Memory effect of caerulein and its analogs in active and passive avoidance responses in the rat

Peptides, Vol. 10, pp. 843-848. © Pergamon Press plc, 1989. Printed in the U.S.A.

0196-9781/89 $3.00 + .00

Memory Effect of Caerulein and Its Analogs in Active and Passive Avoidance Responses in the Rat SHINJI ITOH, AKIRA TAKASHIMA, KEN'ICHI IGANO AND KEN INOUYE

Shionogi Research Laboratories, Fukushima-ku, Osaka 553, Japan

R e c e i v e d 20 D e c e m b e r 1988

ITOH, S., A. TAKASHIMA, K. IGANO AND K. INOUYE. Memory effect of caerulein and its analogs in active and passive avoidance responses in the rat. PEPTIDES 10(4) 843-848, 1989.--Tbe memory effects of caerulein (CER) and its analogs ([des-Gln2]-CER and [LeuS,Nlea]-CER) were compared with that of cholecystokinin octapeptide (CCK-8) using active and passive avoidance responses in rats. In the active avoidance test, single subcutaneous (SC) injection of CER and its analogs immediately after the learning trials at doses of 10 and 100 ng/kg prevented extinction of learned task for 10 days, and at a dose of 1000 ng/kg for at least 15 days, but the effect of CCK-8 was somewhat weaker. In the saline control group, the number of responses decreased after 5 days. In the passive avoidance response, electroconvulsive shock (ECS)-induced amnesia was partially prevented by CCK-8 at doses of 100 and 1000 ng/kg SC, while CER and its analogs at doses of more than 100 ng/kg totally prevented the ECS-induced amnesia. Intraperitoneal administration of scopolamine caused complete amnesia which was also partially prevented by CCK-8, while CER could totally prevent the amnesia following SC injection of 2 p.g/kg. These results indicate that CER and its analogs are more effective than CCK-8 for preventing experimental amnesia. Caerulein

Caerulein analog

Active avoidance response

Passive avoidance response

Memory

METHOD

CAERULEIN (CER) is chemically closely related to cholecystokinin octapeptide (CCK-8), and both peptides have been shown to elicit a variety of central actions following either intracerebroventricular or peripheral administration. These peptides reduce food intake and spontaneous exploratory activity, and cause ptosis and hypothermia. The peptides potentiate the action of barbiturates, suppress motility changes induced by apomorphine, excitations induced by TRH and methamphetamine, convulsions caused by picrotoxine, harman, thiosemicarbazide and isoniazid, and tremors produced by oxotremorine, harmine and ibogaine. Although these actions of CER are similar to those of CCK-8, CER is generally more potent than CCK-8 [cf. (21)]. Recently, we found that CCK-8 could prevent experimental amnesia in a one-trial passive avoidance paradigm (12,13) and prevent extinction of the learned task in an active avoidance response in the rat (10). Similar observations were reported by other investigators (5,6). In addition, the anticholinergic agent, scopolamine, induced a marked amnesia which could be partially prevented by pretreatment of rats with CCK-8 (l l). Thus, we thought it of interest to examine the effect of CER on memory processes in comparison with that of CCK-8. In the present study, CER and its analogs were found to possess markedly potent actions for preventing experimental amnesia and enhancing the performance of a learned task for a long period of time following a single peripheral administration.

Male Wistar rats (n = 627), weighing approximately 250 g, were housed at a constant room temperature of 25 +- I°C with 12-hr light-dark cycles, light on at 0700 hr, and maintained with free access to standard rat biscuits and water. All experiments were carried out between 0900 and 1500 hr.

Active Avoidance Response A platform jumping active avoidance behavior was tested by the method of Gamzu (9) with some modification. Before initiation of a series of trials, the rats were allowed to become accustomed to the experimental box (40 × 35 cm and 40 cm high) for l0 min, and then placed on a platform 0 8 x l0 cm) which was located I 1 cm above the floor in one comer of the box. At the beginning of each trial, a buzzer was sounded as a conditioned stimulus, then the rat was dropped from the platform onto the grid floor by hand. The buzzer continued for 15 sec and was followed by a 2-mA foot-shock for 15 sec. The rat could avoid or escape this by jumping onto the platform. A 10-sec safety period was set between the shock termination and the onset of the next trial. Each rat received two 5-trial sessions approximately 4 hr apart, one in the morning and one in the afternoon. Immediately after 10 trials, test peptide or physiological saline solution was injected subcutaneously (SC). Similar 10-trial sessions without foot-shock were repeated on the next day (day l) and 5, 10, and 15 days later. The

843

844

ITOH, TAKASHIMA, IGANO AND INOUYE

TABLE I COMPOUNDS USED IN THE PRESENT STUDY

1 2 3 4

CCK-8 Caerulein (CER) Des-GIn-~-CER [LeuS,NleS]-CER

tests were conducted 10 times per day and the number of avoidance responses out of 10 was recorded. The total number of avoidance responses of the four sessions are also indicated in the tables.

Passive Avoidance Response Passive avoidance response was tested in a one-trial learning paradigm. The apparatus consisted of a dark box (45 x 40 cm and 35 cm high) with black walls and a grid floor through which electric foot-shock could be introduced. A runway (14 x 30 cm and 12 cm high) was attached to the dark box, which extended from an opening (14 x 12 cm) in the center of one wall of the dark box. The shock chamber was kept dark, and a 60-W lamp was fixed 60 cm above the center of the runway on which the rat was first placed. The latency for entering the dark box was generally a few seconds. One trial was given on day 1, and three on day 2. Immediately after the third trial on day 2, the rat received a single inescapable scrambled foot-shock (0.2 mA, 2 sec). After this learning trial, the rat was removed from the apparatus. Memory retention was tested 24 hr after the learning trial. The rat was placed on the runway, and the latency of entering the dark box was measured up to the maximum of 300 sec. Amnesia was induced by electroconvulsive shock (ECS), by passing a 60 mA current for 0.4 sec through alligator clips attached to the ears immediately after termination of the footshock. The peptide solution or saline was injected SC soon after the ECS. In another experiment, amnesia was induced by scopolamine. CER was first injected SC at 30 min before the learning trials, and 15 min later scopolamine was given intraperitoneally (IP) at a dose of 0.5 mg/kg. Thereafter, the passive avoidance response was tested as described above.

Hot Plate Test Following IP injection of scopolamine at a dose of 0.5 mg/kg, rats were placed on a hot stainless steel plate maintained at 57 --- I°C. The nociceptive response was characterized by vigorous licking of the paws, jumping up and squealing. In the present study, the jumping up latency was measured because of its distinctness.

Drugs Peptides used in the present experiments were CCK-8 (Protein Res. Found., Osaka), CER, [des-Gin2]-CER and [LeuS,NleS] CER (Table 1). CER and its analogs were synthesized in our laboratories (8). The purity of these peptides was more than 99% as estimated by TLC and HPLC. The peptides were dissolved in physiological saline solution and diluted to appropriate concentrations. They were injected SC in doses from 1-1000 ng/kg, as noted in the text. Scopolamine hydrobromide (Wako Pure Chemical Indust., Osaka) and 13-endorphin (Protein Res. Found., Osaka) were also dissolved and diluted in the saline solution.

Analysis of Data The data of active avoidance response on each day after

H-Asp-Tyr(SO3H)-Met-Gly-Trp-Met-Asp-Phc-N H: Pyr-Gln-Asp-Tyr(SO3H)-Thr-Gly-Trp-Met-Asp-Phe-NH z Pyr----Asp-Tyr(SO3H)-Thr-Gly-Trp-Met-Asp-Phe-NH2 Pyr-Gln-Asp-Tyr(SO3H)-Leu-Gly-Trp-Nle-Asp-Phe-NH:

learning trials were analyzed by the multiple comparison of Dunnett following the analysis of variance (ANOVA), and the differences between the number of learning trials and those of successive test days in each rat were calculated and the rate of extinction in different dose groups were analyzed by the paired t-test. The data of passive avoidance response were analyzed by the nonparametric ranking test of Kruskat-Wallis and by the Mann-Whitney U-test. RESULTS

Effect on Active Avoidance Response When saline solution was injected SC immediately after the learning trials, on the next day (day 1) there was no difference in the number of responses compared with the value of learning trials, but extinction of the avoidance task occurred on day 5, and markedly on days 10 and 15. When CCK-8 was given at a dose of 1 ng/kg (approximately 0.25 ng per rat), the number of responses was not different from the level of the saline control, and extinction of learned task occurred on days 10 and 15. Administration of CCK-8 in doses of 10 ng/kg or more caused significant increase in the number of responses on day 1. and CCK-8 at doses of 10 and 100 ng/kg did not cause extinction of the learned task on days 5 and 10. Administration of I000 ng/kg was more effective to prevent the extinction. When these data after CCK-8 administration were compared with the saline control, as shown in Table 2, the preventive effect of CCK-8 on extinction was dose-dependent. The total scores of four sessions were also higher in rats receiving 100 and 1000 ng/kg than the control group. The effects of CER and its analogs were similar to that of CCK-8. However, increases in the number of responses on day 1 were apparently observed following CER administration in doses from 1 to 1000 ng/kg, while other CER analogs caused such an increase only at a dose of 1000 ng/kg. The extinction of learned task occurred in rats receiving CER or its analogs at a dose of I ng/kg on days 10 and 15. However, resistance to extinction was observed in rats receiving CER in doses from 10 to 1000 ng/kg on days 5 and 10, and 1000 ng/kg of CER on days 5, 10 and 15, in rats receiving [des-Gln2]-CER in a dose of 10 ng/kg on day 10, in doses 100 and 1000 ng/kg on days 5, 10 and 15. and in rats receiving [LeuS,Nles]-CER in a dose of 10 ng/kg on days 10 and 15, and in doses of 100 and 1000 ng/kg on days 5, 10 and 15. The results indicate that these peptides increase the resistance to extinction of learned task in a dose-related manner. The effects of CER are shown in Table 3, [des-Gln2]-CER in Table 4, and [Leu ~, NleS]-CER in Table 5.

Effect on ECS-lnduced Amnesia in Passive Avoidance Response In our previous study, ECS-induced amnesia was found to be effectively prevented by CCK-8 (12). Thus, in the present study, the effects of CER and its analogs on the prevention of ECS-

MEMORY EFFECT OF CAERULEIN

845

TABLE 2 EFFECT OF SC INJECTION OF CCK-8 ON PLATFORM JUMPING ACTIVE AVOIDANCE RESPONSE Number of Avoidance

Peptide Saline CCK-8

Dose (ng)

No. of Rats

I 10 1130 1000

10 12 10 10 10

Learning Trial F=0.56 df= 4.47 p>0.05 6.2 6.8 6.3 6.4 6.4

Day 1 F=0.38 df= 4,47 p>0.05

± 0.3 --+ 0.2 ± 0.3 ± 0.2 ___ 0.3

6.8 8.1 7.7 7.8 8.0

Day 5 F=2.94 df= 4,47 p<0.05

: 0.9 ± 0.7 ± 0.7.~ _+. 0.8+ ± 1.0:~

3.0 4.9 5.9 7.1 7.1

Day 10 F=3.65 df= 4,47 p<0.05

± 0.6§

1.6 ± 0.7§

-+- 1.2 +-- 0.9 ± 0.8* ± 1.0"

3.0 4.2 6.3 6.4

± -+± ±

1.1§ 1.3 1.0" 1.2"

Day 15 F = 2.79 df= 4,47 p<0.05 0.9 2.0 3.1 4.0 5.3

+-- 0.2§ ± 0.9§ --- 1.35 ± 1.0:]: +-- 1.3"

Total F=2.97 df= 4,47 p<0.05 12.4 18.0 20.7 25.2 26.0

-+ 2.1 ± 3.2 -± 3.7 +-- 3.2* - 4.1"

Peptide was injected immediately after learning trials. *p<0.05, tp<0.01 vs. saline control (Dunnett's test); 1:p<0.05, §p<0.01 vs. learning trials (paired t-test).

C C K - 8 injected in d o s e s o f 1 and 10 ng/kg SC i m m e d i a t e l y after the ECS treatment did not affect the latency. H o w e v e r , injection o f C C K - 8 in d o s e s o f 100 and 1000 n g / k g increased it significantly, the m e d i a n latencies b e i n g 83 sec and 113 sec, respectively. T h e results o f the K r u s k a l - W a l l i s test were H = 17.50

induced a m n e s i a were c o m p a r e d with those o f C C K - 8 . A s reported p r e v i o u s l y , ECS produced c o m p l e t e m e m o r y deficit in the 24-hr retention test o f the passive a v o i d a n c e response o f the rat, the latency o f entering the dark box b e c o m i n g only a few seconds.

TABLE 3 EFFECT OF SC INJECTION OF CAERULEIN ON PLATFORM JUMPING ACTIVE AVOIDANCE RESPONSE Number of Avoidance

Peptide Saline Caerulein

Dose No. of (ng) Rats

I 10 100 1000

16 16 16 16 16

Learning Trial F=0.17 df=4,75 p>0.05 7.0 6.8 6.8 6.8 6.9

± 0.2 +-- 0.3 -+- 0.2 - 0.2 ± 0.2

Day 1 F = 1.56 df=4,75 p>0.05 7.1 8.3 8.8 8.8 8.9

= : ± ± ±

Day 5 F=6.17 df=4,75 p<0.01

0.6 0.5§ 0.7§ 0.7§ 0.6~.

2.9 5.3 6.8 7.6 7.9

± 0.7§

± 0.8 ±- 0.9t -+- 0.9t _+. 0.8+

Day 10 F = 6.75 df=4,75 p<0.01 1.4 3.6 5.3 6.5 6.4

Day 15 F=3.73 df=4,75 p<0.01

+-- 0.5§ 1.3 ± 0.5§ -+- 0.9§ 1.4 ± 0.6§ ± 0.9t 2.7 + 0.6§ ± 0.9,~ 3.6 ± 0.9.~ ± 0.9+ 4.6 _ 0.9+:[:

Total F = 6.36 df=4,75 p<0.01 12.6 ± 1.9

18.5 23.6 26.4 27.9

± 2.5 ±- 2.5* ± 2.8,~ _+ 2.7+

Peptide was injected immediately after learning trials. *p<0.05, *p<0.01 vs. saline control (Dunnett's test); ~:p<0.05, §p<0.01 vs. learning trials (paired t-test).

TABLE 4 EFFECT OF SC INJECTION OF Ides-Gln2]-CER ON PLATFORM JUMPING ACTIVE AVOIDANCE RESPONSE Number of Avoidance Learning Trial F = 1.67 Peptide Saline [des-Gln2]-CER

Dose No. of (ng) Rats

1 10 100 1000

16 12 12 12 12

df=4,59 p>0.05 7.0 7.1 6.7 6.3 6.7

-'4-± --±

0.2 0.2 0.3 0.2 0.3

Day I F=0.66 df=4,59 p>0.05 7.1 7.8 8.1 7.8 8.5

± + -+ --±

0.6 0.7 0.7 0.8 0.5§

Day 5 F=4.18 df=4,59 p<0.01 2.9 5.6 6.1 6.8 8.0

± 0.7§ -+- 1.1 +-- 1.0 --- 1.2" -'- 1.0t

Day 10 F=7.29 df=4,59 p<0.01 1.4 3.1 5.2 6.1 6.9

--+ 0.5§ -+- 0.8§ -+- 1.1+ -+- 1.0t" ± 1.0+

Day 15 F=3.54 df=4,59 p<0.05 1.3 2.5 3.6 4.8 5.4

± 0.5§ -+- 0.8§ -'- 1.0~: --_ 1.1" -+- 0.9t

Total F=4.66 df=4,59 p<0.01 12.6 19.0 23.2 25.3 28.6

± 1.9 ± 2.8 +__ 3.2* _+_ 4.0* -± 3.0t

Peptide was injected immediately after learning trials. *p<0.05, t0.01 vs. saline control (Dunnett's test); $p<0.05, §p<0.01 vs. learning trials (paired t-test).

846

ITOH, TAKASHIMA, IGANO AND INOUYF

TABLE 5 EFFECT OF SC INJECTION OF [LeuS,NIe~]-CER ON PLATFORM JUMPING ACTIVE AVOIDANCE RESPONSE

Number of Avoidance Learning Trial F=0.16 df=4,59 p>0.05

Dose No. of (ng) Rats

Peptide Saline [LeuS,NleS]-CER

I 10 100 I000

16 12 12 12 12

7.0 7.1 7.0 6.8 7.1

Day I F= 1 . 0 8 df=4,59 p>0.05

_+ 0.2 ± 0.2 _+_0.2 -_+ 0.3 ± 0.3

7.1 7.6 8.2 8.2 8.8

± ± ± ±

0.6 0.7 0.8 0.5 0.7.+

Day 5 F=5.92 df=4,59 p<0.01 2.9 5.6 6.5 7.4 8.1

--_ 0.7§ ± 1.1 ± 1.0" ± 0.8t --- 0.7t

Day 10 F=8.19 df=4.59 p<0.01 1.4 3.6 5.3 6.9 7.2

± 0.5§

__. 1.23; = 0.8~ _+_0.9t _+ 0.9t

Day 15 F= 3.69 df=4,59 p<0.01 1.3 2.4 3.3 4.8 4.8

Total F= 5.92 df=4,59 p<0.01

-+ 0.5§ 12.6 z 0.9§ 19.1 ± 0.8§ 23.3 -+- l . l t 27.3 -+ 0.9~:1:28.5

~ -+ ± ± ±

1.9 3.7 2.9* 2.9,* 2.5,*

Peptide was injected immediately after learning trials. *p<0.05, "l'p<0.01 vs. saline control (Dunnen's test); :t:p<0.05, §p<0.01 vs. learning trials (paired t-test).

(p<0.01). These data are shown in Fig. 1, together with the results of CER and its analogs. Administration of CER and its analogs had a significant preventive effect on ECS-induced amnesia even at a small dose of 10 ng/kg. When these peptides were given in doses of I00 and 1000 ng/kg SC, the effect was surprisingly marked, the median latency values always being the maximum level of 300 sec, as shown in Fig. 1. The effects of Kruskal-Wallis test were H = 20.93, (p<0.001) for CER, H =20.86 (p<0.001) for [des-Gin2] CER, and H = 25.01 (p<0.001) for [LeuS,NleS]-CER. Thus, CER and its analogs are clearly more potent than CCK-8 for preventing ECS-induced amnesia in the passive avoidance response of the rat.

Effect of Scopolamine-Induced Amnesia in Passive Avoidance Response Scopolamine causes pronounced memory deficit in rodents and human. Our previous study indicated that in a passive avoidance response, IP administration of scopolamine in a dose of 0.5 mg/kg

30C

,~ ~.

!l

ccK-8

at 15 min before the training trials produced marked memory impairment in the rat. However, pretreatment with CCK-8 prevented the amnesia (11). CER is expected also to prevent the amnesia induced by scopolamine. Similar to our previous experiment, different doses of CER from 0.1 to 10 I~g/kg were injected SC at 30 min before the learning trials and 15 min later scopolamine was given IP in a dose of 0.5 mg/kg. The results of the 24-hr retention test are shown in Fig. 2 (Kruskal-Wallis test, H = 24.93, p<0.001). The dose-response curve of the median latencies in groups receiving different doses of CER was an inverted U-shape. The most effective dose was 2 Ixg/kg and the median latency at this dose was the same level as that observed in the saline control without scopolamine. The results indicate that the scopolamineinduced amnesia was totally prevented by this dose of CER. In this figure, the median latencies following different doses of CCK-8 which had been reported in our previous paper (11) are indicated as closed circles. The maximal effective dose of CCK-8 was 10 i~g/kg. This difference clearly indicates that CER is more potent than CCK-8 for preventing the amnesia.

Effect of Scopolamine on Nociception

c..,o,.~ In : :

100

There might be a question whether scopolamine itself could affect nociception and consequently reduce the latency of passive avoidance. In order to examine this possibility, the effect of scopolamine on pain perception was tested on a hot plate under the same time schedule as the above experiment. Rats were treated with CER or saline 30 rain before the test, and scopolamine in a

..L. m o

300 ~Des- Gin -CER

K

TABLE 6

~ to,us. N~aJ-CE~

EFFECT OF CAERULEIN (CER, SC)o SCOPOLAMINE (SCP, IP) AND 13-ENDORPHIN (~-END, IP) ON NOCICEPTION IN HOT PLATE TEST

200

Treatment

,o0 SC No. of rats 20

15

14

15

15

20

15

15

18

Dose, ng/kg Saline

1

10

100

1000

Saline

1

10

100 1000

IP

No. of Rats

Saline SCP 0.5 mg/kg Saline SCP 0.5 rng/kg 13-END 0.5 mg/kg

10 I0 10 10 10

Latency Sec

16

FIG. ]. Preventive affects of CCK-8, caerul¢in (CER), [des-G[n2]-CER and [LeuS,NleS]-CER on ECS-induced amnesia in passive avoidance response. Median latencies in sec and the ranges between the 25--75th percentiles (vertical lines) are shown. *p
Saline Saline CER 1 p.g/kg CER 1 O,g/kg Saline

4 4 5 4 t6

Mean +_ SEM. */7<0.01 vs. saline + saline (Dunnett's test).

--+ 0.3 ± 0.5 ± 1.3 ± 0.4 ± 3.2*

MEMORY EFFECT OF CAERULEIN

300

847

I

200

Caerulein (p~lkg, SC ) No. of rats

SalVe

0

0.1

0.5

12

12

12

2

5

10

Scopolamine 0.5 mglkg. IP ~

I 12

1

12

13

13

I

I

20

50 1

12

FIG. 2. Preventive effect of caerulein (CER) on scopolamine-induced amnesia in passive avoidance response. Median latencies in sec and the ranges betweenthe 25-75th percentiles(vertical lines) are shown. Closed circles are the medianlatenciesfrom a similarexperimenton CCK-8 which was reported in our previous paper (11). *p<0.02, **p<0.002 vs. scopolamine alone without CER.

dose of 0.5 mg/kg was injected IP 15 min before the hot plate test and the latency of jumping up was observed. As shown in Table 5, scopolamine had no effect, though an endogenous opioid, 13endorphin, in a dose of 0.5 mg/kg IP, caused significantly long latency. Thus, the above noted possibility would be excluded. DISCUSSION

The effect of CCK-8 on the latency of passive avoidance response in the rat was first reported by Fekete et al. (6,7), who indicated that IP administration of 400 nM (457 Izg)/kg of CCK-8 immediately after the learning trials increased the latency, and that ICV injection of CCK antiserum caused a shorter latency, suggesting that the endogenous CCK in the brain might be a modulator of memory processes. On the other hand, Cohen et al. (2) reported that IP administration of CCK-8 in doses of 20-3840 izg/kg reduced the conditioned avoidance response in a dosedependent manner, and they noted that CCK-8 facilitates extinction of the avoidance. Such controversial results may be due to different experimental procedures and dosages of the peptide. The CCK-8 content in the whole brain of the rat from a report of Beinfeld et al. (1) is estimated to be approximately 550 ng. Thus, the results obtained by injection of such large amounts of CCK-8 may not be physiological effects. When we used CCK-8 in doses of 1-1000 ng (ICV) or 0.1-10 ixg/kg (SC) in the test of passive avoidance of the rat (12,13), we found that the administration of CCK-8 even in such small doses could prevent experimental amnesia. We also found that SC injection of CCK-8 in doses of 10 and 100 ng/kg was most effective for preventing extinction of the learned task in an active avoidance response of the rat (10). Furthermore, the effect of CCK-8 for preventing scopolamine-induced amnesia was an inverted U-shape curve and SC injection of 10 izg/kg of CCK-8 caused the most potent effect (11). Other researchers reported that CCK-8 produced a reduction of the passive avoidance latency at doses from 30 to 500 Ixg/kg IP (4) and that [P injection of CCK-8 at doses of 100 and 200 Ixg/kg did not cause a significant effect on the 24-hr retention test of passive avoidance latency, though longer latencies were observed in the 48-hr test (5). If they had used smaller doses of CCK-8, different results might have been obtained. Our present study indicated that CCK-8 at a dose of 119(2ng, but not 100 Ixg, increased the latency of the 24-hr retention test. From the pharmacological viewpoint, large doses of neuropeptides cannot elucidate their physiological

roles in the central functions, since the exogenous neuropeptides may influence other neurohumoral systems which can cause different responses. When the memory effect of CER was compared with that of CCK-8, CER was always more effective in both active and passive avoidance responses. In the passive avoidance response, CCK-8 at doses of 100 ng/kg or more partially prevented ECS-induced amnesia, while CER and its analogs completely prevented it at doses of 100 and 1000 ng/kg. When the median latencies were compared, the value after 100 ng/kg of CCK-8 was similar to that of 10 ng/kg of CER and its analogs. Furthermore, scopolamineinduced amnesia was totally prevented by 2 p,g/kg of CER, but the same dose of CCK-8 had no effect. In the active avoidance test, CCK-8 at a dose of 100 ng/kg or more prevented extinction of the learned task on days 5 and 10, while a similar effect was obtained by l0 ng/kg of CER. Thus, the memory effect of CER and its analogs is nearly 10 times more potent than that of CCK-8. As to the experiment on scopolamine-inducedamnesia, scopolamine was injected IP 15 min before the training trials. This procedure may raise a question whether scopolamine affects pain perception for the foot-shock, or not. Accordingly, the effect of scopolamine on the pain perception was tested using a hot plate method. The results indicated that scopolamine may not affect the pain perception. It is surprising that both CER and CCK-8 have long-lasting central actions for more than 2 weeks following a single SC administration. Meek et al. (15) reported that the turnover of CCK-8 in the cerebral cortex of the rat is considerably slower (half life= 16 hr) than that of the biogenic amines and amino acid transmitters (half life = less than 4 hr). However, the long-lasting central effects of these peptides could not be explained simply by their long half-lives. They may cause some substantial neurochemical changes in the central neurons, but at present, no evidence is available for this. We first found that some symptoms of chronic schizophrenic patients could be improved by intramuscular injection of CER at a dose of 0.3 or 0.6 txg/kg, and the beneficial effects lasted for 2-3 weeks after a single injection (16). This finding was soon confirmed by others (17). Although the results of clinical application of CER have been confusing (18), such long-lasting effects of CER were also observed in the rat (14). In the present study, CER or its analogs were given, the number of responses in active avoidance increased on day 1, and thereafter it did not significantlydecrease, compared with the value of learning trials, throughout the period of observation for 15 days, while in the saline control group, the number decreased 5 days after the learning trials. The results suggest that CER and related peptides are effective not only for the acquisition of new information but also for consolidation of the memory. In passive avoidance, CER and its analogs were markedly effective for preventing experimental amnesia. According to the results, these peptides are likely to enhance memory processes, since the amnesia was totally prevented by CER and its analogs, and the response (median latencies were up to 300 sec) increased beyond that of the saline controls without ECS treatment (median latencies were usually 200 sec). In relation to this, biochemical events induced by amnesic treatments need to be elucidated. We have previously reported that the scopolamine-induced decrease in the acetylcholine content in the frontal and temporal cortices and the hippocampus could be prevented by CCK-8 in a dose-related manner (19). Such experiments should be carried out employing CER. This may be more effective for preventing the decrease of acetylcholine content in these brain regions which are thought to be involved in memory processes. In general, the central actions of CER are more potent than those of CCK-8. The structural difference between CCK-8 and CER is, as shown in Table I, in the N-terminus amino acids; CER

848

ITOH, TAKASHIMA, IGANO AND INOUYE

has Pyr-Gln-, while CCK-8 does not have them. Since the activity of [des-Gln2]-CER was the same as that of the parent molecule, glutamine at position 2 may not have particular significance in CER action. Moreover, in CER molecule, substitution of threonine at position 5 by leucine and that of methionine at position 8 by norleucine did not affect the memory effect. The potent memory effect of CER compared with that of CCK-8 might be attributed to the N-terminus pyroglutamate, which not only protects its degradation by endogenous amino-peptidases but also stimulates the N-methyl-D-aspartate (NMDA) receptors of neurons involved in learning and memory. The excitatory amino acids, most notably L-glutamate and L-aspartate, are considered to participate in long-term potentiation (LTP) of hippocampal syn-

apses in learning and memory circuits [cf. (3,20)]. The pyroglutamate of CER and aspartate of CCK-8 may affect the NMDA receptor activity as chemical signals, and consequently to enhance memory processes. This possibility should be clarified in further investigations, including electrophysioiogical method. Although various problems remain to be studied in the central actions of CER, this peptide shows promise for clinical application because of its efficacy by peripheral administration in doses of less than 1 p,g/kg. ACKNOWLEDGEMENT We thank Ms. Y. Maeda for her expert technical assistance.

REFERENCES 1. Beinfeld, M. G.; Meyer, D. K.; Eskay, R. L.; Jensen, R. T.; Brownstein, N. J. The distribution of cholecystokinin immunoreactivity in the central nervous system of the rat as determined by radioimmunoassay. Brain Res. 212:51-57; 1981. 2. Cohen, S. L.; Knight, M.; Tamminga, A.; Chase, T. N. Cholecystokinin-octapeptide effects on conditioned avoidance behavior, stereotypy and catalepsy. Eur. J. Pharmacol. 83:213-222; 1982. 3. Collingridge, G. L.; Bliss, T. V. P. NMDA receptors--their role in long-term potentiation. Trends Neurosci. 10:288-293; 1987. 4. Deupree, D.; Hsiao, S. Cholecystokinin octapeptide, proglumide, and passive avoidance in rats. Peptides 8:25-28; 1987. 5. Deupree, D.; Hsiao, S. Cholecystokinin octapeptide increases passive avoidance latencies in rats. Physiol. Behav. 42:203-205; 1988. 6. Fekete, M.; K~td~, T.; Penke, B.; Telegdy, G. Modulation of passive avoidance behavior by cholecystokinin octapeptide in rats. Neuropeptides 1:301-307; 1981. 7. Fekete, M.; Lonovics, J.; Telegdy, G. Modulation of passive avoidance behavior of rats by intracerebroventricular administration of cholecystokinin antiserum. Neuropeptides 1:363-369; 198 I. 8. Fujimoto, M.; lgano, K.; Watanabe, K.; Irie, I.; Inouye, K.; Okabayashi, T. Effects of caemlein-related peptides on cholecystokinin receptor bindings in brain and pancreas. Biochem. Pharmacol. 34:1103-1107; 1983. 9. Gamzu, E. Animal behavioral models in the discovery of compounds to treat memory disfunction. Ann. NY Acad. Sci. 444:370--393; 1985. 10. ltoh, S.; Katsuura, G. Cholecystokinin octapeptide prevents extinction of active avoidance behavior in the rat. Drug Dev. Res. 10:171-175; 1987. 11. Itoh, S.; Takashima, A.; Katsuura, G. Preventive effect of cholecys-

12. 13. 14. 15. 16. 17. 18. 19.

20. 21.

tokinin octapeptide on scopolamine-induced memory impairment in the rat. Drug. Dev. Res. 12:63-70; 1988. Katsuura, G.; Itoh, 8. Preventive effect of cholecystokinin octapeptide on experimental amnesia in rats. Peptides 7:105-110; 1986. Katsuura, G.; ltoh, S. Prevention of experimental amnesia by peripherally administered cholecystokinin octapeptide in the rat. Drug Dev. Res. 7:269--276; 1986. Matsubara, K.; Matsushita, A. Long-lasting reduction of amphetamine-induced hyperactivity in rats after combined admires'tration of caerulein with haloperidol. Eur. J. Pharmacol. 101:157-158; 1984. Meek, J. L.; Iadarola, M. J.; Giorgi, O. Cholecystokinin turn-over in brain. Brain Res. 276:375-378; 1983. Moroji, T.; Watanabe, N.; Anki, N.; Itoh, S. Antipsychotic effects of caerulein, a decapeptide chemically related to cholecystokinin octapeptide, on schizophrenia. Int. Pharmacopsychiat. 17:255-273; 1982. Nair, N. P. V.; Bloom, D. M.; Nestoros, J. N. Cholecystokinin appears to have antipsychotic properties. Prog. Neuropsychopharmacol. Biol. Psychiatry 6:509-512; 1982. Nair, N. P. V.; Lal, S.; Bloom, D. M. Cholecystokinin and schizophrenia. Prog. Brain Res. 65:237-258; 1986. Takashima, A.; Itoh, S. Prevention of scopolamine effect on the decrease of acetylcholine content by peripherally administered cholecystokinin octapeptide in some regions of the rat brain. Drug Dev. Res. 14:67-74; 1988. Watkins, J. C.; Evans, R. H. Excitatory amino acid transmitters. Annu. Rev. Pharmacol. Toxicol. 21:165-204; 1987. Zetler, G. Caerulein and its analogues: neuropharmacological properties. Peptides 6(Suppl. 3):33--46; 1985.