Metabolism of the neurotoxin in MPTP by human liver monoamine oxidase B

Volume 186, number 2

July 1985

FEBS 2678

Metabolism of the neurotoxin in MPTP by human liver monoamine oxidase B Richard R Fritz, Creed W. Abell+, Nutan T. Patel, Wleslaw Gessner*+ and Arnold Brossl* Department of Human Blologwal Chemistry and GenetIcs, Dwwon of Blochemutry, Unrverslty of Texas Medical Branch. Galveston, TX 77550, and *Laboratory of Chemistry. Natronal Institute of Arthritis, Diabetes, and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, MD 20205, USA

Received 11 May 1985 The neurotoxm

I-methyl-4-phenyl-1,2,3,6-tetrahydropyrldlne (MPTP) was oxldlzed to dlhydropyrldme MPDP+ and pyrldme MPP+ by preparations of monoamme oxldase B (MAO B), mcludmg pure human hver MAO B monoclonal antlbody complex Km+vp values for MPTP and benzylamme, a preferred MAO B substrate, were 316 and 64,uM, respectively 4-Phenyl-1,2,3,6_tetrahydropyrldlne (PTP), the nor derlvatlve of MPTP, was also a substrate (KmTapp=221 PM) MPDP+, MPTP, and MPP+, but not PTP, were found to be irreversible mhlbltors of MAO B Our studies support the hypothesis that MPTP IS oxldtzed m prtmate bram by MAO B to MPDP+, which IS then converted to MPP+, a maJor metabohte found m the substantla mgra MPTP

MPDP+

MPP+

Monoamine oxldase

1. INTRODUCTION Several studies suggest that 1-methyl-Cphenyl1,2,3,6_tetrahydropyridme (MPTP) must be metabohzed to produce Parkmson-hke symptoms [l-4]. Monoamme oxidase (MAO, EC 1.4.3.4), the major intracellular enzyme which degrades biogenie ammes m mammals [5], is thought to play an important role in this process. Chiba et al. [l] found that m rat brain mitochondrial fracttons, MPTP can be converted by MAO B, one of the two forms of this enzyme, to its correspondmg pyridmmm derivattve, MPP+, a maJor metabohte in primates [6-71. Furthermore, Langston et al. [2] have shown that pretreatment of primates with the MAO B mhtbnor, deprenyl, prevents the development of MPTP-mduced parkinsonism, while studies by Heikktla et al. [3] demonstrate in mice that clorgyhne, an MAO A inhibitor, does not pre+ TO whom correspondence should be addressed + Visiting Scientist from A Mrckiewicz Umversity, Poznan, Poland

Dopamme rnhlbrtron

vent the dopamme-depleting effect of the drug. Our studies [8] show that MPTP is metabolized to MPP+ by MAO B via the drhydropyridine MPDP+. Simultaneously, Salach et al 191 demonstrated that homogeneous preparations of bovine liver MAO B and human placental MAO A oxrdize MPTP with K,,, values similar to those of benzylamme. Both forms of the enzyme were rrreversrbly inactivated when incubated with high concentrations (l-5 mM) of MPTP. Collectively, these data led us to examine the kinetic properties of human MAO B with MPTP as substrate, smce highly purified preparations of human or other primate MAO B had not previously been studied.

2. MATERIALS

AND METHODS

One source of enzyme used in these studies was pure, catalytically active MAO B : monoclonal antibody complex (MAO B : MAO B-lC2) prepared from human autopsy liver as described by Pate1 et al. [lo], using a well characterized monoclonal anPublished

224

Substantza mgra

00145793/85/$3

by Elsevler 30 0

Science Pubhshers

B V (Blomedlcal

D~vrs~on)

1985 Federation of European Biochemical Societies

July 1985

FEB.5 LETTERS

Volume 186, number 2

tibody [MAO B-1C2) which recognizes human MAO B but not MAO A [ll,l2]. MAO B:MAO B-1C2 was studied because MAO B has not yet been separated from MAO B-lC2 in catalytically active form, nor have preparations of pure human MAO B with hrgh catalytic activity been obtained by other methods. Liver was the most appropriate source of human MAO B because autopsy brain is difficult to obtain. The substrate preferences and molecular activities of human MAO B in hver and brain appear to be essentrally identical [ 131. Crude extracts of human hver mitochondria and partially purified bovine liver MAO B were prepared and studied as descrrbed in [8,9]. MPTP and 4-phenyl-1,2,3,6_tetrahydropyridine (PTP) were purchased from Aldrich. MPDP+ and MPPt were prepared and chara~erlzed as m [S] (see fig. 1 for structures). These compounds exhibit

different UV spectra, and the oxidation of MPTP 181 and PTP (unpublished) into then corresponding dihydropyridmmm derivatives can be followed spectrophotometrrcally at 340 nm (EM = 1 30 x 104) in aqueous solutions.

gdg-~ I

CH3

MPTP

MPDP’

A

I+

I

C”3

CH3

MPP’

B

c

F1g.1 Structures of MPTP, MPDP” and MPP+

Table 1 Kmettc parameters of MPTP, PTP and benzylamlne oxtdauon by human and bovine hver MAO

preparatrons Source of enzyme

Substrate

Km.aPP fuM)

Partially purified bovrne liver MAO B Human liver mnochondrtal (1 5% Trlton X-100)

extracts

Pure human hver MAO B *MAO B- 1C2

MPTP

v max,app

(nmol/min per mg)

benzy~amine

179 128

34 6 76 9

MPTP PTP benzylamme

152 282 34

2 47 4 17 11 2

MPTP PTP benzylamme

316 221 64

229 196 880

Partrally purtfted bovine MAO B was assayed spectrophotometricaliy m trtphcate wrth 9 different concentrattons of MPTP (40-650 ,uM, A 340~~)and benzylamme (25-250 pM, AzsD,,,,,) Km,app and Vmax,appvalues were calculated by linear regresston analysts of the data plotted as l/v against l/S The correlatton coefftctents of the regresston lines were 0 99 for MPTP and 0 98 for benzylamme and the values of the trrphcate pomts were + 8% SD Protein concentrations used for the enzyme activny measurements were 11 pg protem per sample for both substrates Human hver mttochondrrai extracts were assayed m trtphcate for actrvrty at 9 concentratrons of MPTP (20-200 FM), PTP (20-480 PM) and benzylamme (20-480 PM) Values for IY,,,+~~and Vmax,ar,pp were calculated as stated above with correiatton coefftctents of 0 99, 0.97, 0.98, respecttvely. The means for the triphcates were 4 Solo SD for benzylamme and MPTP and rt 10% SD for PTP. Protem Concentrations used for these measurements were 242 pg for MPTP and PTP and 48 fig for benzytamme per sample analyzed Pure human hver MAO B MAO B-ICI? was assayed at 9 concentrattons tn triphcate of MPTP (40-400/1M), PTP (40-4OOpM) and benzylamme (10-160 pM). iG,,app and V,,,ax,appvalues were obtained as described above, correlatron coefftctents were 0 98,O 98,0 99, respectrvely. The means of the trrphcates were + 5% SD for MPTP and PTP and k 8% SD for benzylamme MAO B protem concentratrons were 6.7 gg per sample for MPTP and PTP and 0 67 pg per sample for benzylamme

22s

Volume 186, number 2

FEBS LETTERS

July 1985

3. RESULTS Table 1 shows the kinetic properties for the oxidation of MPTP and PTP to their correspondmg drhydropyridimum analogs by pure human liver MAO B : MAO B-lC2 and detergent extracts of human liver mitochondrra. Km,app values for MPTP were similar, although not identical, m these 2 preparations. Km,app values for benzylamme, however, were approx 5-fold lower than for MPTP A comparison of MPTP and benzylamme oxidation by partially purified bovme MAO B (about 80% pure as Judged by exammation on SDS gels), which was prepared accordmg to Salach [ 141, gave similar K,,,,appvalues (179 and 129 PM, respectively). In all MAO B preparations studied, however, benzylamine was oxidized at a 2-4-fold faster rate than MPTP. MAO B catalytrc activity declined progressively when MPTP was used as substrate, suggesting that the enzyme was being inactivated. Therefore, the effects of MPTP, MPDP+, MPP+ and PTP on benzylamme oxidation of MAO B:MAO B-lC2 were assessed as a function of time. Fig.2 shows that after a short initial lag, enzyme inactivation by MPTP followed first-order kinetics. MPDP+ and MPP+ also inactivated the enzyme complex, but PTP (1 mM) had no effect. The pseudo first-order inactivation of constants for the rate MAO B:MAO B-lC2 were 3.92 x lo-’ mu-’ for MPTP (1 mM), 3.48 x lo-’ mu-’ for MPDP+ (0.2 mM), and 1.77 x 10e2 mu-’ for MPP+ (10 mM) Attempts to recover actrvrty by overnight dialysis against buffer were unsuccessful. incubation of [3H]MPTP with Furthermore, MAO B : MAO B-lC2 resulted m the mcorporation of tritium into protein, which could not be removed by extensive washing with buffer. These results support the concept that MPTP can be oxidized by human MAO B to MPDP+, an intermediate in the formation of the metabolite MPP+. Using crude preparations of MAO B, we have shown that MPTP is oxidized to MPDP+ at a 5-fold faster rate than MPDP+ 1s converted to MPP+ [8]. However, since MPDP+ chemically undergoes rapid disproporttonatron at the pH of the mcubatron mixture [ 15,161 and has spectral properties in water (max 343 nm) which are similar to those of MPP+ (max. 293 nm), it is presently not possible to determine whether the 226

I

1,

I

‘h, 20

40

60

60

: 20

100 Time

: 40

: 60

: 60

4

100

ImInI

Fig 2. Irreversible

mactlvatton of pure human hver MAO B.MAO B-lC2 by MPTP (1 mM, panel A), MPDP+ (0 2 mM, panel B), and MPP+ (10 mM, panel MAO B.MAO B-lC2 (62pg protein) was C) suspended m 50 mM KHzPO~/K~HPO~ buffer, pH 7.4, and incubated at 30°C with 1 mM MPTP, 0.2 mM MPDP+, and 10 mM MPP+, respectively At the times mdtcated, trtphcate lop1 ahquots were removed, added to the assay cuvettes, and diluted to 1 0 ml wtth the assay solutron, whtch contamed 3 mM benzylamme m 50 mM potassmm phosphate buffer, pH 7 4 Imttal rates were determined from 0 to 2 mm by measurmg the increase m absorption at 250 nm using a Beckman DU-7U spectrophotometer Panel D shows linear regression analyses of the data; the correlatton coefftcients m all cases were >0 99 (0) Control or 1 mM PTP (panel A), (A) 1 mM MPTP, (A) 0 2 mM MPDP+; (0) 10 mM MPP+ The calculated rate constants k (mu-‘) were 3 92 x lo-’ for 1 mM MPTP, 3 48 x lo-* for 0.2 mM MPDP+ and 1 77 x IO-* for 10 mM MPP+ Overmght dtalysts vs phosphate buffer after lOO-fold ddutton drd not affect the acttvtty measurements formation of MPP+ from MPDP+ is the result of this chemical transformatron or addrtronal enzymatic oxidation. MPTP and its oxidation products were found to inhibit MAO B irreversibly with potencres ranked as follows: MPDP+ > MPTP > MPP+. In contrast, PTP did not inactivate MAO B. The N-

methyl substrtuent is clearly Important for nreversable enzyme mhibition.

Volume 186, number 2

FEBS LETTERS

4. DISCUSSION The relevance of MAO inactivation to the neurotoxrc effects of MPTP in vivo IS unclear. Patients treated for endogenous depression with irreversible inhibitors of MAO A and B (e.g., tranylcypromme) do not develop parkinsonian symptoms [ 171. Furthermore, if irreversible inhrbition of MAO B were the crucial event for neurotoxicity, many MAO B-containing neurons would presumably undergo similar damage. In fact, however, we have shown that MAO B 1s not found m the highly sensitive cells of the substantra nigra. Our localization studies of MAO A and MAO B m Macaca cynomolgus monkey brain, performed with MAO type-specific monoclonal antibodies, revealed that MAO B is localized largely in serotonergic neurons and not dopammergic while neurons, MAO A is localized in catecholammergic regions, including the substantia nigra [18]. Given these patterns of drstrrbutron, we hypothesize that MAO B catalyzes the oxidation of MPTP to MPDP+ and MPP+ in serotonergrc neurons which impinge either on dopammergic cells in the substantra nigra or their terminals m the striatum, or m MAO B-containing, astrocytic ghal cells. These metabolites may be transported out of these cells and taken up by neurons in the substantra mgra. After this work was completed, Javitch et al. [19] also proposed that MPP+ may be produced m astrocytes and transported into neurons in the substantra nigra. Previous studies show that 48-72 h after MPTP administration to monkeys, MPP+ accumulates in the substantra nigra, but decreases in surrounding regrons of brain [20]. In rats, MPP+ 1s transported into neurons through the dopamme uptake system in striatal preparations [21]. The biologrcal target(s) of MPP+ m neurons of the substantra mgra 1s unknown, but one possible candidate is MAO A. In vrtro preparations of pure human placental MAO A converts MPTP to MPP+, an irreversible mhrbitor of this enzyme [9]; however, MPTP is a reversible competitrve mhibitor of rat brain MAO A [22]. If MPP+ is an effective m vrvo mhrbitor of prrmate neuronal MAO A, its accumulation in the substantra mgra could interfere with the function of this important amme-degrading enzyme and thereby contribute to specific neurotoxicity.

July 1985

Collectively, the results of our studies of human MAO B and those of others using MAO B from other mammals [l-4,9] support the hypothesis that MPTP IS oxidized by MAO B to compounds that can cause significant mhibrtron of both MAO A and MAO B. These molecular events could contribute to specrfic neuronal toxicity in the substantia nigra and lead to Parkinson’s disease. ACKNOWLEDGEMENTS We thank B. Witkop, G. Hillman, K. Johnson, and C. Denney for helpful comments on the manuscript. Supported by the Multidrsciphnary Research Program on Schizophrema, Umversrty of Texas Medical Branch, Galveston.

REFERENCES 111Chtba, K , Trevor, A and Castagnoh, N Jr (1984) Brochem Btophys Res. Commun. 120, 574-578 J.W , Irwin, I., Langston, E.B and Forno, L.S. (1984) Sctence 225, 1480-1482. [31 Hetkktla, R.E , Manzmo, L., Cabbat, F S. and Duvotsm, R.C. (1984) Nature 311, 467-469 141 Cohen, G. and Mytdineou, C. (1985) Life Scl. 36, 237-242 PI Von Korff, R.W (1979) m. Monoamme Oxrdase* and Altered Functions Structure, Functton, (Singer, T P et al eds) pp l-6, Academic Press, New York 161 Markey, S P., Johannessen, J.N , Chmeh, C C , Burns, R.S. and Herkenham, M.A (1984) Nature 3 11, 464-467 171 Langston, J.W , Irwin, I , Langston, E B. and Forno, L S (1984) Neurosct Lett 48, 87-92 PI Gessner, W., Brossr, A., Shen, R.S., Fritz, R R and Abell, C.W. (1984) Helv. Chum. Acta 67, 2037-2042 (91 Salach, J.I., Singer, T.P , Castagnoh, N Jr and Trevor, A (1984) Brochem. Brophys Res Commun. 125, 831-835. WI Patel, N T , Frttz, R R and Abell, C W (1984) Brochem Brophys Res. Commun. 125, 748-754 ill1 Denney, R M., Frrtz, R.R., Pate], N.T and Abell, C W (1982) Science 215, 1400-1403. 1121 Denney, R.M., Patel, N T , Frrtz, R.R. and Abell, C.W (1982) Mol. Pharmacol. 22, 500-508 1131 Denney, R.M., Fritz, R R , Patel, N T , Wrden, S.G. and Abell, C.W. (1983) Mol. Pharmacol 24, 60-68.

121Langston,

227

Volume 186, number 2

FEBS LETTERS

[14] Salach, J I Jr (1978) Methods Enzymol. 53, 495-501. [15] Castagnoh, N Jr, Chlba, K and Trevor, A J (1985) Life Scl 36, 225-230. [ 161 Gessner, W , Brossl, A , Shen, R.-S. and Abell, C W. (1985) FEBS Lett , m press [17] Rlgal, F. and Zanflan, E (1983) Mod Probl Pharmacopsychlat. 19, 162-169 [18] Westlund, K N., Denney, R.M., Kochersperger, L M., Rose, R.M. and Abell, C W (1985) Science, submitted

228

July 1985

(191 Javltch, J A , D’Amato, R -J., Stnttmatter, S M and Snyder, S H (1985) Proc Nat1 Acad SCI USA 82, 2173-2177 [20] Irwin, I. and Langston, J.W (1985) Life Sa 36, 207-212 [21] Javltch, J A. and Snyder, S H. (1984) Eur J Pharmacol 106, 455-456 [22] Fuller, R W. and Steranka, L R (1985) Life Sa 36. 243-247.