UK). Tissue culture biochemicals (tissue culture grade) and kainic acid were purchased from Sigma (Poole, Dorset, UK), MPP + was from Research Biochemicals Incorporated (Natick, MA, USA). IMR 32 and C1300 N2A cells were grown and differentiated as previously described (Thomas and Anderton, 1991). Differentiated IMR 32 cells were exposed to MPP + or kainic acid in 10 ml of complete medium, l-ml samples of conditioned medium were removed at various times for analysis of NSE and LDH levels; the samples were replaced by 1 ml of the appropriate medium. At the end of the experiment the cells were washed three times with phosphate buffered saline (PBS) before freezing. NSE and LDH levels were measured in differentiated and undifferentiated *To whom all correspondence should be addressed. IMR 32 and C1300 N2A cells. Cells were resuspended Abbreviations: ANOVA- analysis of variance; HIFCS = in PBS (1 ml) and sonicated three times using a Dawes heat-inactivated foetal calf serum; IRMA=immuno- Ultrasonic Generator Type 7532B on half power for radiometric assay; LDH = lactate dehydrogenase; MEM =modified Eagle's medium; MPP + -- 1-methyl-4- 15 sec. Protein concentrations were measured using an phenylpyridinium; MPTP = 1-methyl-4-phenyl-l,2,3,6tetrahydropyridine; NSE = neurone-specific enolase; automated hicinchoninic acid protein assay (Smith PBS = phosphate buffered saline; SEM = standard error et al., 1985). NSE was measured using the coated bead 'Prolifigen' NSE immunoradiometric assay (IRMA) of the mean. 439

440

S.M. THOMASet al.

(Sangtec Medical, Dietzenbach, Sweden). This is a two-site, monoclonal antibody radiometric assay-the antibodies recognized different epitopes on the 7-subunit of NSE. NSE activity in the sonicated cells was measured by the method outlined in the kit, although the method was modified to improve its sensitivity for the measurement of NSE in the medium. Thus, the beads, coated with one antibody, were preincubated with 25/tl of standard diluted with 225/~1 of fresh medium, or 250/tl of conditioned medium samples. The beads were then washed before incubating with the second antibody labelled with radiolabelled iodine (1251). This modification allowed a ten-fold increase in sensitivity. LDH activity was measured by automated fluorimetric and spectrophotometric methods. The fluorimetric method was a modification of the spectrophotometric method of Demetriou et al. (1974) and the automated spectrophotometric method was adapted from the Sigma LDH Isoenzymes kit (kit No. 705-C), using phenazine methosulphate (PMS) and tetranitroblue tetrazolium (TNBT) (Tsou et al., 1956). Data were analysed by one-way analysis of variance (ANOVA) with the Scheffe multiple-range tests and correlation analyses. Results

Differentiation of the two cell lines, IMR 32 and C1300, led to an increase in cellular levels of NSE. In the human-derived IMR 32 cells it increased from 0.138 #g/mg cellular protein to 0.393 #g/mg on differentiation, while in the mouse-derived C1300 cells it increased from 0.059/~g/mg protein to 0.107 #g/mg. Experiments to validate the measurement of NSE

in the mouse cell line showed complete parallelism between assay standards, NSE in IMR 32 cells and NSE in the C1300 cells. Measurement of the intracellular levels of LDH and NSE showed a clear concentration-related decrease for both these two proteins after 48 hr exposure to MPP ÷ (Fig. IA). NSE and LDH showed a two-phase pattern over the concentration range employed for kainic acid (Fig. 1B). At 1 0 - r i the concentration of both cytoplasmic proteins was reduced in comparison with control values, and at the higher toxin dose of 10 -5 i their levels increased back towards the control values. LDH showed an increase above the mean control value at the lowest dose of 10 - 7 M. Measurement of total LDH and NSE in the medium was performed at all the time points. Correlation between the two methods used to measure LDH was good (r = 0.87, where n = 67), although both methods gave considerable values for LDH in the control medium samples while levels of NSE (the specific neuronal cell protein) in the control medium approached zero (mean = 0.01 # g/litre; SEM = 0.004/~ g/litre), the limit of detection of the assay. Figure 2A shows the increase in the amount of NSE in the media over time in the control samples. At the 48-hr time point the control medium content of 14.7 ng NSE represented approximately 12-14% of the total cellular content of NSE. At no time did any dose of MPP ÷ increase the extracellular level of NSE above the level found in the control, thus suggesting that these doses of MPP + were not causing significant increase in cell death. The levels of NSE in the medium at 48 hr for the highest MPP ÷ dose were significantly lower than the control values and showed a dose pattern that paralleled the intracellular levels of NSE (Fig. 1A). Correlation

(A) t.2

(B) 1.2

"

i

IINSE JLDH I

llllm -

"I

~o.e -

~D "a

>O.B

c.

'1

C o

uO.B . o

~0.4

-

r.t U

0.2-

O-

ConLro]s

t0-B M

,--,

e~ tO

N

°e¢

Iii

t

IINsE JLDH

°I

10-7 H

t 0 - 6 14

L 4.1 C Q

uO.6 o ID

0.2

0

"

controls

i 0 - 7 14

tO-6 M

t0-5 M

Fig. 1. Intracellular levels of NSE and LDH activity (fluorimetric assay) after 48 hr exposure to different concentrations of MPP ÷ (A) or kainic acid (B). Data expressed relative to mean value in respectivecontrol. Data are means _ SEM, n = 4. *Indicates significant difference (P < 0.05) to control value using Scheffe's multirange ANOVA test.

Effects of neurotoxins on neuroblastoma cells

(A)

i media ~ 2 hr ~ 2 3 hr [-]29 hr ml46 hr

IB

15

441

(e)

Ilmoaio hr

~2 50

~23 hr E]29 hr

|

,IJ cO

| 12 4J

u

i 30

9 "o m r-

1

,., 6

¢,.

0 control

tO-BM

tO-7N

tO-6N

control

IO-B M

10-7 N

t0-6 M

Fig. 2. Extracellular levels of NSE (A) and LDH activity (fluorimetric assay) (B) after exposure to different concentrations of MPP + over time periods of up to 48 hr. Data are means 5- SEM, n = 4. *Indicates significant difference (P < 0.05) to control value using Scheffe's multirange ANOVA test. between intracellular N S E and L D H levels with the m e d i u m content o f the two proteins at 48 hr were r = 0.79 and r = 0.66, respectively, where n = 16. The time-course for extracellular L D H (Fig. 2B) shows no

significant increase in the release o f this protein at the three dose levels o f M P P +. Kainate did not cause any significant increase in the m e d i u m levels o f N S E or L D H levels above the

i.odia

(A)

I~

IB

(8)

|,oalo

hr

,,

.o

1

2.,

o 12 | 3O

n

i

'° 10

• control

0 lO-7H

lO-6N

IO-SN

control

10-7 N

10-6 H

t0-5 N

Fig. 3. Extracellular levels of NSE (A) and LDH activity (fluorimetric assay) (B) after exposure to different concentrations of kainic acid over time periods of up to 48 hr. Data are means + SEM, n = 4.

442

S. M. THOMASet al.

control value at any respective time point, suggesting that the neurotoxin concentrations were not causing increased cell death (Fig. 3A and 3B). While the 48-hr medium levels of NSE were not significantly different between dose levels, the pattern paralleled that seen in the intracellular levels (Fig. 1B), with the 10-6M kainate concentration being lower than the 10-7M or 10 -5 i concentration. A correlation coefficient (r) of 0.52 was obtained between the intracellular and 48-hr extracellular levels of NSE in the kainate-exposed samples. The medium levels of LDH at each time point were also not significantly different from the respective control value (Fig. 3B). Discussion

Increased cell death, as measured by leakage into the medium of the two cytoplasmic proteins LDH and NSE was absent during the experiment, although effects on the intracellular levels of LDH and NSE were apparent. MPP ÷ significantlyreduced the cellular levels of the proteins after 48-hr of exposure to the toxin in a simple concentration fashion, while kainate produced a biphasic response in these cytoplasmic proteins. These results were interpreted as suggesting that the toxin levels caused alteration in the metabolic synthesis or degradation of the proteins. This is consistent with an earlier study on protein degradation (Thomas and Anderton, 1991) in which a different assay for cell death was used. It has been shown previously (Thomas and Anderton, 1991) that both MPP ÷ and kainic acid have a non-linear, concentration-dependent effect on protein degradation in both differentiated human (IMR 32) and mouse neuroblastoma (C1300) cell lines. The concentration-dependency was biphasic, with an increase in protein degradation at lower doses of neurotoxin and a decrease at higher doses, although the biphasic response was at different doses for the two toxins. These effects on protein degradation after 24 hr occurred in the absence of increased cell death. We report here results from a further study of the effects of similar concentrations of these two neurotoxins on two specific cytoplasmic proteins--LDH and NSE. The medium levels of NSE and LDH at 48 hr suggested that they reflected the intracellular concentration rather than indicating increased cell death. NSE is a specific neuronal cell protein and measurement of its leakage may have advantages over LDH leakage as a meaurement of neuronal cell death. Assays of intracellular NSE may also provide a valuable indication of altered neuronal cytoplasmic protein metabolism. LDH levels, as measured by the two enzymic methods, were high in all samples of medium that contained HIFCS, including fresh media that had not been placed on cells, and any release of LDH from cells into medium had to be measured

above this high baseline signal. However, NSE levels in freshly prepared medium approximated to zero (at the limit Of detection of the adjusted assay) (approx. 0.02#g/litre), and leakage (0.3#g/litre in media samples) was clearly measurable in all samples after only 2 hr incubation. The immunochemical assay system developed for human NSE also proved to be equally suitable for measuring mouse-derived NSE from C1300 cells. Measurement of NSE may also prove useful in detecting specific neuronal cell cytoplasmic leakage or altered cytoplasmic protein metabolism in in vitro toxicological studies using mixed cell culture systems. Altered metabolism of specific proteins in cultured cells may provide a useful parameter for assessing sublethal toxicity of some compounds. Acknowledgements--S.M.T. was funded partly by a grant from Action Research for the Crippled Child and partly by the charity Fund for Replacement of Animals in Medical Experiments (FRAME). REFERENCES

Anderton B. H. (1987) Alzheimer's disease. Tangled genes and proteins. Nature, London 329, 106-107. Cole G. M., Wu K. and Timiras P. S. (1985) A culture model for age-related human neurofibrillary pathology. International Journal of Developmental Neuroscience 3 (1), 23-32. Demetriou J. A., Drewes P. A. and Gin J. B. (1974) Enzymes. In Clinical Chemistry Principles and Practise. Edited by R. J. Henry, D. C. Cannon and J. W. Winkelman. pp. 815-1002. Harper and Row, MD. Gupta M., Notter M. F. D., Felters S. and Gash D. M. (1985) Differentiation characteristics of human neuroblastoma ceils in the presence of growth modulators and anti-mitotic drugs. Developmental Brain Research 19, 21-29. Prasad K. N., Mandal B. and Kumar S. (1973) Human neuroblastoma cell culture: effect of 5'-bromodeoxyuridine on morphological differentiation and levels of neural enzymes. Proceedings of the Society for Experimental Biology and Medicine 144, 38-42. Smith P. K., Krohn R. I., Hermanson G. T., Mallia A. K., Gartner F. H., Provenzano M. D., Fujimoto E. K., Goeke N. M., Olson B. J. and Klenk D. C. (1985) Measurement of protein using bicinchoninic acid. Annals of Biochemistry 150, 76-85. Thomas S. M. and Anderton B. H. (1991) Effects of methylphenylpyridinium and kainic acid on protein degradation in differentiated neuroblastoma cultures. Toxicology in Vitro 5, 173-180. Tsou K. C., Cheng C. S., Nachlas M. M. and Seligman A. M. (1956) Synthesis of some p-nitrophenyl substituted tetrazolium salts as electron acceptors for the demonstration of dehydrogenases. Journal of the American Chemical Society 78, 391. Tumilowicz J. J., Nichols W W., Cholan J. J. and Green A. E. (1970) Definition of a continuous human cell line derived from neuroblastoma. Cancer Research 30, 2110-2218.