Epilepsiu, 33(6):994-1000, 1992 Raven Press, Ltd., New York 0 International League Against Epilepsy
Anticonvulsant Effect of DN-1417, a Derivative of Thyrotropin-Releasing Hormone, and Liposome-Entrapped DN- 1417, on Amygdaloid-Kindled Rats Norio Mori and Toshihiko Fukatsu Department of Neuropsychiatry, Fukushima Medical College, Fukushima, Japan
Summary: The effects of y-butyrolactone-y-carbony1-Lhistidyl-L-propinamide citrate (DN-1417), a derivative of thyrotropin-releasing hormone, and liposome-entrapped DN-1417 (DN-L) were examined in amygdaloid-kindled rats. The animals were tested 2 h after intraperitoneal (i.p.) drug administration and then again every 24 h without further drug treatment. DN-1417 did not suppress the kindled seizure at 2 h but did beginning 1-6 days after injection. DN-L suppressed the kindled seizure at 2 h and had a more prolonged anticonvulsant effect than DN-
1417. After liposomes were given i.p. once daily for 2 weeks, there was no morphologic evidence that liposomes damaged brain neurons. These results, together with previously published data, suggest that as drug delivery vehicles, liposomes can enhance the effectiveness of drugs in the CNS without producing overt brain damage. Key Words: Liposomes-Thyrotropin-releasing hormone-y-Butyrolactone-y-carbonyl-L-histidyl-L-propinamide citrate (DN- 1417)--Amygdala-Kindling.
Thyrotropin-releasing hormone (TRH) is widely distributed in the brain (Hokfelt et al., 1975) and has been reported to have various central effects unrelated to endocrine action through the pituitarythyroid axis (Nemeroff et al., 1975). y-Butyrolactone-y-carbonyl-L-histidyl-L-propinamide citrate (DN-1417), an analogue of TRH, has more profound central activity than TRH, with only - v 3 0 of the thyroid-stimulating hormone-releasing activity (Fukuda et al., 1980). Clinical studies have shown that DN-1417 has beneficial effects on various aspects of degenerative myoclonus epilepsy, LennoxGastaut syndrome, and infantile spasms (Inanaga and Inoue, 1981; Ueda et al., 1983; Matsumoto et al., 1987; Inanaga et al., 1989). The anticonvulsant effect of intravenous (i.v.) DN-1417 in amygdaloid (AM)-kindled cats was not dose-related, however, and there was considerable interindividual variation (Sato et al., 1984), whereas intracerebroventricular (i.c.v.) DN-1417 had a dose-dependent anticonvulsant effect in AM-kindled rats (Sato et al., 1985). In the Senegalese baboon, Pupio pupio, i.v.
DN-1417 had no effect on photosensitivity or cortically kindled seizures, whereas i.c.v. injection had a significant effect (Sakai et al., 1991). Therefore, the blood-brain barrier (BBB) appears to pose a serious obstacle to DN-1417. Liposomes, microscopic vesicles composed of one or several lipid membranes surrounding discrete aqueous compartments, can encapsulate water-soluble chemicals in their aqueous spaces. Because liposomes can enhance the effectiveness of biologically active materials or reduce toxicity or both, they may be useful as drug delivery vehicles (Gregoriadis et al., 1976; Ostro and Cullis, 1989; Rande, 1989). Recent studies have demonstrated liposome-mediated brain uptake of BBB-impermeable substances, including P-galactosidase (Takada et al., 1982), superoxide dismutase (SOD) (Yusa et al., 1984; Chan et al., 1987), and y-aminobutyric acid (GABA) (Loeb et al., 1982, 1986). Indeed, intraperitoneal (i.p.) injection of liposomeentrapped GABA prevented penicillin- or isoniazidinduced epileptogenesis in rats, whereas free GABA had little effect (Loeb et al., 1982, 1986). We sought to determine whether DN-1417, entrapped in liposomes (DN-L) and administered i.p. exerts a more marked anticonvulsant effect than free DN-1417 in AM-kindled rats. We also studied
Received August 1989; revision accepted January 1992. Address correspondence and reprint requests to Dr. N. Mori at Department of Neuropsychiatry , Fukushima Medical College, 1 Hikarigaoka, Fukushima-shi 960-12, Japan.
994
LIPOSOMES A N D TRH DERIVATIVE IN KINDLED RATS the effect of chronic i.p. administration of liposomes on brain neurons. MATERIALS AND METHODS Preparation of AM kindling Male Wistar rats weighing 200-250 g, under pentobarbital anesthesia (50 mg/kg, i.p.), had a bipolar electrode (stainless-steel wire 0.2 mm in diameter with tip separation of 0.5 mm) implanted into the left AM. Extradural screw electrodes were placed over the bilateral sensorimotor cortexes. One week postoperatively, the afterdischarge (AD) threshold was determined in the left AM and daily AM stimulation was delivered bipolarly in a 1-s train with a constant 60-Hz square-wave current at the AD threshold (ADT). The developing seizure was classified into five stages (Racine, 1972). All animals were kindled at the left AM until a stable stage 5 seizure was evoked for 5 successive days. Subsequently, the stimulus intensity was gradually reduced and the last intensity to induce a stage 5 seizure was designated the generalized seizure triggering threshold (GST) Preparation of test compounds DN-1417 was dissolved in deionized water at 1 or 2 mg/ml. To prepare liposomes, 100 pmol L-aphosphatidylcholine (Sigma, St. Louis, MO, U.S.A.), 28.6 pmol cholesterol (Wako Pure Chemical, Osaka, Japan), and 14.3 pmol stearylamine (Wako Pure Chemical) were dissolved in chloroform at a molar ratio of 7:2: 1. After the chloroform was removed by nitrogen gas and subsequent rotary evaporation for 2 h, a thin dry phospholipid film was formed. The film was dispersed by mechanical shaking in 10 ml deionized water at -50°C. To prepare DN-L, the film was dissolved in 10 ml DN1417 solution (1 or 2 mglml). Both liposomes and DN-L were prepared on the day of the experiment and used without sonication. Assessment of anticonvulsant effect In a previous study with i.v. administration, 1 or 2 mg/kg DN-1417 had an anticonvulsant effect on kindled AM seizure in cats (Sato et al., 1984). In a preliminary dose-response experiment, we noted that i.p. injection of the same quantities produced little or no effect, but 4 mg/kg was effective. Therefore, we used 4 and 8 mg/kg in this study. The kindled animals were divided into six groups: They received deionized water i.p. (n = 7), liposomes (n = 7), 4 mglkg DN-1417 (n = 8), 8 mg/kg DN-1417 (n = 6), 4 mg/kg DN-L (n = lo), or 8 mg/kg DN-L (n = 5 ) , each in a 4-ml/kg vol. A preliminary experiment showed that i.p. DN-1417 had no effect in 6 2
995
h, but liposomes readily cross the BBB (Kimelberg and Papahadjopoulos, 1971; Jonah et al., 1975; Loeb et al., 1982, 1986), suggesting an earlier onset of central effects of DN-L. Therefore, 2 h after the drug treatments, the left AM was stimulated at the previously established GST. If AD was not evoked at the GST, the stimulus intensity was increased in 40-pA steps until an AD was elicited. Subsequently, the animals were tested every 24 h at the GST or at the increased ADT. Assessment of neurotoxicity of liposomes after chronic treatment We found no published data regarding untoward effects of systemically administered liposomes. To test the neurotoxicity of systemic chronic administration of liposomes, we administered 4 ml/kg of liposomes i.p. to a separate group of animals (n = 5 ) once daily for 2 weeks, after which we examined their brain tissue. Histologic examination On completion of the experiment, all animals were perfused with physiologic saline and 3% formaldehyde under deep pentobarbital anesthesia. Their brains were serially sectioned and stained with cresyl violet. All electrodes were located in the intended structures. RESULTS Neither deionized water nor liposomes had any anticonvulsant effect (Fig. l), but both DN-1417 and DN-L were effective. Although DN-1417 was totally ineffective at 2 h, AD was completely suppressed between 24 and 48 h (Table 1). This suppression occurred in 6 of 8 animals that received 4 mg/kg DN-1417 and in 2 of 6 animals that received 8 mgikg DN-1417 and lasted for 2-6 days. However, when the kindled AM was restimulated 10 min later at 40-400 pA above the GST, AD occurred in all animals (Fig. 2). The AD was associated with the kindled seizure in all animals except one, and that animal had a stage 1 manifestation. The low dose (4 mg/kg) of DN-1417 was more effective than the high dose (8 mg/kg). Unlike DN-1417, DN-L completely suppressed the AD 2 h after treatment (Table 2), but restimulation 10 min later at 40-160 pA above the GST reactivated the kindled seizure (Fig. 3). Furthermore, DN-L increased the GST or ADT for a significantly longer time than did DN-1417 (Table 3). The low dose (4 mg/kg) of DN-L was more effective than the high dose (8 mg/kg). Injection of deionized water, liposomes, DN1417, or DN-L i.p. produced no specific neuronal Epilepsiu, Vol. 33, N o . 6 . 1992
N . MORI AND T . FUKATSU
996
Psec FIG. 1. Effect of deionized water or liposomes (LIPO) on kindled seizure. The animals were tested 2 h after drug treatment and then again every 24 h until five consecutive stage 5 seizures were evoked at the generalized seizure triggering threshold (GST). Deionized water had no effect (left). EEGs were recorded 2 h, 2 days, and 4 days after treatment in an animal (animal 1) given deionized water. Liposomes had no effect (right). EEGs were recorded 2 h, 2 days, and 4 days after treatment in an animal (animal 1) given liposomes. L-AM, left amygdala; L-CX and R-CX, left and right sensorimotor cortex. Numbers (30, 64, 60) in top left corners refer to time (in seconds) after cessation of (amygdaloid) AM stimulation.
damage anywhere in the brain. Neither was there any neuronal damage in a n y brain region after chronic administration of liposomes.
does not damage brain neurons even after 2-week treatment, although liposomes injected directly in mouse brain caused hemorrhagic necrosis at the injection site (Adams et al., 1977). DN-1417 administered i.p. had a delayed but prolonged anticonvulsant effect in AM-kindled rats. This finding is comparable to that obtained in AMkindled cats (Sato et al., 1984). DN-1417 given i.v. had a long-term anticonvulsant effect on kindled AM seizure in cats. The effect of DN-1417 on Lennox-Gastaut syndrome has also been reported to persist for 30-70 days after the last intramuscular injection (Ueda et al., 1983). The extrapituitary ef-
DISCUSSION Our results confirm the antiepileptic efficacy of DN-1417, which has been demonstrated clinically (Inanaga and Inoue, 1981; Ueda et al., 1983; Matsumoto et al., 1987; Inanaga et al., 1989) and experimentally (Sato et al., 1984, 1985; Sakai et al., 1991). They also indicate that entrapment of DN-1417 in liposomes potentiates its effect. In addition, our results show that administration of liposomes i.p.
TABLE 1. Efiect ofDN-1417 on GST or ADT Dose of DN- 1417/ rat no. 4 mg/kg 1 2 3 4 5 6
7 8 8 mg/kg 1 2 3 4 5 6
Time after injection 2h
I
0
+ 120 + 40 0 + 160 0 + 80 0 + 40
0 0 0 0 0 0
0 0 0 0 0 + 80
0 0 0 0 0 0 0
2
4
3
+120 + 40 0 +I60 0 + 80 0 + 40 0
+ 80 0 0 0 + 80
+120 + 40 0 +400” 0
0 0 0 0
-
+I20 + 40 0 0 0 0 0 0
5
7
6
8
9
10
1 1 (day) -
~~
0
+I20 0
0 0
0 0
0
0
0
0
0
0
0
0
0
0
+I20
0 0
0
0
0
+ 80
+ 80
0
0
0
0 0 0 0
0 0 0 0
0
0
0
0
0
GST, generalized seizure triggering threshold; AD, afterdischarge. When AD was not induced at the previously established GST, stimulus intensity was increased in 40-pA steps until an AD was elicited. Values indicated by 0 (GST), +40 (GST plus 40 PA), +SO (GST plus 80 PA) refer to stimulus intensity required to evoke an AD. a AD was associated with stage 5 seizure in all animals except one, in which an AD was associated with stage 1 manifestation. Epilepsia, Vol. 33, No. 6, 1992
LIPOSOMES AND TRH DERIVATIVE IN KINDLED RATS
-
2hr
DN-1417 (4 mglkg, # I )
u --; -
L-AML-cx R-CXZ GST
4
L L
r r r r w -
strg, 5
6 days
-?
. C
---
A X
N
-
7 P
-
7
*
L
p -
stage 5
4 days
-
1-AM L-CX’-: ACXGGST
DN-1417 (8 rnglkg, #2)
2 hr
“rrrenrr*rr*-
997
-.
/--
L.AN, La-
-
_c__
7
stage 5
FIG. 2. Effect of DN-1417 on kindled seizure. EEGs were recorded 2 h , 6 days, and 7 days after treatment in an animal (animal 1) given 4 mg/kg DN-1417 (left).In this animal, DN-1417 had no effect at 2 h but, after discharge (AD) was suppressed from day 1 through day 6 after treatment although restimulation at 120 yA above the generalized seizure triggering threshold (GST) evoked stage 5 seizure. At 7 days, stage 5 seizure was evoked at the GST. EEGs were recorded 2 h, 4 days, and 5 days after treatment in an animal (animal 2) given 8 mg/kg DN-1417 (right). In this animal, there was no effect at 2 h, but AD was suppressed from day 2 through day 4 after treatment although stage 5 seizure was elicited at 80 yA above the GST. This animal responded to GST stimulation 5 days after treatment. L-AM,left amygdala; L-CX and R-CX, left and right sensorimotor cortex.
seizures at 2 h. Furthermore, the effect of DN-L lasted longer than that of free DN-1417. Although clarifying the mechanism of DN-L activity in the CNS was not a goal of this study, two hypotheses can be advanced. First, DN-L may enter the brain more rapidly than DN-1417 because liposomes readily cross the BBB (Kimelberg and Papahadjopoulos, 1971; Jonah et al., 1975; Loeb et al., 1982, 1986). Second, liposomes may delay enzymatic degradation of DN-L, thus prolonging its action (Kimelberg and Papahadjopoulos, 1971; Jonah et al., 1975; Gregoriadis, 1976; Ostro and Cullis, 1989; Rande, 1989). Loeb et al. (1982, 1986) demonstrated that i.p. injection of liposome-entrapped GABA
fects of TRH and DN-1417 have been shown to be correlated with acetylcholine (Malthe-Sorenssen et al., 1978; Narumi et al., 1983), catecholamines (Keller et al., 1974; Bennett et al., 1981), serotonin (Rastogi et al., 1981), GABA (Cott and Engel, 1977), glutamate (Renaud et al., 1979), and opioids (Tache et al., 1977), all of which are known to modulate some aspect of CNS excitability relevant to the problem of epilepsy. Enhancement of catecholaminergic mechanisms appears to be responsible for the anticonvulsant effects of TRH and DN-1417 (Bennett et al., 1981; Miyamoto et al., 1981; Sat0 et al., 1985). Unlike free DN-1417, DN-L suppressed kindled
TABLE 2. Effect of DN-L on GST or ADT Dose of DN-L/ rat no.
4 mdkg
Time after injection
2h
1
1
0
2 3 4
0
5
6
7 8 9 10
8mdk 1 2 3 4 5
0 +I20
2 +40
+120 0 +so +80 +I20 +SO +80 +160 +160 +160 0 +SO +SO +40 0 0 +I20 +160 + I 2 0 +I20 +I20 +I20 0 0 0 0
+40 0 0
+40 +40 +40 0
+sc +so
+40 +40 +40 0 +80
9
3
4
5
6
+40“ +I20
0 +320”
+SO +I60 +SO 0 +120 +I20 0
+80 +120 +120 +80 0 +SO 0 +I20 +120 0
0 0 0 +200 0 0
0 0 0 +120 +I20 +I20 0 0 0 0 0 0 +80 +80 +80 0 0 +I20 +I20 +I20 +120
0 +I20
0 +I20
+40 +40
0 0
+40
+40 0
0 0 0
0 0 0
0 0 0
0 0 0
0
0
0
0
0
0
+so
0 +80
+so
+so
+320 0 +320 0 +I20 +I20
7
8
10
0 0 0 +I20 +I20 +I20
12
11
13
15
14
17 (day)
I6
0
0
0
0
+SO
+80
0
0
0
0
0
+I20
+I20
+I20
+I20
+I20
+I20
+I20
+I20
+I20
+I20
+I20
+I20
+I20
+I20
Abbreviations as in Table 1. * AD was associated with stage 5 seizure in all animals except one, in which an AD was with stage 1 manifestation.
Epilepsia, Vol. 33, N o . 6, 1992
N . MORI AND T . FUKATSU
998
2 hr
DN-L (4mg/kg, #4)
2 hr
12 days
DN-L (8 mg/kg, #2)
3 days
2SeC
FIG. 3. Effect of DN-L on kindled seizure (left). EEGs were recorded 2 h, 12 days, and 13 days after treatment in an animal (animal 4) given 4 mg/kg DN-L. In this animal, after discharge (AD) was suppressed at 2 h and the effect lasted 12 days after treatment. Increasing the stimulus intensity, reactivated stage 5 seizure however. At 13 days, stage 5 seizure was elicited at the generalized seizure triggering threshold (GST) (right). EEGs were recorded 2 h, 3 days, and 4 days after treatment in an animal (animal 2) given 8 mg/kg DN-L. In this animal, AD was suppressed from the second hour through day 3 after treatment, although stage 5 seizure was elicited by increasing stimulus intensity. At 4 days, stage 5 seizure was elicited at the GST. L-AM, left amygdala; L-CX and R-CX, left and right sensorimotor cortex.
prevents penicillin- or isoniazid-induced epileptogenesis in rats, whereas GABA alone has little effect. Entrapment of SOD in liposomes has also been reported to block oxygen-induced brain damage (Yusa et al., 1984; Chan et al., 1987). Without liposomes, this enzyme had no such effect. These findings support the contention that as drug delivery vehicles liposomes can enhance the actions of drugs in the CNS without producing overt brain damage. In our study, the lower dose, when given as i.p. DN-1417 or i.p. DN-L, had a more marked effect than the higher dose. The anticonvulsant effect of i.v. DN-1417 evaluated in AM-kindled cats was not dose-related (Sato et al., 1984), whereas i.c.v. DN1417 had a dose-dependent anticonvulsant effect in AM-kindled rats (Sato et al., 1985). In the Senegalese baboon Papio papio, i.v. DN-1417 had no effect on photosensitivity or cortically kindled seiTABLE 3 . Duration of seizure suppression after DN-1417 or D N - L treatment Treatment
DN-1417 (mg/kg) 4 8 DN-L (mg/kg) 4 8
n
Duration of seizure suppression (days)
8 6
2.1 f 0.8 0.8 5 0.5
10 5
6.8 2.8
f 2
1.9“ 0.7“
Values are mean f SEM. p < 0.05 versus treatment of identical dose of DN-1417 (Student’s t test). a
Epilepsia, Vol. 33, NO. 6 , 1992
zures, whereas i.c.v. injection had a significant effect (Sakai et al., 1991). Thus, the effect of DN-1417 on the CNS appears to depend on its quantity in the brain, but DN-1417 cannot penetrate the BBB sufficiently. Therefore, the result of our study may mean that less DN-1417 or DN-L enters the brain at the higher dose than at the lower dose. Because both DN-1417 and DN-L were given systemically, absorption and/or metabolism in the circulation and in organs other than the brain may affect the amount that can cross the BBB, but this issue remains to be studied. DN-1417 or DN-L blocked AD generation at the GST, but stage 5 or stage 1 seizure was evoked by increasing stimulus intensity, indicating that the “all-or-none” property was lost. Because the interstimulus interval was 10 min, previous stimulations may have diminished the effect of subsequent stimulations, resulting in stage 1 seizure. Acknowledgment: This work was supported by grants from the Ministry of Health and Welfare of Japan. DN1417 was supplied by Takeda Pharmaceutical Company, Ltd.
REFERENCES Adams DH, Joyce G, Richardson VJ, et al. Liposome toxicity in the mouse central nervous system. J Neurof Sci 1977;31: 173-9. Bennett GW, Marsden CA, Metcaff G, et al. Analogues of thyrotropin releasing hormone (TRH) stimulate in vivo release of endogenous dopamine from rat brain regions. Br J Pharmacol 1981;74:227-8.
LIPOSOMES A N D TRH DERIVATIVE IN KINDLED RATS Chan PH, Longar S, Fishman RA. Protective effects of liposome-entrapped superoxide dismutase on posttraumatic edema. A n n Neurol 1987;21:540-7. Cott J, Engel J Jr. Antagonism of analeptic activity of thyrotropin releasing hormone (TRH) by agents which enhance GABA transmission. Psychopharmacology (Berlin) 1977;52: 1145-9. Fukuda N, Nishimura 0, Shikata M, et al. Synthesis and pharmacology of TRH analog to separate central nervous action from endocrine activity. Chem Pharm Bull (Tokyo) 1980;28: 1667-72. Gregoriadis G . The carrier potential of liposomes in biology and medicine. N Engl J Med 1976;295:704-10. Hokfelt J, Fux K, Jonansson 0 , et al. Distribution of thyrotropin-releasing hormone (TRH) in the central nervous system as revealed with immunohistochemistry. Eur J Pharmacol 1975;34:89-96. Inanaga K , Inoue Y. Effect of a thyrotropin releasing hormone analog in patients with myoclonus epilepsy. Kurume Med J 1981;28:201-10. Inanaga K, Kumashiro H, Fukuyama Y, et al. Clinical study of oral administration of DN-1417, a TRH analog, in patients with intractable epilepsy. Epilepsia 1989;30:438-45. Jonah MM, Cerny EA, Rahman YE. Tissue distribution of EDTA encapsulated within liposomes of varying surface property. Biochim Biophys Actu 1975;401:336-48. Keller HH, Bartholine G , Pletscher A. Enhancement of central noradrenaline turnover by thyrotropin releasing hormone. Nature 1974;48:528-9. Kimelberg HK, Papahadjopoulos D. Phospholipid-protein interactions: membrane permeability correlated with monolayer “penetration.” Biochim Biophys Actu 1971;133:805-9. Loeb C, Benassi E, Besio G , et al. Liposome-entrapped GABA modifies behavioral and electrographic changes of penicillininduced epileptic activity. Neurology 1982;32:1234-8. Loeb C, Besio G, Mainardi P, et al. Liposome-entrapped yaminobutyric acid inhibits isoniazid-induced epileptogenic activity in rats. Epilepsia 1986;27:98-102. Malthe-Sorenssen D, Cheney DL, Costa E, et al. Modulation of the turnover rate of acetylcholine in rat brain by intraventricular injection of thyrotropin releasing hormone, somatostatin, neurotensin and angiotensin 11. J Neurochem 1978;31: 381-91. Matsurnoto A, Kumagai T, Takeuchi T, et al. Clinical effects of thyrotropin-releasing hormone in childhood: a comparative study with ACTH therapy. Epilepsia 1987;28:49-50. Miyamoto M, Fukuda N, Narumi S, et al. y-Butyrolactone-ycarbonyl-L-histidyl-L-propinamide citrate (DN-1417): a novel analog with potent effects on the central nervous system. Life Sci I981 ;28:861-9. Narumi S, Nagai M, Miyamoto M, et al. Thyrotropin releasing hormone (TRH) and its analog (DN-1417): interaction with phenobarbital in choline uptake and acetylcholine synthesis of the rat brain slice. Life Sci 1983;32:1637-45. Nemeroff CB, Prange AJ, Bessete JrG, et al. Thyrotropin releasing hormone (TRH) and its p-alanin analogue potentiation of the anticonvulsant potency of phenobarbital in mice. Psychopharmucol Commun 1975;1:305-7. Ostro MJ, Cullis PR. Use of liposomes as injectable delivery systems. A m J Hosp Pharm 1989;46:157&87. Racine RJ. Modification of seizure activity by electrical stimulation. 11. Motor seizure. Electroencephalogr Clin Neurophysiol 1972;32:281-94. Rande VV. Drug delivery systems. 1. Site-specific drug delivery using liposomes as carriers. J Clin Pharmacol 1989;29:68594. Rastogi RB, Singhal RL, Lapierre YD. Effects of MK-771, a novel TRH analogue, on brain dopaminergic and serotonergic systems. Eur J Pharmacol 1981;73:307-12. Renaud LP, Blume HW, Pittman QJ, et al. Thyrotropin releasing hormone (TRH) selectively depresses glutamate excitation of cerebral cortical neurons. Science 1979;205: 1275-7. Sakai S, Baba H, Sato M, et al. Effect of DN-1417 on photo-
999
sensitivity and cortically kindled seizure in Senegalese baboons, Papio papio. Epilepsia 1991;32:16-21. Sato M, Kajita S , Okamoto M, et al. Anticonvulsant effects of thyrotropin-releasing hormone analog (DN-1417) and changes in brain rnonoamines. Psychiar Neurol Jpn 1985;87: 176-85. Sato M, Morimoto K, Wada JA. Antiepileptic effects of thyrotropin-releasing hormone and its new derivative. Epilepsia 1984;25:537-44. Tache Y, Lis M, Collu R. Effects of thyrotropin releasing hormone on behavior and hormonal changes induced by betaendorphin. Life Sci 1977;21:84 1-6. Takada G , Onodera H , Tada K. Delivery of fungal p-galactosidase to rat brain by means of liposomes. Tohoku J Exp Med I982 ;136:219-29. Ueda S , Nakamura J, Inanaga K. Clinical effects of TRH analog (DN-1417) on the Lennox syndrome. J Jpn Epilepsy So; 1983;1:3 1-9. Yusa T, Crapo JD, Freeman BA. Liposome-mediated augmentation of brain SOD and catalase inhibits CNS 0, toxicity. J Appl Physiol 1984;57:674-81.
RkSUME Les auteurs prtsentent les resultats d’une etude a long terme de 50 patients qui ont benCfici6 d’une rbsection du nCocortex temporal avec preservation des structures lirnbiques profondes, comme traitement chirurgical d’une Cpilepsie temporale rebelle. La pkriode de suivi allait de 3 ri 15 ans. Les investigations EEG pre-operatoires etaient fondtes sur les decharges critiques seulement. Trois facteurs emergent comme predictifs d’une Cvolution post-operatoire favorable: ( I ) un foyer unilatCral net, temporal anterieur ou moyen (p < 0.01); (2) un debut sttrLotypC des crises temporales (p < 0.005) et (3) un plus grand volume de tissu reseque lors de la chirurgie (p < 0.05). Les rCsultats globaux montrent que 62% des patients ont bkntficie d’un resultat considere comme “guerison” ou “presque gukrison”, selon les critkres modifies de Crandall (categories Crandall I et 11). Ceux qui ont bCneficie d’une reduction significative des crises mais qui continuent a presenter une epilepsie rebelle (Crandall 111) n’ont pas CtC considCres comme bCnCficiant d’un bon resultat. L’Cvolution globale se compare de faGon favorable avec celle relevCe dam d’autres centres utilisant des approches chirurgicales diffkrentes, et indique que la nkocortectomie est une procedure acceptable chez des patients hautement stlectionnes, mCme lorsque des moyens limites sont disponibles. P. Genton, Marseille)
RESUMEN Se presentan 10s resultados de un estudio de seguimiento a largo plazo de 50 pacientes que fueron sometidos a una resecci6n del neocortex temporal con preservaci6n de las estructuras limbicas profundas como tratamiento quirdrgico de epilepsia del 16bulo temporal resistente. El periodo de seguimiento oscil6 entre 3 y 15 anos. Las investigaciones electroencefalogrdficas preoperatorias se basaron solamente en las descargas interictales. Tres factores han sido interpretados como elementos de predicci6n de un buen resultado: (1) un claro foco anterior y temporal medio unilateral (p < 0.01), (2) comienzo estereotipado del ataque del 16bulo temporal (p < 0.005) y (3) mayor cantidad de volumen de tejido estirpado durante la cirugia (p < 0.05). Los resultados generales mostraron que un 62% se seguieron de “curaci6n” o “casi curaci6n” segdn una versi6n modificada de 10s criterios de Crandall (Crandall I y 11). Aquellos casos que experimentaron una reducci6n significativa del ndmero de ataques pero que continuaron sufriendo epilepsia resistente al tratamiento (grupo I11 de Crandall) se clasificaron como resultados pobres. En general este resultado, m8s favorable que 10s obtenidos en otros centros utilizando diferentes tkcnicas quirdrgicas, indica que la neurocorticectomfa es un procedimiento apropiado Epilepsia, Vol. 33, No. 6 , 1992
1000
N . MORI AND T. FUKATSU
en una poblacion bien seleccionada incluso cuando solo se dispone de una tecnologia limitada. (A. Portera-Sanchez, Madrid)
ZUSAMMENFASSUNG Es werden die Langzeitergebnisse von 50 Patienten dargestellt, die wegen unkontrollierbarer Temporallappenepilepsie einer tempoarlen Neokortektomie unter Erhaltung der tiefen limbischen Strukturen unterzogen wurden. Die Nachbeobachtung liegt zwischen 3 und 15 Jahren. Praoperative EEG-Untersuchungen umfaRten nur die interiktalen Entladungen. Fur eine giinstige Prognose sprechen drei Faktoren: 1. eindeutiger, einseitiger
Epilepsia, Vol. 33, N o . 6 , 1992
Fokus temporal vorne bis mitte (p 0.01); 2. stereotyper Beginn des Temporallappenanfalls (p < 0.01); 3. Entfernung eines groRen Gewebevolumens bei der Operation ( p < 0.05). Das Gesarntergebnis lag bei 62% geheilter oder fast geheilter Patienten gemaR den rnotifizierten Crandall-Kriterien (I 11). Patienten mit Anfallsreduktion aber unbeeinflufibarer Epilepsie wurden nicht als gebessertbetrachtet. Das Gesamtergebnis ist im Vergleich rnit anderen Zentren rnit unterschiedlichem chirurgischen Vorgehen giinstig und zeigt, daR die Neokortektomie ein angemessenes Vorgehen ist bei einer hochselektionierten Patientengruppe, selbst wenn die zur Vergiigung stehenden diagnostischen Moglichkeiten begrenzt sind.
+
(C. G. Lipinski, HeidelberglNeckarRemiind)
Anticonvulsant Effect of DN-1417, a Derivative of Thyrotropin-Releasing Hormone, and Liposome-Entrapped DN- 1417, on Amygdaloid-Kindled Rats Norio Mori and Toshihiko Fukatsu Department of Neuropsychiatry, Fukushima Medical College, Fukushima, Japan
Summary: The effects of y-butyrolactone-y-carbony1-Lhistidyl-L-propinamide citrate (DN-1417), a derivative of thyrotropin-releasing hormone, and liposome-entrapped DN-1417 (DN-L) were examined in amygdaloid-kindled rats. The animals were tested 2 h after intraperitoneal (i.p.) drug administration and then again every 24 h without further drug treatment. DN-1417 did not suppress the kindled seizure at 2 h but did beginning 1-6 days after injection. DN-L suppressed the kindled seizure at 2 h and had a more prolonged anticonvulsant effect than DN-
1417. After liposomes were given i.p. once daily for 2 weeks, there was no morphologic evidence that liposomes damaged brain neurons. These results, together with previously published data, suggest that as drug delivery vehicles, liposomes can enhance the effectiveness of drugs in the CNS without producing overt brain damage. Key Words: Liposomes-Thyrotropin-releasing hormone-y-Butyrolactone-y-carbonyl-L-histidyl-L-propinamide citrate (DN- 1417)--Amygdala-Kindling.
Thyrotropin-releasing hormone (TRH) is widely distributed in the brain (Hokfelt et al., 1975) and has been reported to have various central effects unrelated to endocrine action through the pituitarythyroid axis (Nemeroff et al., 1975). y-Butyrolactone-y-carbonyl-L-histidyl-L-propinamide citrate (DN-1417), an analogue of TRH, has more profound central activity than TRH, with only - v 3 0 of the thyroid-stimulating hormone-releasing activity (Fukuda et al., 1980). Clinical studies have shown that DN-1417 has beneficial effects on various aspects of degenerative myoclonus epilepsy, LennoxGastaut syndrome, and infantile spasms (Inanaga and Inoue, 1981; Ueda et al., 1983; Matsumoto et al., 1987; Inanaga et al., 1989). The anticonvulsant effect of intravenous (i.v.) DN-1417 in amygdaloid (AM)-kindled cats was not dose-related, however, and there was considerable interindividual variation (Sato et al., 1984), whereas intracerebroventricular (i.c.v.) DN-1417 had a dose-dependent anticonvulsant effect in AM-kindled rats (Sato et al., 1985). In the Senegalese baboon, Pupio pupio, i.v.
DN-1417 had no effect on photosensitivity or cortically kindled seizures, whereas i.c.v. injection had a significant effect (Sakai et al., 1991). Therefore, the blood-brain barrier (BBB) appears to pose a serious obstacle to DN-1417. Liposomes, microscopic vesicles composed of one or several lipid membranes surrounding discrete aqueous compartments, can encapsulate water-soluble chemicals in their aqueous spaces. Because liposomes can enhance the effectiveness of biologically active materials or reduce toxicity or both, they may be useful as drug delivery vehicles (Gregoriadis et al., 1976; Ostro and Cullis, 1989; Rande, 1989). Recent studies have demonstrated liposome-mediated brain uptake of BBB-impermeable substances, including P-galactosidase (Takada et al., 1982), superoxide dismutase (SOD) (Yusa et al., 1984; Chan et al., 1987), and y-aminobutyric acid (GABA) (Loeb et al., 1982, 1986). Indeed, intraperitoneal (i.p.) injection of liposomeentrapped GABA prevented penicillin- or isoniazidinduced epileptogenesis in rats, whereas free GABA had little effect (Loeb et al., 1982, 1986). We sought to determine whether DN-1417, entrapped in liposomes (DN-L) and administered i.p. exerts a more marked anticonvulsant effect than free DN-1417 in AM-kindled rats. We also studied
Received August 1989; revision accepted January 1992. Address correspondence and reprint requests to Dr. N. Mori at Department of Neuropsychiatry , Fukushima Medical College, 1 Hikarigaoka, Fukushima-shi 960-12, Japan.
994
LIPOSOMES A N D TRH DERIVATIVE IN KINDLED RATS the effect of chronic i.p. administration of liposomes on brain neurons. MATERIALS AND METHODS Preparation of AM kindling Male Wistar rats weighing 200-250 g, under pentobarbital anesthesia (50 mg/kg, i.p.), had a bipolar electrode (stainless-steel wire 0.2 mm in diameter with tip separation of 0.5 mm) implanted into the left AM. Extradural screw electrodes were placed over the bilateral sensorimotor cortexes. One week postoperatively, the afterdischarge (AD) threshold was determined in the left AM and daily AM stimulation was delivered bipolarly in a 1-s train with a constant 60-Hz square-wave current at the AD threshold (ADT). The developing seizure was classified into five stages (Racine, 1972). All animals were kindled at the left AM until a stable stage 5 seizure was evoked for 5 successive days. Subsequently, the stimulus intensity was gradually reduced and the last intensity to induce a stage 5 seizure was designated the generalized seizure triggering threshold (GST) Preparation of test compounds DN-1417 was dissolved in deionized water at 1 or 2 mg/ml. To prepare liposomes, 100 pmol L-aphosphatidylcholine (Sigma, St. Louis, MO, U.S.A.), 28.6 pmol cholesterol (Wako Pure Chemical, Osaka, Japan), and 14.3 pmol stearylamine (Wako Pure Chemical) were dissolved in chloroform at a molar ratio of 7:2: 1. After the chloroform was removed by nitrogen gas and subsequent rotary evaporation for 2 h, a thin dry phospholipid film was formed. The film was dispersed by mechanical shaking in 10 ml deionized water at -50°C. To prepare DN-L, the film was dissolved in 10 ml DN1417 solution (1 or 2 mglml). Both liposomes and DN-L were prepared on the day of the experiment and used without sonication. Assessment of anticonvulsant effect In a previous study with i.v. administration, 1 or 2 mg/kg DN-1417 had an anticonvulsant effect on kindled AM seizure in cats (Sato et al., 1984). In a preliminary dose-response experiment, we noted that i.p. injection of the same quantities produced little or no effect, but 4 mg/kg was effective. Therefore, we used 4 and 8 mg/kg in this study. The kindled animals were divided into six groups: They received deionized water i.p. (n = 7), liposomes (n = 7), 4 mglkg DN-1417 (n = 8), 8 mg/kg DN-1417 (n = 6), 4 mg/kg DN-L (n = lo), or 8 mg/kg DN-L (n = 5 ) , each in a 4-ml/kg vol. A preliminary experiment showed that i.p. DN-1417 had no effect in 6 2
995
h, but liposomes readily cross the BBB (Kimelberg and Papahadjopoulos, 1971; Jonah et al., 1975; Loeb et al., 1982, 1986), suggesting an earlier onset of central effects of DN-L. Therefore, 2 h after the drug treatments, the left AM was stimulated at the previously established GST. If AD was not evoked at the GST, the stimulus intensity was increased in 40-pA steps until an AD was elicited. Subsequently, the animals were tested every 24 h at the GST or at the increased ADT. Assessment of neurotoxicity of liposomes after chronic treatment We found no published data regarding untoward effects of systemically administered liposomes. To test the neurotoxicity of systemic chronic administration of liposomes, we administered 4 ml/kg of liposomes i.p. to a separate group of animals (n = 5 ) once daily for 2 weeks, after which we examined their brain tissue. Histologic examination On completion of the experiment, all animals were perfused with physiologic saline and 3% formaldehyde under deep pentobarbital anesthesia. Their brains were serially sectioned and stained with cresyl violet. All electrodes were located in the intended structures. RESULTS Neither deionized water nor liposomes had any anticonvulsant effect (Fig. l), but both DN-1417 and DN-L were effective. Although DN-1417 was totally ineffective at 2 h, AD was completely suppressed between 24 and 48 h (Table 1). This suppression occurred in 6 of 8 animals that received 4 mg/kg DN-1417 and in 2 of 6 animals that received 8 mgikg DN-1417 and lasted for 2-6 days. However, when the kindled AM was restimulated 10 min later at 40-400 pA above the GST, AD occurred in all animals (Fig. 2). The AD was associated with the kindled seizure in all animals except one, and that animal had a stage 1 manifestation. The low dose (4 mg/kg) of DN-1417 was more effective than the high dose (8 mg/kg). Unlike DN-1417, DN-L completely suppressed the AD 2 h after treatment (Table 2), but restimulation 10 min later at 40-160 pA above the GST reactivated the kindled seizure (Fig. 3). Furthermore, DN-L increased the GST or ADT for a significantly longer time than did DN-1417 (Table 3). The low dose (4 mg/kg) of DN-L was more effective than the high dose (8 mg/kg). Injection of deionized water, liposomes, DN1417, or DN-L i.p. produced no specific neuronal Epilepsiu, Vol. 33, N o . 6 . 1992
N . MORI AND T . FUKATSU
996
Psec FIG. 1. Effect of deionized water or liposomes (LIPO) on kindled seizure. The animals were tested 2 h after drug treatment and then again every 24 h until five consecutive stage 5 seizures were evoked at the generalized seizure triggering threshold (GST). Deionized water had no effect (left). EEGs were recorded 2 h, 2 days, and 4 days after treatment in an animal (animal 1) given deionized water. Liposomes had no effect (right). EEGs were recorded 2 h, 2 days, and 4 days after treatment in an animal (animal 1) given liposomes. L-AM, left amygdala; L-CX and R-CX, left and right sensorimotor cortex. Numbers (30, 64, 60) in top left corners refer to time (in seconds) after cessation of (amygdaloid) AM stimulation.
damage anywhere in the brain. Neither was there any neuronal damage in a n y brain region after chronic administration of liposomes.
does not damage brain neurons even after 2-week treatment, although liposomes injected directly in mouse brain caused hemorrhagic necrosis at the injection site (Adams et al., 1977). DN-1417 administered i.p. had a delayed but prolonged anticonvulsant effect in AM-kindled rats. This finding is comparable to that obtained in AMkindled cats (Sato et al., 1984). DN-1417 given i.v. had a long-term anticonvulsant effect on kindled AM seizure in cats. The effect of DN-1417 on Lennox-Gastaut syndrome has also been reported to persist for 30-70 days after the last intramuscular injection (Ueda et al., 1983). The extrapituitary ef-
DISCUSSION Our results confirm the antiepileptic efficacy of DN-1417, which has been demonstrated clinically (Inanaga and Inoue, 1981; Ueda et al., 1983; Matsumoto et al., 1987; Inanaga et al., 1989) and experimentally (Sato et al., 1984, 1985; Sakai et al., 1991). They also indicate that entrapment of DN-1417 in liposomes potentiates its effect. In addition, our results show that administration of liposomes i.p.
TABLE 1. Efiect ofDN-1417 on GST or ADT Dose of DN- 1417/ rat no. 4 mg/kg 1 2 3 4 5 6
7 8 8 mg/kg 1 2 3 4 5 6
Time after injection 2h
I
0
+ 120 + 40 0 + 160 0 + 80 0 + 40
0 0 0 0 0 0
0 0 0 0 0 + 80
0 0 0 0 0 0 0
2
4
3
+120 + 40 0 +I60 0 + 80 0 + 40 0
+ 80 0 0 0 + 80
+120 + 40 0 +400” 0
0 0 0 0
-
+I20 + 40 0 0 0 0 0 0
5
7
6
8
9
10
1 1 (day) -
~~
0
+I20 0
0 0
0 0
0
0
0
0
0
0
0
0
0
0
+I20
0 0
0
0
0
+ 80
+ 80
0
0
0
0 0 0 0
0 0 0 0
0
0
0
0
0
GST, generalized seizure triggering threshold; AD, afterdischarge. When AD was not induced at the previously established GST, stimulus intensity was increased in 40-pA steps until an AD was elicited. Values indicated by 0 (GST), +40 (GST plus 40 PA), +SO (GST plus 80 PA) refer to stimulus intensity required to evoke an AD. a AD was associated with stage 5 seizure in all animals except one, in which an AD was associated with stage 1 manifestation. Epilepsia, Vol. 33, No. 6, 1992
LIPOSOMES AND TRH DERIVATIVE IN KINDLED RATS
-
2hr
DN-1417 (4 mglkg, # I )
u --; -
L-AML-cx R-CXZ GST
4
L L
r r r r w -
strg, 5
6 days
-?
. C
---
A X
N
-
7 P
-
7
*
L
p -
stage 5
4 days
-
1-AM L-CX’-: ACXGGST
DN-1417 (8 rnglkg, #2)
2 hr
“rrrenrr*rr*-
997
-.
/--
L.AN, La-
-
_c__
7
stage 5
FIG. 2. Effect of DN-1417 on kindled seizure. EEGs were recorded 2 h , 6 days, and 7 days after treatment in an animal (animal 1) given 4 mg/kg DN-1417 (left).In this animal, DN-1417 had no effect at 2 h but, after discharge (AD) was suppressed from day 1 through day 6 after treatment although restimulation at 120 yA above the generalized seizure triggering threshold (GST) evoked stage 5 seizure. At 7 days, stage 5 seizure was evoked at the GST. EEGs were recorded 2 h, 4 days, and 5 days after treatment in an animal (animal 2) given 8 mg/kg DN-1417 (right). In this animal, there was no effect at 2 h, but AD was suppressed from day 2 through day 4 after treatment although stage 5 seizure was elicited at 80 yA above the GST. This animal responded to GST stimulation 5 days after treatment. L-AM,left amygdala; L-CX and R-CX, left and right sensorimotor cortex.
seizures at 2 h. Furthermore, the effect of DN-L lasted longer than that of free DN-1417. Although clarifying the mechanism of DN-L activity in the CNS was not a goal of this study, two hypotheses can be advanced. First, DN-L may enter the brain more rapidly than DN-1417 because liposomes readily cross the BBB (Kimelberg and Papahadjopoulos, 1971; Jonah et al., 1975; Loeb et al., 1982, 1986). Second, liposomes may delay enzymatic degradation of DN-L, thus prolonging its action (Kimelberg and Papahadjopoulos, 1971; Jonah et al., 1975; Gregoriadis, 1976; Ostro and Cullis, 1989; Rande, 1989). Loeb et al. (1982, 1986) demonstrated that i.p. injection of liposome-entrapped GABA
fects of TRH and DN-1417 have been shown to be correlated with acetylcholine (Malthe-Sorenssen et al., 1978; Narumi et al., 1983), catecholamines (Keller et al., 1974; Bennett et al., 1981), serotonin (Rastogi et al., 1981), GABA (Cott and Engel, 1977), glutamate (Renaud et al., 1979), and opioids (Tache et al., 1977), all of which are known to modulate some aspect of CNS excitability relevant to the problem of epilepsy. Enhancement of catecholaminergic mechanisms appears to be responsible for the anticonvulsant effects of TRH and DN-1417 (Bennett et al., 1981; Miyamoto et al., 1981; Sat0 et al., 1985). Unlike free DN-1417, DN-L suppressed kindled
TABLE 2. Effect of DN-L on GST or ADT Dose of DN-L/ rat no.
4 mdkg
Time after injection
2h
1
1
0
2 3 4
0
5
6
7 8 9 10
8mdk 1 2 3 4 5
0 +I20
2 +40
+120 0 +so +80 +I20 +SO +80 +160 +160 +160 0 +SO +SO +40 0 0 +I20 +160 + I 2 0 +I20 +I20 +I20 0 0 0 0
+40 0 0
+40 +40 +40 0
+sc +so
+40 +40 +40 0 +80
9
3
4
5
6
+40“ +I20
0 +320”
+SO +I60 +SO 0 +120 +I20 0
+80 +120 +120 +80 0 +SO 0 +I20 +120 0
0 0 0 +200 0 0
0 0 0 +120 +I20 +I20 0 0 0 0 0 0 +80 +80 +80 0 0 +I20 +I20 +I20 +120
0 +I20
0 +I20
+40 +40
0 0
+40
+40 0
0 0 0
0 0 0
0 0 0
0 0 0
0
0
0
0
0
0
+so
0 +80
+so
+so
+320 0 +320 0 +I20 +I20
7
8
10
0 0 0 +I20 +I20 +I20
12
11
13
15
14
17 (day)
I6
0
0
0
0
+SO
+80
0
0
0
0
0
+I20
+I20
+I20
+I20
+I20
+I20
+I20
+I20
+I20
+I20
+I20
+I20
+I20
+I20
Abbreviations as in Table 1. * AD was associated with stage 5 seizure in all animals except one, in which an AD was with stage 1 manifestation.
Epilepsia, Vol. 33, N o . 6, 1992
N . MORI AND T . FUKATSU
998
2 hr
DN-L (4mg/kg, #4)
2 hr
12 days
DN-L (8 mg/kg, #2)
3 days
2SeC
FIG. 3. Effect of DN-L on kindled seizure (left). EEGs were recorded 2 h, 12 days, and 13 days after treatment in an animal (animal 4) given 4 mg/kg DN-L. In this animal, after discharge (AD) was suppressed at 2 h and the effect lasted 12 days after treatment. Increasing the stimulus intensity, reactivated stage 5 seizure however. At 13 days, stage 5 seizure was elicited at the generalized seizure triggering threshold (GST) (right). EEGs were recorded 2 h, 3 days, and 4 days after treatment in an animal (animal 2) given 8 mg/kg DN-L. In this animal, AD was suppressed from the second hour through day 3 after treatment, although stage 5 seizure was elicited by increasing stimulus intensity. At 4 days, stage 5 seizure was elicited at the GST. L-AM, left amygdala; L-CX and R-CX, left and right sensorimotor cortex.
prevents penicillin- or isoniazid-induced epileptogenesis in rats, whereas GABA alone has little effect. Entrapment of SOD in liposomes has also been reported to block oxygen-induced brain damage (Yusa et al., 1984; Chan et al., 1987). Without liposomes, this enzyme had no such effect. These findings support the contention that as drug delivery vehicles liposomes can enhance the actions of drugs in the CNS without producing overt brain damage. In our study, the lower dose, when given as i.p. DN-1417 or i.p. DN-L, had a more marked effect than the higher dose. The anticonvulsant effect of i.v. DN-1417 evaluated in AM-kindled cats was not dose-related (Sato et al., 1984), whereas i.c.v. DN1417 had a dose-dependent anticonvulsant effect in AM-kindled rats (Sato et al., 1985). In the Senegalese baboon Papio papio, i.v. DN-1417 had no effect on photosensitivity or cortically kindled seiTABLE 3 . Duration of seizure suppression after DN-1417 or D N - L treatment Treatment
DN-1417 (mg/kg) 4 8 DN-L (mg/kg) 4 8
n
Duration of seizure suppression (days)
8 6
2.1 f 0.8 0.8 5 0.5
10 5
6.8 2.8
f 2
1.9“ 0.7“
Values are mean f SEM. p < 0.05 versus treatment of identical dose of DN-1417 (Student’s t test). a
Epilepsia, Vol. 33, NO. 6 , 1992
zures, whereas i.c.v. injection had a significant effect (Sakai et al., 1991). Thus, the effect of DN-1417 on the CNS appears to depend on its quantity in the brain, but DN-1417 cannot penetrate the BBB sufficiently. Therefore, the result of our study may mean that less DN-1417 or DN-L enters the brain at the higher dose than at the lower dose. Because both DN-1417 and DN-L were given systemically, absorption and/or metabolism in the circulation and in organs other than the brain may affect the amount that can cross the BBB, but this issue remains to be studied. DN-1417 or DN-L blocked AD generation at the GST, but stage 5 or stage 1 seizure was evoked by increasing stimulus intensity, indicating that the “all-or-none” property was lost. Because the interstimulus interval was 10 min, previous stimulations may have diminished the effect of subsequent stimulations, resulting in stage 1 seizure. Acknowledgment: This work was supported by grants from the Ministry of Health and Welfare of Japan. DN1417 was supplied by Takeda Pharmaceutical Company, Ltd.
REFERENCES Adams DH, Joyce G, Richardson VJ, et al. Liposome toxicity in the mouse central nervous system. J Neurof Sci 1977;31: 173-9. Bennett GW, Marsden CA, Metcaff G, et al. Analogues of thyrotropin releasing hormone (TRH) stimulate in vivo release of endogenous dopamine from rat brain regions. Br J Pharmacol 1981;74:227-8.
LIPOSOMES A N D TRH DERIVATIVE IN KINDLED RATS Chan PH, Longar S, Fishman RA. Protective effects of liposome-entrapped superoxide dismutase on posttraumatic edema. A n n Neurol 1987;21:540-7. Cott J, Engel J Jr. Antagonism of analeptic activity of thyrotropin releasing hormone (TRH) by agents which enhance GABA transmission. Psychopharmacology (Berlin) 1977;52: 1145-9. Fukuda N, Nishimura 0, Shikata M, et al. Synthesis and pharmacology of TRH analog to separate central nervous action from endocrine activity. Chem Pharm Bull (Tokyo) 1980;28: 1667-72. Gregoriadis G . The carrier potential of liposomes in biology and medicine. N Engl J Med 1976;295:704-10. Hokfelt J, Fux K, Jonansson 0 , et al. Distribution of thyrotropin-releasing hormone (TRH) in the central nervous system as revealed with immunohistochemistry. Eur J Pharmacol 1975;34:89-96. Inanaga K , Inoue Y. Effect of a thyrotropin releasing hormone analog in patients with myoclonus epilepsy. Kurume Med J 1981;28:201-10. Inanaga K, Kumashiro H, Fukuyama Y, et al. Clinical study of oral administration of DN-1417, a TRH analog, in patients with intractable epilepsy. Epilepsia 1989;30:438-45. Jonah MM, Cerny EA, Rahman YE. Tissue distribution of EDTA encapsulated within liposomes of varying surface property. Biochim Biophys Actu 1975;401:336-48. Keller HH, Bartholine G , Pletscher A. Enhancement of central noradrenaline turnover by thyrotropin releasing hormone. Nature 1974;48:528-9. Kimelberg HK, Papahadjopoulos D. Phospholipid-protein interactions: membrane permeability correlated with monolayer “penetration.” Biochim Biophys Actu 1971;133:805-9. Loeb C, Benassi E, Besio G , et al. Liposome-entrapped GABA modifies behavioral and electrographic changes of penicillininduced epileptic activity. Neurology 1982;32:1234-8. Loeb C, Besio G, Mainardi P, et al. Liposome-entrapped yaminobutyric acid inhibits isoniazid-induced epileptogenic activity in rats. Epilepsia 1986;27:98-102. Malthe-Sorenssen D, Cheney DL, Costa E, et al. Modulation of the turnover rate of acetylcholine in rat brain by intraventricular injection of thyrotropin releasing hormone, somatostatin, neurotensin and angiotensin 11. J Neurochem 1978;31: 381-91. Matsurnoto A, Kumagai T, Takeuchi T, et al. Clinical effects of thyrotropin-releasing hormone in childhood: a comparative study with ACTH therapy. Epilepsia 1987;28:49-50. Miyamoto M, Fukuda N, Narumi S, et al. y-Butyrolactone-ycarbonyl-L-histidyl-L-propinamide citrate (DN-1417): a novel analog with potent effects on the central nervous system. Life Sci I981 ;28:861-9. Narumi S, Nagai M, Miyamoto M, et al. Thyrotropin releasing hormone (TRH) and its analog (DN-1417): interaction with phenobarbital in choline uptake and acetylcholine synthesis of the rat brain slice. Life Sci 1983;32:1637-45. Nemeroff CB, Prange AJ, Bessete JrG, et al. Thyrotropin releasing hormone (TRH) and its p-alanin analogue potentiation of the anticonvulsant potency of phenobarbital in mice. Psychopharmucol Commun 1975;1:305-7. Ostro MJ, Cullis PR. Use of liposomes as injectable delivery systems. A m J Hosp Pharm 1989;46:157&87. Racine RJ. Modification of seizure activity by electrical stimulation. 11. Motor seizure. Electroencephalogr Clin Neurophysiol 1972;32:281-94. Rande VV. Drug delivery systems. 1. Site-specific drug delivery using liposomes as carriers. J Clin Pharmacol 1989;29:68594. Rastogi RB, Singhal RL, Lapierre YD. Effects of MK-771, a novel TRH analogue, on brain dopaminergic and serotonergic systems. Eur J Pharmacol 1981;73:307-12. Renaud LP, Blume HW, Pittman QJ, et al. Thyrotropin releasing hormone (TRH) selectively depresses glutamate excitation of cerebral cortical neurons. Science 1979;205: 1275-7. Sakai S, Baba H, Sato M, et al. Effect of DN-1417 on photo-
999
sensitivity and cortically kindled seizure in Senegalese baboons, Papio papio. Epilepsia 1991;32:16-21. Sato M, Kajita S , Okamoto M, et al. Anticonvulsant effects of thyrotropin-releasing hormone analog (DN-1417) and changes in brain rnonoamines. Psychiar Neurol Jpn 1985;87: 176-85. Sato M, Morimoto K, Wada JA. Antiepileptic effects of thyrotropin-releasing hormone and its new derivative. Epilepsia 1984;25:537-44. Tache Y, Lis M, Collu R. Effects of thyrotropin releasing hormone on behavior and hormonal changes induced by betaendorphin. Life Sci 1977;21:84 1-6. Takada G , Onodera H , Tada K. Delivery of fungal p-galactosidase to rat brain by means of liposomes. Tohoku J Exp Med I982 ;136:219-29. Ueda S , Nakamura J, Inanaga K. Clinical effects of TRH analog (DN-1417) on the Lennox syndrome. J Jpn Epilepsy So; 1983;1:3 1-9. Yusa T, Crapo JD, Freeman BA. Liposome-mediated augmentation of brain SOD and catalase inhibits CNS 0, toxicity. J Appl Physiol 1984;57:674-81.
RkSUME Les auteurs prtsentent les resultats d’une etude a long terme de 50 patients qui ont benCfici6 d’une rbsection du nCocortex temporal avec preservation des structures lirnbiques profondes, comme traitement chirurgical d’une Cpilepsie temporale rebelle. La pkriode de suivi allait de 3 ri 15 ans. Les investigations EEG pre-operatoires etaient fondtes sur les decharges critiques seulement. Trois facteurs emergent comme predictifs d’une Cvolution post-operatoire favorable: ( I ) un foyer unilatCral net, temporal anterieur ou moyen (p < 0.01); (2) un debut sttrLotypC des crises temporales (p < 0.005) et (3) un plus grand volume de tissu reseque lors de la chirurgie (p < 0.05). Les rCsultats globaux montrent que 62% des patients ont bkntficie d’un resultat considere comme “guerison” ou “presque gukrison”, selon les critkres modifies de Crandall (categories Crandall I et 11). Ceux qui ont bCneficie d’une reduction significative des crises mais qui continuent a presenter une epilepsie rebelle (Crandall 111) n’ont pas CtC considCres comme bCnCficiant d’un bon resultat. L’Cvolution globale se compare de faGon favorable avec celle relevCe dam d’autres centres utilisant des approches chirurgicales diffkrentes, et indique que la nkocortectomie est une procedure acceptable chez des patients hautement stlectionnes, mCme lorsque des moyens limites sont disponibles. P. Genton, Marseille)
RESUMEN Se presentan 10s resultados de un estudio de seguimiento a largo plazo de 50 pacientes que fueron sometidos a una resecci6n del neocortex temporal con preservaci6n de las estructuras limbicas profundas como tratamiento quirdrgico de epilepsia del 16bulo temporal resistente. El periodo de seguimiento oscil6 entre 3 y 15 anos. Las investigaciones electroencefalogrdficas preoperatorias se basaron solamente en las descargas interictales. Tres factores han sido interpretados como elementos de predicci6n de un buen resultado: (1) un claro foco anterior y temporal medio unilateral (p < 0.01), (2) comienzo estereotipado del ataque del 16bulo temporal (p < 0.005) y (3) mayor cantidad de volumen de tejido estirpado durante la cirugia (p < 0.05). Los resultados generales mostraron que un 62% se seguieron de “curaci6n” o “casi curaci6n” segdn una versi6n modificada de 10s criterios de Crandall (Crandall I y 11). Aquellos casos que experimentaron una reducci6n significativa del ndmero de ataques pero que continuaron sufriendo epilepsia resistente al tratamiento (grupo I11 de Crandall) se clasificaron como resultados pobres. En general este resultado, m8s favorable que 10s obtenidos en otros centros utilizando diferentes tkcnicas quirdrgicas, indica que la neurocorticectomfa es un procedimiento apropiado Epilepsia, Vol. 33, No. 6 , 1992
1000
N . MORI AND T. FUKATSU
en una poblacion bien seleccionada incluso cuando solo se dispone de una tecnologia limitada. (A. Portera-Sanchez, Madrid)
ZUSAMMENFASSUNG Es werden die Langzeitergebnisse von 50 Patienten dargestellt, die wegen unkontrollierbarer Temporallappenepilepsie einer tempoarlen Neokortektomie unter Erhaltung der tiefen limbischen Strukturen unterzogen wurden. Die Nachbeobachtung liegt zwischen 3 und 15 Jahren. Praoperative EEG-Untersuchungen umfaRten nur die interiktalen Entladungen. Fur eine giinstige Prognose sprechen drei Faktoren: 1. eindeutiger, einseitiger
Epilepsia, Vol. 33, N o . 6 , 1992
Fokus temporal vorne bis mitte (p 0.01); 2. stereotyper Beginn des Temporallappenanfalls (p < 0.01); 3. Entfernung eines groRen Gewebevolumens bei der Operation ( p < 0.05). Das Gesarntergebnis lag bei 62% geheilter oder fast geheilter Patienten gemaR den rnotifizierten Crandall-Kriterien (I 11). Patienten mit Anfallsreduktion aber unbeeinflufibarer Epilepsie wurden nicht als gebessertbetrachtet. Das Gesamtergebnis ist im Vergleich rnit anderen Zentren rnit unterschiedlichem chirurgischen Vorgehen giinstig und zeigt, daR die Neokortektomie ein angemessenes Vorgehen ist bei einer hochselektionierten Patientengruppe, selbst wenn die zur Vergiigung stehenden diagnostischen Moglichkeiten begrenzt sind.
+
(C. G. Lipinski, HeidelberglNeckarRemiind)
