September 2000
`
`Chem. Pharm. Bull. 48(9) 1293—1298 (2000)
`
`1293
`
`Interactions between Local Anesthetics and Na1 Channel Inactivation
`Gate Peptides in Phosphatidylserine Suspensions as Studied by 1H-NMR
`Spectroscopy
`
`Yoshihiro KURODA,* Kazuhide MIYAMOTO, Kazufumi TANAKA, Yoshitaka MAEDA, Junya ISHIKAWA,
`Ryo-ichi HINATA, Akira OTAKA, Nobutaka FUJII, and Terumichi NAKAGAWA
`Graduate School of Pharmaceutical Sciences, Kyoto University, Kyoto 606–8501, Japan.
`Received March 29, 2000; accepted May 26, 2000
`
`Interactions between local anesthetics and a sodium channel inactivation gate peptide (Ac–GGQDIFM-
`TEEQK–NH2, MP-1A), which was dissected from the cytoplasmic linker between domains III and IV of the
`sodium channel aa-subunit (G1484—K1495 in rat brain type IIA), have been studied by 1H-NMR spectroscopy.
`Changes in 1H-NMR chemical shifts of the aromatic proton resonances of dibucaine (pH 7.0) and lidocaine (pH
`6.0 and 9.0) in phosphatidylserine (PS) suspensions were observed. The effects of substitution of glutamine
`(F1489Q; MP-2A) or D-phenylalanine (MP-1A9) for L-phenylalanine (F1489) in MP-1A and the effects of substi-
`tution of neutral amino acid residues for the corresponding acidic amino acid residues (D1487N, MP-1NA;
`E1492Q, MP-1QEA; E1493Q, MP-1EQA) in MP-1A, on the aromatic 1H-NMR cheimcal shift changes of dibu-
`caine and lidocaine were also investigated. From these results it was concluded that: the aromatic ring of pheny-
`lalanine of MP-1A and the aromatic ring of the cationic form of dibucaine or lidocaine are interacting by pp–pp
`stacking; the tertiary amine nitrogen of dibucaine is interacting electrostatically with D1487, whereas that of li-
`docaine is interacting with E1492.
`1H-NMR; dibucaine; lidocaine; phosphatidylserine; Na1 channel; inactivation gate peptide
`
`Key words
`
`thetics, because there exist negatively charged amino acids
`only on both sides of the IFM motif (D1487, E1492, E1493)
`in the III–IV linker (Fig. 1). We have proposed a new concept
`that can satisfy the requirements from both the specific re-
`ceptor theory and the non-specific general perturbation the-
`ory: a local anesthetic molecule, located at the polar head-
`group region of the so-called boundary lipids in the vicinity
`of the Na1 channel pore, is also interacting by p-p stacking
`interactions with the phenylalanine residue in the III-IV
`linker, and by electrostatic interactions with one of the nega-
`tively charged amino acid residues.17) We have studied inter-
`actions between a model peptide MP-1 (Ac–GGQDIFM-
`TEEQK–OH), which was dissected from the III–IV linker
`(G1484–K1495, Fig. 1), and dibucaine (Fig. 2) by 1H-NMR
`sepctroscopy. We found that the quinoline ring of dibucaine
`can interact with the aromatic ring of Phe by p–p stacking of
`the rings. However, we could not specify which negatively
`charged amino acid residues are interacting electrostatically
`with the tertiary amine nitrogen of dibucaine. In the present
`study,
`in order
`to
`specify
`the negatively charg-
`ed amino acid residue, we have synthesized MP-1A
`(Ac–GGQDIFMTEEQK–NH2), which was amidated at the
`C-terminus of MP-1, and some related peptides in which
`some acidic amino acid residues (Asp, Glu) are substituted
`by the corresponding neutral amino acid residues (Asn, Gln),
`and investigated the interactions with dibucaine in sonicated
`phosphatidylserine (PS) suspensions by 1H-NMR spec-
`troscopy. In order to obtain information on the role of the
`cationic charge at the tertiary amine nitrogen of a drug, we
`also studied the interactions between the peptides and lido-
`caine at both pH 6.0 (cationic form) and pH 9.0 (neutral
`form), since we could not use dibucaine under alkaline con-
`ditions on account of its low solubility.
`
`Local anesthetics are chemicals that block action poten-
`tials in excitable membranes.1,2) Their receptor sites are now
`considered to be within a sodium channel a-subunit.3,4) This
`hypothesis is called a specific receptor theory.4) However,
`there still remains the hypothesis that considers the receptor
`site to be located within the lipid membrane,5,6) because a
`wide variety of different chemicals which include neutral and
`both negatively and positively charged molecules can act as
`local anesthetics.5,7) There are good correlations between the
`membrane concentrations of anesthetics and their poten-
`cies,8) suggesting that local anesthetics are binding at a hy-
`drophobic region of the lipid membrane. This hypothesis is
`called a non-specific general perturbation theory.4) Since both
`theories have their own rationalizations, a new concept which
`involves both theories is necessary for elucidating the molec-
`ular mechanisms of local anesthesia.
`The sodium channel a-subunit consists of four homolo-
`gous domains, I—IV, each with six transmembrane seg-
`ments, S1—S6, (Fig. 1).9) The cytoplasmic linker between
`domains III and IV of the sodium channel a-subunit (III–IV
`linker) is known to play a decisive role in a fast inactivation
`process.10,11) The inactivation gate is considered to close by
`hydrophobic interactions between the three adjacent hy-
`drophobic amino acids, Ile-Phe-Met (I1488-F1489-M1490,
`Fig. 1), and their receptors which are considered to be com-
`posed of the short S4–S5 loops of both domains III12) and
`IV.13—15) Bennett et al. have demonstrated that removal of the
`fast inactivation by mutations in the IFM motif to QQQ in
`human heart sodium channels (hH1) results in loss of a high-
`affinity inactivated-state block of the channel by lidocaine.16)
`Their results suggest that the III–IV linker, especially the
`IFM motif, is a determinant of the local anesthetic binding
`site. In our previous paper,17) we addressed the III–IV linker,
`especially the three adjacent hydrophobic amino acids (IFM)
`moiety as a docking site of charged amine type local anes-
`* To whom correspondence should be addressed.
`
`e-mail : yokuroda@pharm.kyoto-u.ac.jp
`
`© 2000 Pharmaceutical Society of Japan
`
`ALL 2088
`PROLLENIUM V. ALLERGAN
`IPR2019-01505 et al.
`
`

`

`1294
`
`Vol. 48, No. 9
`
`Fig. 1. Schematic Representation for the Sodium Channel a Subunit
`The amino acid sequence of the cytoplasmic linker between domains III and IV of the rat brain type IIA sodium channel is shown at the bottom using one-letter symbol; the
`amino acid sequence corresponding to MP-1A is between G1484 and K1495.
`
`Fig. 2. Structures of Dibucaine and Lodocaine
`
`Experimental
`Materials Dibucaine hydrochloride, lidocaine hydrochloride and bovine
`brain L-a-PS were obtained from Sigma and used without further purifica-
`tion. All the peptides were synthesized automatically by the solid phase
`method using 9-fluorenylmethoxycarbonyl (Fmoc) chemistry on an Applied
`Biosystems 433A peptide synthesizer; their N-termini were acetylated (de-
`noted by Ac–) and their C-termini were amidated (denoted by–NH2). They
`were purified on a reverse-phase C18 high-performance liquid chromatogra-
`phy column using a gradient of 90% A, 10% B to 70% A, 30% B, where A
`is 0.1% trifluoroacetic acid (TFA) in water and B is 0.1% TFA in acetoni-
`trile; the rate of decrease in A was 20%/40 min. They were characterized by
`ion spray mass spectrometry. The amino acid sequences of the peptides were:
`MP-1A, Ac–GGQDIFMTEEQK–NH2; MP-1A9, Ac–GGQDIFMTEEQK–NH2
`(F means D-phenylalanine); MP-2A, Ac–GGQDIQMTEEQK–NH2; MP-1NA,
`Ac–GGQNIFMTEEQK–NH2; MP-1QEA, Ac–GGQDIFMTQEQK–NH2; MP-
`1EQA, Ac-GGQDIFMTEQQK-NH2.
`Preparation of Sample Solutions Single bilayer vesicles (liposomes)
`were prepared by ultrasonic irradiation of an isotonic (310 mOsm, 150 mM)
`phosphate buffer (in D2O) suspension of dried PS for 20 min, cooling in an
`ice/water bath and bubbling with nitrogen gas. A weighed amount of a drug
`(3 mM) dissolved in the buffer and/or a peptide (3 mM) was added to the sus-
`pension of the pre-formed vesicles (15 mM).
`Measurements The 1H-NMR experiments were carried out on a Bruker
`AM-600 (600.13 MHz) spectrometer with a digital resolution of 0.25 Hz per
`
`1H-NMR Spectra of (a) Dibucaine in a Phsophate Buffer (pH 7.0),
`Fig. 3.
`(b) Dibucaine in PS Suspensions (pH 7.0), and (c) Dibucaine and MP-1A in
`PS Suspensions (pH 7.0)
`
`point. Ambient probe temperature was 27 °C. Chemical shifts were refer-
`enced to internal TSP (3-trimethylsilylpropionic acid-d4).
`
`Results and Discussion
`Figure 3 shows a typical example of 1H-NMR spectra of
`(a) dibucaine in a phsophate buffer (pH 7.0), (b) dibucaine in
`PS suspensions (pH 7.0), and (c) dibucaine and MP-1A in PS
`suspensions (pH 7.0), respectively. At pH 7.0, most of the
`58.0).18)
`dibucaine molecules exist as a cationic form (pKa
`Since the peaks due to the drug, PS, and the peptide over-
`lapped with one another at a lower frequency region than that
`of the HDO resonance, we followed the changes in the chem-
`ical shifts of the aromatic proton resonances of the drug and
`the phenylalanine residue of the peptide. The assignments for
`the quinoline ring proton resonances were 3, 6, 7, 8, and 5,
`respectively, from low to high frequency.17,19) The spin-cou-
`pled sharp resonances at around 7.2—7.3 ppm in Fig. 3c
`
`

`

`September 2000
`
`1295
`
`Fig. 4. Changes in Chemical Shifts (in Units of Hz at 600.13 MHz) of the Quinoline Ring Proton Resonances (at 3, 5, 6, 7, and 8 Positions) of Dibucaine
`as Caused by the Peptides
`(a) MP-1A, (b) MP-2A, (c) MP-1A9, (d) MP-1NA, (e) MP-1QEA, and (f) MP-1EQA.
`
`were due to the aromatic ring protons of the phenylalanine
`residue of MP-1A; the assignments were ortho, para, and
`meta protons from low to high frequency, respectively. The
`quinoline ring proton resonances were broadened by the
`presence of PS suspensions (Fig. 3b), indicating that dibu-
`caine partitioned into the lipid bilayer. The resonance at the 5
`position shifted to higher frequency, whereas all the remain-
`ing resonances (8, 7, 6, and 3 positions) shifted to lower fre-
`quency in PS suspensions. All these quinoline ring proton
`resonances shifted slightly to lower frequencies when MP-1A
`was added (spectrum c). Since in the PS suspensions, the res-
`onance at the 5 position and those at the 8, 7, 6, and 3 posi-
`tions shifted to the opposite direction with each other as
`compared to those in the phosphate buffer solution, the
`changes in the chemical shifts from spectrum b to spectrum c
`are not due to the change in the partition of dibucaine into
`the PS bilayers which might be caused by the peptide. Ob-
`served changes in chemical shifts (in units of Hz at 600.13
`MHz) by MP-1A are summarized graphically in Fig. 4 to-
`gether with the results by the other peptides.
`It can be seen that: 1) the quinoline ring proton resonances
`of dibucaine were shifted to a low frequency by MP-1A (Fig.
`4a), whereas to a high frequency by MP-2A (Fig. 4b); 2)
`MP-1A9 (Fig. 4c) caused smaller chemical shift changes than
`did MP-1A; 3) the magnitudes of the changes in chemical
`shifts were the smallest by MP-1NA (Fig. 4d), while the
`
`largest were by MP-1EQA (Fig. 4f); 4) MP-1EQA caused
`chemical shift changes more than twice as much as those
`done by MP-1A (Fig. 4a) or by MP-1QEA (Fig. 4e). Obser-
`vation 1 suggests that the aromatic ring of the phenylalanine
`residue of MP-1A and the quinoline ring of dibucaine are in-
`teracting by p–p stacking with each other. Observation 2
`supports this view and suggests that to interact by the p–p
`stacking, an appropriate stereochemical configuration is re-
`quired. In our previous paper,17) we also deduced this p–p
`stacking interaction by comparing the changes in chemical
`shifts of the quinoline ring proton resonances caused by the
`peptides, MP-1 and MP-2; MP-2 is the F1489Q substituted
`peptide of MP-1. In the case of MP-1, all the quinoline ring
`proton resonances are shifted to a low frequency in a similar
`manner as in MP-1A, but to a much larger extent than by
`MP-1A. Moreover, MP-2 also caused a low frequency shift
`to the quinoline ring proton resonances, although the magni-
`tudes were smaller than those caused by MP-1.17) Since the
`C-termini of both MP-1 and MP-2 are not amidated, the C-
`terminal carboxyl group appears to induce the low frequency
`shift to the quinoline ring proton resonances of dibucaine.
`Observation 3, especially for MP-1NA is rather dramatic and
`means that D1487 is indispensible for the interaction with
`dibucaine, while observation 4 means that E1493 is interfer-
`ing with the binding of dibucaine.
`In Fig. 5, we summarized the changes in chemical shifts of
`
`

`

`1296
`
`Vol. 48, No. 9
`
`1H-NMR Spectra of (a) Lidocaine in a Phsophate Buffer (pH 9.0),
`Fig. 6.
`(b) Lidocaine in PS Suspensions (pH 9.0), and (c) Lidocaine and MP-1A in
`PS Liposomes (pH 9.0)
`
`Fig. 5. Changes in Chemical Shifts (in Units of Hz at 600.13 MHz) of the
`Aromatic Ring Proton Resonances of F1489 as Caused by Dibucaine
`m: meta, p: para, and o: ortho positions. (a) MP-1A, (b) MP-1A9, (c) MP-1NA, (d)
`MP-1QEA, and (e) MP-1EQA.
`
`the aromatic proton resonances of the phenylalanine residues
`of (a) MP-1A, (b) MP-1A9, (c) MP-1NA, (d) MP-1QEA, and
`(e) MP-1EQA in PS suspensions as caused by dibucaine. The
`aromatic proton resonances were shifted to a low frequency
`by dibucaine in MP-1A, MP-1A9, and MP-1NA, whereas to a
`high frequency in MP-1QEA and MP-1EQA. All these
`chemical shift changes may include those due to changes in
`the strength of the interaction of the peptides with PS sus-
`pensions as caused by dibucaine. The changes in chemical
`shifts due to the interaction of a peptide with PS suspensions
`were found to be as large as 21.8 Hz for all the peptides.
`Thus, the observed high frequency shifts of MP-1QEA (Fig.
`5d) and MP-1EQA (Fig. 5e), especially of MP-1EQA can be
`considered to be due to relatively strong interaction of dibu-
`caine with the peptide, which is also seen in Fig. 4f.
`Figure 6 shows a typical example of 1H-NMR spectra of a)
`lidocaine in a phsophate buffer (pH 9.0), b) lidocaine in PS
`suspensions (pH 9.0), and c) lidocaine and MP-1A in PS sus-
`pensions (pH 9.0), respectively. Aromatic proton resonance
`regions are shown here for brevity. At pH 6.0, the overall
`spectral features were the same as in Fig. 6 except that spec-
`trum c contained some amide proton resonances due to the
`peptide. The lidocaine molecules exist as a cationic form at
`57.86).20) The meta
`pH 6.0 and a neutral form at pH 9.0 (pKa
`and para ring proton resonances of lidocaine (Fig. 2), which
`were observed as sharp spin-coupled peaks at around 7.2
`ppm (spectrum a), became a broad single peak and were
`shifted to a low frequency by the interaction with PS lipo-
`somes. The broad peak in spectrum b shifted to a high fre-
`
`Fig. 7. Changes in Chemical Shifts (in units of Hz at 600.13 MHz) of the
`Aromatic Ring Proton Resonances of Lidocaine as Caused by the Peptides
`(a) MP-1A, (b) MP-2A, (c) MP-1A9, (d) MP-1NA, (e) MP-1QEA, and (f) MP-
`1EQA.
`
`quency in the presence of MP-1A (spectrum c). Thus in the
`case of lidocaine, the changes in chemical shifts of the aro-
`matic proton resonances from spectrum b to spectrum c may
`include changes in the partitioning of lidocaine into PS lipo-
`somes which are caused by the peptide. All the observed
`changes in chemical shift of lidocaine by MP-1A and also by
`the other peptides at pH 6.0 and 9.0 are summarized in Fig.
`7. It is seen that: 1) in all the cases (Figs. 7a—f) and at both
`pH 6.0 and 9.0, the aromatic proton resonances of lidocaine
`were shifted to a high frequency by the peptides; 2) these
`high frequency shifts were much larger than those observed
`
`

`

`September 2000
`
`1297
`
`in dibucaine (Fig. 5); 3) MP-1NA (Fig. 7d) caused the largest
`chemical shift changes, whereas MP-1QEA (Fig. 7e) caused
`the least changes at both pH 6.0 and 9.0; 4) MP-1A9 (Fig. 7c)
`caused smaller chemical shift changes than did MP-1A (Fig.
`7a) at pH 6.0; 5) in MP-1A, MP-1NA, and MP-1EQA, each
`of the low frequency shifts was larger at pH 6.0 than the cor-
`responding shift change at pH 9.0; 6) at pH 9.0, the shift
`changes by MP-2A (Fig. 7b) and MP-1A9 (Fig. 7c) were
`larger than by MP-1A (Fig. 7a). Although it is difficult to
`rule out the possibility that the large high frequency shift
`(observations 1 and 2) was due to the decrease in the parti-
`tion of lidocaine into PS liposomes, observed variations in
`the magnitude of the changes in chemical shifts appear to in-
`dicate the differences in the strength and the mode of the in-
`teraction between lidocaine and the peptides. Observation 3
`can thus be considered to mean that D1487 interfered with
`the interaction, while E1492 played an important role in the
`interaction with lidocaine. Moreover, observation 4 appears
`to be a manifestation of the importance of the stereochemical
`configuration around F1489 for the interaction. Observation
`5 suggests that the electrostatic interaction between the posi-
`tive charge at the tertiary amine nitrogen of lidocaine and the
`negative charge at E1492 strengthened the binding of lido-
`caine with the peptide. Although both MP-1A (Fig. 7a) and
`MP-2A (Fig. 7b) caused a high frequency shift, the magni-
`tude was larger in MP-1A than in MP-2A at pH 6.0. Thus the
`p–p stacking interaction may be present between the aro-
`matic ring of F1489 and that of the cationic form of lido-
`caine. However at pH 9.0, since observation 6 indicates that
`the presence of F1489 and the stereochemical configuration
`around F1489 are not important for the binding, the p–p
`stacking interaction is not operating or weak, if any, in the in-
`teraction for the neutral form of lidocaine. The 2H-NMR
`quadrupole splittings of the aromatic ring of tetracaine indi-
`cate that charged and uncharged tetracaine occupy different
`sites in the PS bilayer;21) the latter binds at a more ordered
`environment than the former. In analogy with tetracaine, un-
`charged lidocaine may also be binding with PS at a more or-
`dered environment than its charged counterpart, probably a
`little deeper into the PS bilayer. This may be the reason why
`the p–p stacking interaction was weakened in the interaction
`with the peptide for the neutral form of lidocaine.
`Figure 8 shows changes in the chemical shifts of the aro-
`matic proton resonances of the phenylalanine residues of (a)
`MP-1A, (b) MP-1A9, (c) MP-1NA, (d) MP-1QEA, and (e)
`MP-1EQA in PS suspensions as caused by lidocaine and at
`pH 6.0 and 9.0. It is seen that: 1) except for MP-1QEA (Fig.
`8d), the aromatic protons of the phenylalanine residues
`shifted to a high frequency at pH 6.0 and oppositely at pH
`9.0; 2) the magnitudes were smaller than those observed with
`interactions for dibucaine (Fig. 5); 3) at pH 6.0, MP-1NA
`showed the largest high frequency shift; 4) at pH 9.0, in all
`the cases (Figs. 8a—e), the phenylalanine ring protons
`showed a very small low frequency shift. Observation 3)
`again means that at pH 6.0, D1487 is interfering with the in-
`teraction for lidocaine. The fact that MP-1QEA showed an
`opposite chemical shift change as compared to the other pep-
`tides means that E1492 had been playing a key role in the in-
`teraction. Finally observation 4) implies that the neutral form
`of lidocaine does not interact with F1489 as discussed above.
`In conclusion, the cationic forms of dibucaine and lido-
`
`Fig. 8. Changes in Chemical Shifts (in Units of Hz at 600.13 MHz) of the
`Aromatic Ring Proton Resonances of F1489 as Caused by Lidocaine at pH
`6.0 and pH 9.0
`(a) MP-1A, (b) MP-1A9, (c) MP-1NA, (d) MP-1QEA, and (e) MP-1EQA.
`
`caine interact differently with MP-1A. The tertiary amine ni-
`trogen of dibucaine interacts electrostatically with the nega-
`tive charge of D1487, while its quinoline ring interacts with
`the aromatic ring of F1489. On the other hand, the tertiary
`amine nitrogen of lidocaine interacts electrostatically with
`the negative charge of E1492, while its aromatic ring inter-
`acts with the aromatic ring of F1489.
`Finally, it should be added that since the absolute magni-
`tudes of the presently observed changes in chemical shifts of
`the aromatic proton resonances of the local anesthetics and
`those of the phenylalanine residues of the peptides were
`small, other explanations not including the p–p stacking in-
`teractions cannot be excluded for explaining the differences
`in the magnitudes of the chemical shift changes among the
`peptides. Differences in the magnetic anisotropy effects be-
`tween the side-chains of the amino acids and/or the confor-
`mational changes due to substitution of one of the amino
`acids in a peptide can be a candidate for explaining the rea-
`son for the differences in the observed chemical shift
`changes. To verify these hypotheses and for a more thorough
`discussion, determination of the structures of the peptides in
`solution is required. Structural studies on the peptides includ-
`ing the IMF motif in trifluoroethanol solutions and in mi-
`celles are currently being undertaken and some results have
`been reported elsewhere.22)
`
`

`

`1298
`
`Vol. 48, No. 9
`
`References
`1) Strichartz G. R., Ritchie J. M., “Local Anesthetics,” ed. by Strichartz
`G. R., Springer-Verlag, New York, 1987, pp. 21—52.
`2) Courtney K. R., Strichartz G. R., “Local Anesthetics,” ed. by Strichartz
`G. R., Springer-Verlag, New York, 1987, pp. 53—94.
`3) Richards C. D., Keightley C. A., Hesketh T. R., Metcalfe J. C., Prog.
`Anesthesiol., 2, 337—351 (1980).
`4) Hille B., Prog. Anesthesiol., 2, 1—6 (1980).
`5) Lee A. G., Nature (London), 262, 545—548 (1976).
`6) Trudell J. R., Anesthesiology, 46, 5—10 (1977).
`7) Lee A. G., Anesthesiology, 51, 64—71 (1979).
`8) Seeman P., Pharmacol. Rev., 24, 583—655 (1972).
`9) Noda M., Ikeda T., Kayano T., Suzuki H., Takeshima H., Kurasaki M.,
`Takahashi H., Numa S., Nature (London), 320, 188—192 (1986).
`10) West J. W., Patton D. E., Scheuer T., Wang Y., Goldin A. L., Catterall
`W. A., Proc. Natl. Acad. Sci. U.S.A., 89, 10910—10914 (1992).
`11) Patton D. E., West J. W., Catterall W. A., Goldin A. L., Proc. Natl.
`Acad. Sci. U.S.A., 89, 10905—10909 (1992).
`12) Smith M. R., Goldin A. L., Biophys. J., 73, 1885—1895 (1997).
`13) Tang L., Kallen R. G., Horn R., J. Gen. Physiol., 108, 89—104 (1996).
`
`14) Lerche H., Peter W., Fleischhauser R., Pika-Hartlaub U., Malina T.,
`Mitrovic N., Lehmann-Horn F., J. Physiol., 505. 2, 345—352 (1997).
`15) McPhee J. C., Ragsdale D. S., Scheuer T., Catterall W. A., J. Biol.
`Chem., 273, 1121—1129 (1998).
`16) Bennett P. B., Valenzuela C., Chen L. Q., Kallen R. G., Circ. Res., 77,
`584—592 (1995).
`17) Kuroda Y., Ogawa M., Nasu H., Terashima M., Kasahara M., Kiyama
`Y., Wakita M., Fujiwara Y., Fujii N., Nakagawa T., Biophys. J., 71,
`1191—1207 (1996).
`18) Kuroda Y., Wakita M., Nakagawa T., Chem. Pharm. Bull., 42, 2418—
`2425 (1994).
`19) Kuroda Y., Fujiwara Y., Biochim. Biophys. Acta, 903, 395—410
`(1987).
`20) Truant A. P., Takman F. K., Anesthesia and Analgesia, 38, 478—484
`(1959).
`21) Kelusky E. C., Boulanger Y., Schreier S., Smith I. C. P., Biochim. Bio-
`phys. Acta, 856, 85—90 (1986).
`22) Kuroda Y., Miyamoto K., Matsumoto M., Maeda Y., Kanaori K.,
`Otaka A., Fujii N., Nakagawa T., J. Peptide Res., (2000) in press.
`
`