Age-related memory decline and longevity under treatment with selegiline

Life Sciences, Vol. 55, No,s 25/26, pp. 2155-2163, 1994 Copyright © 1994 Elsevier Science Ltd Printed in the USA. All rights reserved 0024-3205/94 $6.00 + .00

Pergamon 0024-3205(94)00396-3

AGE-RELATED MEMORY DECLINE AND LONGEVITY UNDER TREATMENT WITH SELEGILINE

S. Stoll, U. Hafner, O. Pohl, W. E. MOiler Central Institute for Mental Health, Department of Psychopharmacology, J5, D-68159 Mannheim, Germany

Summary The MAO-B inhibitor selegiline is used in the treatment of Parkinson's disease. Further, benefical effects in Alzheimer's disease have also been described as well as neuroprotective effects, increased longevity and an attenuation of age-related cognitive decline in experiments using rats. Our studies in mice and Syrian hamsters aim at the question whether the effects of selegiline reported in the rat can be generalized to other species. Aged female NMRI-mice (23 mo.) treated with selegiline (0.25 mg/kg, i.p., 3 times a week for 2-3 weeks) showed no treatment effect in the Morris water maze and in passive avoidance learning after 2 and 3 weeks of treatment. However, Syrian hamsters chronically treated with selegiline (0.05 mg/kg/day in the food, starting at 12 months old) showed a 3 month delay in the age-related decline of spontaneous alternation behavior, a measure of longer-term memory, compared to untreated controls. Since treated hamsters also show increased longevity (study still in progress) the data suggest a protective effect of a chronic treatment with selegiline against age-related cognitive and physical decline. Key Words: aging, longevity,memory, polyamines, selcgiline, NMDA receptor

The irreversible MAO-B inhibitor selegiline (L-deprenyl) is in clinical use for the treatment of Parkinson's disease. Several clinical studies further suggest a therapeutic effect in Alzheimers's disease (table 1), Animal studies have revealed three major actions of selegiline that may contribute to its clinical efficacy: an influence on cognitive performance (table 2), neuroprotective effects, and an increase in life span. The neuroprotective action of selegiline (16) may explain the decelerated progression of symptoms in patients with Parkinson's disease that was described for the selegiline treated groups of the DATATOP study after a washout period for selegiline (17). A striking extension of mean and maximum life span has been reported for rats after long term treatment with selegiline (18). This finding has been confirmed by others (19,20). However, smaller, but significant increases were found in these studies, probably due to differences in rat strains and different ages at the time when treatment was started. The increases in maximum life span are important observations and strongly suggest that aging can be slowed down by selegiline (18,19,20,21). Most of the animal data discussed above have been reported for the rat. Apart from the rat, the most standardized short-lived laboratory rodents are the mouse and the Syrian hamster. Therefore, in order to find out whether the results obtained in rats can be generalized to other species we are currently testing the effects of treatments with selegiline in mice and in Syrian hamsters. The following is a progress report on subchronic studies in mice and on a chronic study in hamsters that is close to conclusion. A chronic study in mice is at a too early stage to be discussed already.

2156

M e m o r y and Longevity under Selegiline

Vol. 55, No.s 25/26, 1994

TABLE 1

Survey of Clinical Studies on Cognitive Performance under Treatment with Selegiline in Patients with Alzheimer's Disease Effects

Treatment

Duration of Study

double-blind, placebo-controlled, parallel group

selegiline 10 mg/d

3 too.

selegiline 10 mg/d

3 too.

selegiline 10 mg/d

3 mo.

selegiline 10 mg/d

3 too.

14 short-term memoryT (4) 'visual-spatial functions T intellectual ]functions on the Gottfries, Brane and Steen Scale 1

selegiline 10 mg/d

y.

80 lowered (5) deterioration of the M/VISE

single-blind, parallel group

n

Source

Study Design

20 attention 1, memoryT (acquisition and recall), "stabilization of gross behavioral variables"

(1)

primary and declarative memory T, verbal fluency T

(2)

110

119 memoryt activities of daily living !

(3)

selegiline 10 mg/d 3mo. vs. L-acetylcarnitine 1000 mg/d

401antonomy in daily living I vs. NE memory! vs. (1) concentration Tvs.

1(6)

selegiline 10 mg/d vs. phosphatidylserine 200 mg/d

3 mo.

39 autonomy in daily livingT vs NE memoryT vs. (T)

(7)

selegiline 10 mg/d 3 mo. vs. oxiracetam 1000 mg/d

40 autonomy in daily livingT vs. NE memoryt vs. (T)

(8)

selegiline 10 mg/d and 40 mg/d vs. placebo (in 13 out of 17 patients)

17 Brief Psychiatric

](9,10)

(t)

within subject design

28d. 35 d.

Rating Scale1" at 10 mg/d and 40 mg/d episodic learning and memoryT at 10

mg/d

selegiline I 0 mg/d

180 d.

selective noun reminding task T at 10 mg 20 memory 1 attention T tendency of less deterioration after treatment (90 d.) compared with ~lacebo (90 d.)

Signs and Symbols: T improvement, NE no effect

(ll)

Vol. 55, No.s 25/26, 1994

Memory and Longevity under Selegiline

2157

TABLE 2 Survey of Animal Studies on Cognitive Performance under Treatment with Selegiline Effect

Duration of Study active avoidance learning in sexually sluggish rats I 36 wks.

(12)

spatial learning in rats T

lO d.

(13)

spatial learning in mice I

37 wks.

(14)

active avoidance learning in sexually sluggish rats I 12 mo.

Source

(15)

Signs and Symbols: I improvement, ~ deterioration

Methods Animals Female NMRI-mice had been obtained as retired breeders from Interfauna, Tuttlingen, Germany at the age of 12 months . They were kept in our animal facilities till the age of 23 months. Hamsters were from the German Federal Health Agency (Bundesgesundheitsamt), Berlin.

~,

750

E

600

[] controls • solegiline

.E 450 Q 0 e..

300

"ID

150

0 0

I

I

I

I

|

I

2

3

4

5

days Fig. 1 Morris water maze performance of 23 too. old female NMRI mice treated with 0.25 mg/kg selegiline 3 times a week i.p. The experiment was started atter two weeks of treatment. Error bars represent SEMs.

Young hamsters had been purchased at the age of 1 month, middle-aged as couples of ex-breeders at the age of 9 months. The young animals were kept one per cage, the middle-aged as the couples they had been during breeding. Couples that showed aggressive behavior between

2158

Memory and Longevity under Selcgiline

Vol. 55, No.s 25/26, 1994

the partners were seperated. All experiments described in the following were performed in accordance with the German law on animal rights (Tierschutzgesetz).

140 w "0

120

0 u 0

100

control • selegiline []

T

/~

80 6O u "" I,t --

40 20

0 L-naive

L-before

L-60"

L-24h

Fig. 2 'One way step through passive avoidance learning' of 23 mo. old female NMRI mice treated for three weeks as described in fig. 1.

Treatment Mice were treated subchronically with 0.25 mg selegiline per kg body weight three times a week intraperitoneally (control: n=13, selegiline: N=12). In the chronic hamster experiment food consumption and body weight were measured regularly. Starting at the age of 13 months the middle-aged ex-breeder hamsters were randomly divided into two groups (control, male: n=l 8; control, female: n=lg; selegiline, male: n=lS; selegiline, female: n=lS). The selegiline group received 0.05 mg selegiline per kg body weight daily in the food.

Behavioral testing Spatial learning of the mice was tested in a Morris water maze task (22) beginning after two weeks o f treatment with selegiline. The animals were placed in a round pool of 94 cm diameter. The pool was clockwisely divided into 4 quadrants and filled with water at a temperature of 21 °C. The water had been rendered opaque by adding milk and cacoa. An invisible platform was situated in the center of the first quadrant 8 cm from the wall. The mice were randomly placed into the water at the corresponding positions of the other three quadrants with their head directed to the wall. No starting postition was used twice in a row. The animals could search for the goal platform up to 60 seconds. Mice that did not find the platform were guided to it by the experimenter. The animals had to stay on the platform for 15 seconds before being removed to their home cage. The latency to reach the platform was registered with a stop watch and the distance traveled by a Coulburn Video Path Analyzer. The animals were tested twice a day with an intertest interval of 1 hour for five days. Passive avoidance learning of the mice was tested in a plastic box of 10x20x40 cm after three weeks o f treatment. A wall with a guilloutine door (5x5 cm) divided the box into an bright and a dark compartment. On day 1 o f passive avoidance learning the mice were habituated to the box for 2 minutes with the guilloutine door open. On day 2 the animal was placed into the bright

Vol. 55, No.s 25/26, 1994

Memory and Longevity under Selegiline

2159

compartment for 15 seconds before opening the door. When it had entered the the dark compartment with all four paws the naive latency (Lnaive) was registered. The guilloutine door was lowered then and the mouse stayed in the dark compartment for 15 seconds before it was moved back to its home cage. About 90 minutes later the animal was placed into the bright compartment again, the guilloutine door was opened 15 seconds later, and after clocking the latency (Lbefore) the door was lowered again. However, the mouse received a scrambled foot shock of 0.8 mA for 1 second immediately after the door was closed. 15 seconds later the animal was placed into a small (Macrolon type I) cage for 30 seconds. It was moved to the bright compartment again, the guilloutine door was opened after 15 seconds and the latency 60 seconds after shock (L60") was registered as a measure of short-term memory. The animal was removed from the dark compartment immediately after it had entered to minimize the unlearning of the shock experience. 24 hours later the after shock latency (L24h) was measured in the same way as the naive latency to give a measure of long-term memory. Spontaneous alternation behavior (SAB) was tested in the hamsters every three months using a T-maze with arms o f 10x40x20 cm. The animals were habituated to the maze once for 2 minutes with all doors of the maze open. Then they were habituated daily using the training-test procedure described below (interval between training and test: 15 seconds) until they showed no more stress-related behavior like pronounced defecation. If the daily habituation procedure increased the signs of stress-related behavior habituation was continued every second day. The training-test procedure proper started with three training sessions in which one "horizontal" arm of the T was closed. The animal was placed in a wooden box (10xl0xl0 cm) that was situated at the "vertical" arm of the T. After 15 seconds the guilloutine door to the maze was opened and the animal had the opportunity to go to the open "horizontal" arm. Immediately after entering the open "horizontal" arm the guilloutine door to it was closed and the animal had to stay there for a habituation time of 30 seconds. The hamster was guided to a wooden box at the end of the arm then. After the three training sessions and an interval of 15 seconds, 4 hours or 24 hours the hamster was placed in the T-maze with both "horizontal" arms open for the test session.

c 0 •-

100

p
90

c

80

•

70

-

=

~,,.f

.

.

.

.

.

.

.

.

.

.

.

.

.

i

60 0 o

c m

o

a.

50 40 30 20 10

b~

0 15"

4h

15"

24h

4 too.

4h

24h

17 too. Fig. 3

Spontaneous alternation behavior (SAB) of young (4 mo.) and aged (17 mo.) hamsters. Broken lines indicate the chance level of SAB. Comparisons were performed with the Chi-square test.

2160

Memory and Longevity under Selegiline

,- 1000 -~ 90. II E 80" ~e 70m Me -" o e f-

Vol. 55, No.s 25/26, 1994

p<0.035

60' 5040. 30.

on 2o. Me 10. N 0 15"

4h

24h

control

,.-

100-1

.0

90 '

15"

4h

24h

solegilino

p
80' 70" I° Me

" o e i

60. 504030.

oa, 2o. Me 10. N

0 15"

4h

24h

control

,.0 -~ lib

=

15"

4h

24h

solegiline

10090,

so' 70-

-~ m

•" o e C

so ' 504030-

C

o Q. Me IA

2O10o 15"

4h

24h

control

15"

4h

24h

selegiline Fig. 4

Spontaneous alternation behavior o f aged hamsters (a. 17 mo., b: 20 mo., c. 23 mo.) treated with 0.05 mg/kg selegiline daily in the food starting at the age o f 12 months vs. controls. Broken lines and statistics as in fig. 3.

Vol. 55, No.s 25/26, 1994

Memory and Longevity under Selegiline

2161

Spontaneous alternation took place if the animal entered the arm that had been closed before. After each test session the animal was placed into the T-maze with the guilloutine doors open for two minutes to extinguish the direction that had been trained before. SAB in a group was said to be found for the respective interval between training and test if the percentage of animals in the group showing SAB was above the two-tailed 5 % chance level of the binomial distribution for a probability of SAB being 0.5. The chronic treatment group was tested every three months. The young hamsters were tested once at the age of 4 months for an age comparison. Survival The hamsters in the chronic treatment study were checked twice daily for animals that had died of natural causes. Statistics All statistics were performed using SAS 6.03. Results Behavioral testing after subchronic treatment in mice Aged mice subchronically treated with selegiline neither showed an improvement in Morris water maze performance (fig. 1) nor in passive avoidance learning (fig. 2).

Behavioral testing after chronic treatment in hamsters Young Syrian hamsters exhibit SAB up to 4 hours after training while aged, 17 months old untreated hamsters showed SAB only 15 seconds after training (fig. 3). There is a clear-cut, highly significant age difference in the SAB 4 hours after training (p<0.001, Chi-square=7.78, 1 DF). Treated 17 months old hamsters, however, still featured SAB 4 hours after training (fig. 4a)

100-

•

t/J

o L--

:P °

i

:D L--

=1 f/J

50

control, n - 3 7 n-36

I O selegiline,

N

I

12

15

I

,

!

18 21 months

'

I

24

'

I

27

Fig. 5 Preliminary survival curves of hamsters treated with 0.05 mg/kg selegiline daily in the food starting at the age of 12 months vs. controls.

2162

Memory and Longevityunder Selegiline

Vol. 55, No.s 25/26, 1994

with a significant difference between treated animals and controls for the 4 hour interval (p<0.035, Chi-square=4.91, 1 DF). At the age of 20 months control hamsters had lost the ability for SAB even at the interval of 15 seconds (fig. 4b). Treated animals still showed SAB after the 15 second interval (fig. 4b) with a significant difference versus controls (p<0.05, Chi-square=4.17, IDF). However, no more SAB could be found for them after the 4 hour interval. At the age of 23 months even the treated group had lost the ability to show SAB (fig. 4c). No sex differences in SAB could be detected. Survival

Hamsters chronically treated with selegiline show a marked tendency to live longer than controls (fig. 5). Since the study is still in progress no final analysis of survival can be given at the present time. However, a preliminary survival analysis using the exponential distribution indicates significant differences between the survival curves (fig 5; p<0.0001, Chi-square=32.85, 1 DF). Discussion

Our behavioral data on the effects of a chronic treatment with selegiline suggest that this drug may slow down the progress of age-related deteriorations in cognitive performance. Control hamsters first loose longer-term memory in the SAB paradigm before showing no more SAB at all. Chronically treated Syrian hamsters also loose these abilities, but at a later time (fig. 4). One might argue that this delay of age-related cognitive decline may be due to a merely symptomatic improvement as described for rats after a short-term treatment (13). However, this is probably not a general feature of the drug, since in our experiments subchronically treated mice did no better than controls, although they had been treated with a dose comparable to the doses used in the short-term treatment study in the rat (13). The hamsters received a much lower-dose of selegiline making acute effects even more unlikely. The lack of a consistent acute effect of selegiline in aging rodents is further supported by the study of Knoll (15). He reported differences in cognitive performance between rats treated with 0.25 mg selegiline per kg three times a week and control animals no sooner than six months after starting the treatment. Therefore, it seems most likely that selegiline slows down cognitive decline due to normal aging. This effect may also be important for the beneficial effects in Alzheimer patients (23). However, additional neuroprotective effects may also be relevant. Our hypothesis of a slowing-down of the aging process is further supported by the survival curves of the hamsters in the chronic experiment (fig. 5). Since food intake and body weights of the hamster groups did not differ significantly an effect of dietary restriction that may also increase survival in hamsters (24) can be ruled out. The further progress of the study will show whether selegiline shifts the survival curve to the fight or whether it leads to a rectangularization of the survival curve. Only if the curve is be shifted to the right, can a slowing down of general aging processes then be assumed (21). A possible mechanisms of action for the postponement of age-related cognitive decline by selegiline may be an effect on polyamine metabolism (25). By inhibiting MAO-B the degradation of N-acetylated polyamines could be reduced. Since polyamines and their N-acetylated forms stimulate or inhibit the NMDA-receptor differentially (Hafner et al., unpublished findings) a shift in the ratio of polyamines and their N-acetylated forms could alter the neurotoxicity &glutamate (26). However, other mechanisms including the growth-factor like action of selegiline (16,27) or its effect on survival in immunodeficient mice (28) by a mechanism not yet identified might be similarly relevant. Further research is needed to uncover the basic mechanism(s) underlying the actions of selegiline on aging processes. References

1. A. AGNOLI, N. MARTUCCI, G. FABBR1NI, A.E. BUCKLEY and M. FIOAVANTI, Dementia ! 109-114 (1990).

Vol. 55, No.s 25/26, 1994

Memory and Longevityunder Selegitine

2163

2. C. LOEB and C. ALBANO, abstract from the European conference on Parkinson's disease and extrapyramidal disorders (1990); cited in P.N. TARIOT, L.S. SCHNEIDER, S.V. PATEL, B. GOLDSTEIN, Inhibitors ofMonoamine Oxidase B, I. Szelenyi (ed.), 301-317, Birkh/iuser Verlag, Basel (1993). 3. A. MANGONI, M.P. GRASSI, L. FRATOLLA, R. PIOLTI, S. BASSI, A. MOTTA, A. MARCONE and S. SMIRNE, Eur. Neurol. 31 100-107 (1991). 4. E. MARTIGNONI, G. BONO, F. BLANDINI, E. SINFORIANI, P. MERLO and G. NAPPI,J. Neural Transm. [P-DSect] 3 15-25 (1991). 5. P.J. RIEKKINEN, E.-L. HELKALA, T. H)UqNINEN, M. VANHANEN, H. AALTONEN and E. HEINONEN, abstract from the symposion Past, Present and Future of Parkinson's Disease and New Aspects in the Treatment of Dementia at the XVth World Congress of Gerontology (1993) 6. N. CAMPI, G. P. TODESCHINIand L. SCARZELLA, Ciin. Therap. 12 306-314 (1990). 7. A. MONTEVERDE, P. GNEMMI, F. ROSSI and A. MONTEVERDE, Clin. Therap. 12 315-322 (1990). 8. A. FALSAPERLA, P. PRETI and C. OLIANI, Clin. Therap. 12 376-384 (1990). 9. P.N. TARIOT, R.M. COHEN, T. SUNDERLAND, P.A. NEWHOUSE, D. YOUNT, A.M. MELLOW, H. WEINGARTNER, E.A. MUELLER and D.L. MURPHY, Arch. Gen. Psychiatry 44 427-433 (1987). 10. P.N. TARIOT; T. Sunderland, H. WEINGARTNER, D.L. MURPHY, J.A. WELKOWITZ, K. THOMPSON and R.M COHEN, Psychopharmacology 91 489-495 (1987). 11. G.L. PICCININ, G FINALI and M. PICC1RILLI, Clin. Neuropharm 13 147-163 (1990) 12. J. KNOLL, Acta Neurol. Scand. 126 83-91 (1989). 13. R. BRANDEIS; M. SAPIR, Y. KAPON, G. BORELLI, S. CADEL and B. VALSECCHI, Pharmacol. Biochem. Behav. 39 297-304 (1991). 14. A.J. BARBER, P.H. YU and A.A. BOULTON, Life Sci 53 739-747 (1993). 15. J. KNOLL, Inhibitors of Monoamine Oxidase B, I. Szelenyi (ed.), 145-167, Birkh~user Verlag, Basel (1993). 16. P.T. SALO and W.G. TATTON, J. Neurosci. Res. 31 394-400 (1992). 17. PARKINSON STUDY GROUP, N. Engl. J. Med. 328176-183 (1993). 18. J. KNOLL, Mech. Ageing Dev. 46237-262 (1988). 19. N.W. MILGRAM, R.G. RACINE, P. NELLIS, A. MENDONCA and G.O. IVY, Life Sci. 47415-420 (1990). 20. K. KITANI, S. KANAI, Y. SATO, M. OHTA, G.O. IVY and M.-C. CARILLO, Life Sci. 52 281-288 (1992). 21. L. HAYFLICK, Exp. Gerontol. 20145-159 (1985). 22. M.D. LINDNER, A.H. BALCH and C.P. VANDERMAELEN, Behav. Neur. Biol. 58 94-102 (1992). 23. K. WESSEL, Inhibitors ofMonoamine Oxidase B., I. Szelenyi (ed.), 253-275, Birkhauser Verlag, Basel,(1993,). 24. E. STUCHLIKOVA, M. JURICOV,/~-HO~OV,~ and Z. DEYL, Exp. Geronto110 141-144 (1975). 25. M. GERLACH, P. RIEDERER and M.B. YOUDIM, Eur. J. Pharmacol. 22697-108 (1992). 26. R.T. BARTUS, JAGS 3___88680-695 (1990). 27. W.G. TATTON, abstract from the symposion Past, Present and Future of Parkinson's Disease and New Aspects in the Treatment of Dementia at the XVth World Congress of Gerontology (1993) 28. H.-J. FREISLEBEN;, F. LEHR and J. FUCHS, J. Neural Transm. Supp141359-364 (1994).