Vol. 42, No. 3, July 1997
BIOCHEMISTRY and MOLECULAR BIOLOGY INTERNATIONAL Pages 641-647
SYNTHESIS AND IMAGING OF BLOOD-BRAIN-BARRIER PERMEABLE NITROXYLPROBES FOR FREE RADICAL REACTIONS IN BRAIN OF LIVING MICE Hiroaki Sano, Ken-ichiro Matsumoto, and Hideo Utsumi* Department of Biophysics, Faculty of Pharmaceutical Sciences, Kyushu University, Higashi-ku, Fukuoka 812-82, Japan Received April 14, 1997
SUMMARY: Three different lipophilic nitroxyl-probes having capability to pass the blood-brain barrier were, for the ftrst time, synthesized to estimate free radical reactions in brain of living animals. Two of the three were designed to be hydrolyzed by esterase and remain in cell. All 3 probes had high n-octanolfouffer partition coefficients and gave 2 signal components in in vivo ESR spectra at head of living mice after intravenous injection. The ESR parameters of 2 components agreed with those of probes dissolved in water and lipidic phases. The ESR-CT imaging on the nilroxyl-proves after intravenous injection revealed that all probes presented in both encephalon and extracranial region of head. Tissue distribution of the nitroxyl-probes demonstrated that the newly synthesized lipophilic nitroxyl-probes had capability to pass the blood-brain barrier and accumulated in brain than that of hydrophilic probe. KEY WORDS: ESR, EPR, ESR-imaging, in vivo ESR, free radicals, nitoroxide probe, nitroxyl radicals, spin probe, blood-brain barrier, active oxygen species, INTRODUCTION Free radical reactions in brain are one of most interesting subjects for brain-research. Free radicals and reactive oxygen species have been implicated in brain ischemia-reperfusion injury (1), brain tumor (2), aging (3), familial amyotrophic lateral sclerosis (4), and other several neurodegenerative diseases (5). There have been, however, few works which determined in vivo free radical reactions in brain directly, because of lack of adequate method. Nitroxyl radicals are susceptible to oxygen concentration, active oxygens and biological redox systems, and are widely used as probes for in vivo ESR measurement (6-10). In vivo ESR signal of exogenous nitroxyl radicals decreased gradually in living mice (6) and its decay rate depended on physiological and pathological conditions such as aging (11), hyperoxia (12), ischemia-reperfusion injury (13), carbon-tetrachloride induced liver injury (14), etc. The enhancement of decay rate by oxidative stress was inhibited by exogenous antioxidants (15). This work was in part supported by Grant-in-Aids for Research on Priority Areas, for Cooperative Research, for General Science Research, and for Developmental Scientific Research from the Ministry of Education, Science, Sports and Culture of Japan, by Naito Foundation, and by the Cosmetology Research Foundation. Abbreviations: PROXYL; 2,2, 5, 5-te~amethyl-pyrrolidine-1-oxyl, doxyl-butane; 2-ethyl-2,4,4trimethyl-oxazolidine-3-oxyl *To whom all correspondence should be addressed. (Utsumi H: TEL; +81-92-642-6621, FAX; +81-92-642-6626, E-mall;
[email protected]) 1039-9712/97/030641~)7505.00/0 641
Copyright 9 1997 by Academic Press Australia. All rights of reproduction in any form reserved.
Vol. 42, No. 3, 1997
BIOCHEMISTRY and MOLECULAR BIOLOGY INTERNATIONAL
Most commercially available nitroxyl-probes are water-solubility and have no capability to pass the blood-brain barrier, causing it difficult to apply in vivo ESR measurement to evaluation of free radical reactions in brain. The desirable natures of nitroxyl-probes for investigating the brain redox reaction are high capabilities to pass the blood-brain barrier, to remain/accumulate in brain and to be susceptible moderately towards redox reaction in the brain. Lipophilic ester compounds such as radioactive probes for positron emission tomography were reported to pass the blood-brain barrier and to be metabolized specifically by esterase into hydrophilic metabolites in brain (16). In this study, we synthesized, for the first time, lipophilic and blood-brain-barrier permeable nitroxyl-probes,
3-methoxycarbonyl-2,2,5,5-tetramethyl-pyrrolidine-l-oxyl (methoxycarbony-
PROXYL), 3-ethoxycarbonyl-2,2,5,5-tetra-methylpyrrolidine-l-oxyl (ethoxycarbonyl-PROXYL) and 2-ethyl-2,4,4-trimethyl-oxazolidine-3-oxyl(doxyl-butane). The former two were designated to be metabolized by esterase into hydrophilic ones and to remain in the brain. We succeeded in the ESR-CT imaging of these nitroxyl-probes at head of living mice.
MATERIALS AND METHODS Synthesis of lipophilic nitroxyl-probes: 3-Methoxycarbonyl- and 3-ethoxycarbonyl-2,2, 5,5tetramethyl-pyrrolidine-l-oxyl were synthesized by esterification of 3-carboxy-2,2,5,5-tetramethylpyrrolidine-l-oxyl (carboxy-PROXYL) as follows. Carboxy-PROXYL (5.0g, Aldrich Chemical Co. Inc., Milwaukee, WI) was dissolved in methanol or ethanol (200 ml), and concentrated hydrochloric acid (2.0 ml) was added to the solution under stirring in ice bath. The mixture was kept at room temperature for 2 days. After reoxidizing the reduced products from nitroxyl radicals during reaction by babbling oxygen gas, 100 ml of ice-cold water was added and then the products were extracted with ethylether. The presence of several compounds involving carboxy-PROXYL, its ester and their reduced forms, were confLrrned by TLC and ESR experiments. The ether layer was washed once with saturated sodium hydrogen bicarbonate and twice with saturated sodium chloride solution. After drying with magnesium sulfate, ether was evaporated and applied to silica gel column chromatography (Wakogel C-200, Wako Pure Chemicals, LTD., Osaka, elution with hexane-ether). The purity was estimated with TLC, ESR and MS measurements. 2-Ethyl-2,4,4-trimethyl-oxazofidine-3-oxyl(doxyl-butane) was synthesized by the method of Keana, et al. (17) with slight modification because of high volatility of butanone, and purified with column chromatography as described above. Partition coefficients between n-octanol and buffer: Nitroxyl-probes (150 mM) were dissolved in isotonic phosphate-buffered saline (pH 7.4). One ml of diluted nitroxyl-probe solution (1 mM) was vigorously mixed with 1.0 ml of n-octanol, and then the mixture was centrifuged at 3,000 r.p.m, for 5 min. The amounts of nitroxyl-probe in both n-octanol and buffer were calculated from double-integrated ESR signal intensity as described below. Mn2+ was employed for correction of sensitivity between n-octanol and buffer. In vivo ESR measurement and ESR-CT imaging; Female ddY mice (4-5 weeks old, 17-23g body weight) were used throughout this study. Mice were anesthetized with Nembutal (Dinabot, Osaka), and injected with 100 I.tl of nitroxyl nitroxyl-probe intravenously via tail vein. In vivo ESR spectra were measured with an L-band ESR spectrometer (JEOL, JES-RE-3L) at 1.2 GHz of microwave frequency, 1.0 mW of the power, and 0.1 mT of 100 kHz field modulation. 2D-image of nitroxyl-probes at mice head was obtained with a home-made ESR-CT system as reported previously (18). Head of mouse was restrained in a loop-gap resonator (35 mm in diameter, 5 mm in axial length) of an L-band ESR spectrometer having field gradient coils (x-, yand z-axis, 0.60 mT/cm). The spectra with and without field gradient were alternately obtained, changing the direction in 10-degree steps, providing 18 projections. The spectral data under field
642
Vol. 42, No. 3, 1997
BIOCHEMISTRYond MOLECULAR BIOLOGY INTERNATIONAL
gradient was deconvoluted with those under zero-field gradient by Fourier transform method, and image was reconstructed from 18 projections of the deconvoluted data by a filtered back projection. Determination of tissue distribution of nitroxyl-probes; Tissue distribution of nitroxyl-probes was determined as follows. Blood was obtained by cutting carotid of mice 3 rain. after intravenous injection of nitroxyl-probes (100 gl, 150 mM). Brain, liver, kidney, heart, lung, spleen, stomach and femoral muscle were rapidly dissected and homogenized in 9-fold ice cold phosphate-buffered saline (pH 7.4). Blood or homogenates was transferred into capillary tubes and then observed with an X-band spectrometer (JEOL RE-IX) at 9.4 GHz of microwave frequency, 5.0 mW of the power, and 0.1 mT of 100 kHz field modulation. The spin-recovery was determined by calibration of double-integrated signal intensity with that of 1,1-diphenyl-2-picrylhydrazyl as a standard. The reduced form of nitroxyl-probes was oxidized by addition of potassium ferricyanide (19).
RESULTS and DISCUSSION Partition coefficients of carboxy-, methoxycarbonyl-, ethoxycarbonyl-PROXYL and doxylbutane between n-octanol and buffer were determined as indices of lipophilicity of nitroxyl-probes (Table 1). Doxyl-butane has the largest partition coefficient and carboxy-PROXYL has the lowest one. The introduction of ester to carboxy-PROXYL increased lipophilicity, and methoxycarbonyland ethoxycarbonyl-PROXYL had relatively large partition coefficients, indicating the possibility that 3 lipophilic nitroxyl-probes, methoxycarbonyl-, ethoxycarbonyl-PROXYL and doxyl-butane, pass the blood-brain barrier. Figure 1 demonstrates typical ESR spectra of nitroxyl-probes at head of mice after intravenous injection. Carboxy-PROXYL gave triplet lines with the same peak height and 1.60 mT of hyperfine splitting, which agreed with the previous observation using the other hydrophilic probes (12). The ESR spectra of methoxycarbonyl-, ethoxycarbonyl-PROXYL and doxyl-butane composed of two Iriplet components. The hyperfine splittings of each component were 1.45 mT and 1.60 mT, which are indicated with cross and circle, respectively, in figure 1, agreed with those of nitroxyl radical dissolved in lipidic and water phase, respectively (20). This indicates that methoxycarbonyl-, ethoxycarbonyl-PROXYL and doxyl-butane should present in both water and lipidic phase such as membranes in mice head after intravenous injection. The lipidic signals of methoxycarbonyl- and ethoxycarbonyl-PROXYL disappeared within 5 min, while that of doxylbutane remained significantly even 10 min after injection (data not shown). Methoxycarbonyl- and ethoxycarbonyl-PROXYL are conceivably hydrolyzed by the action of esterase as reported in ester
Table 1. Partition coefficients between n-octanol and buffer nitroxyl-probes n-octanol and buffer carboxy-PROXYL 0.02 methoxycarbonyl-PROXYL 8.7 ethoxycarbonyl-PROXYL 4.5 doxvt-butane 11.0 Partition coefficients between n-octanol and buffer were determined from the amount of nitroxyl-probe in n-octanol and buffer, which were obtained from ESR signal intensity as described in text.
643
Vol. 42, No. 3, 1997
BIOOHEMISTRYand MOLECULAR BIOLOGY INTERNATIONAL
(a)
(b) ),
1.0mT',
'm'4 I
I
o
• |
(c) -..--~,
1.0mT',
I .-~mT!
o •
(d)
',
;
~
1.-~mT
o x
',
,
~m,q t Figure 1. Typical in vivo E S R spectra at h e a d of mice a f t e r i n t r a v e n o u s injection of (a) carboxy-PROXYL, (b) methoxycarbonyl-PROXYL, (c) ethoxycarbonyl-PROXYL, (d) doxyl-butane. A hundred [tl of nitroxyl-probes (150 raM) were injected into tail vein of mice anesthetized with Nembutal, and then in vivo ESR spectra were measured at head after indicated minutes with an L-band ESR spectrometer (JEOL, JES-RE-3L) at 1.2 GHz of microwave frequency, 1.0 mW of the power, and 0.1 mT of 100 kHz field modulation. Symbols, • and O, indicate lipophilic and hydrophilic signal components, hyperfine splittings of which are 1.45 and 1,60 mT, respectively.
derivatives of radioactive (16) and fluorescent probes (21). If so, hydrolysis of ester moiety of methoxycarbonyl- and ethoxycarbonyl-PROXYL and their conversion to carboxy-PROXYL should contribute to disappearance of the lipidic signal. ESR-CT imaging was performed to determine the distribution of nitroxyl-probes at head of mice after intravenous injection. The lipophilic probes gave the 2 signal components as shown in figure 1. Therefore, central peak (h(0)) of triplet lines was used for the imaging to avoid the disturbance due to the occurrence of 2 signal components in deconvolution process with Fourier transform method. Figure 2a, b, c, and d demonstrate 2D-image of carboxy-, methoxycarbonyl-, ethoxycarbonyl-PROXYL and doxyl-butane at head of mice after intravenous injection, respectively. Distinguished contrasts were obtained not only in exlracranial domain but also in encephalon region by methoxycarbonyl-, ethoxycarbonyl-PROXYL and doxyl-butane, while carboxy-PROXYL distributed only in extracranial domain. These images clearly indicate that methoxycarbonyl-, ethoxycarbonyl-PROXYL and doxyl-butane will pass the blood-brain barrier and reach to encephalon region. In order to connrm wnemer or -or me fipophilic nitroxyl-D~be~ r ~ s the blood-brain barrier, ESR signal intensities of nitroxyl nitroxyl-probes were measured in collected blood and isolated tissues. The spin-concentration of nitroxyl nitroxyl-probe was calculated by comparing the peak area of nitroxyl signal with that of 1,1-diphenyl-2-picrylhydrazyl as a standard, and the spin
644
Vol. 42, No. 3, 1997
BIOCHEMISTRYand MOLECULAR BIOLOGY INTERNATIONAL
lencephalon re,qionl
E
0 0 0
ea
2.00cm
\
Icontours of a mouse head I
Figure 2. 2D-ESR-CT imaging a t h e a d of mice a f t e r i n t r a v e n o u s injection of (a) carboxy-PROXYL, (b) methoxycarbonyl-PROXYL, (c) ethoxycarbonyl-PROXYL, (d) doxyl-butane. 2D-image of nitroxyl-probes at mice head was obtained with a home-made ESR-CT system. Head of mouse was restrained in a loop-gap resonator of an L-band ESR spectrometer having field gradient coils (x-, y- and z-axis, 0.60 mT/cm). The spectra with and without field gradient were alternately obtained, changing the direction in 10-degree steps. The 2D-image was reconstructed from 18 projections by a filtered back projection.
Table 2. Tissue distribution of nitroxyl-probes in mice after intravenous injection carboxymethoxycarbonylPROXYL PROXYL (a) intact nitroxyl radicals (~tmol spin/g-tissue) brain 0.00i-_0.00 0.02+0.00 blood 1.18+0.06 0.37+0.07 fiver 0.08+0.02 0.01_+0.00 kidney 0.51+0.21 0.02+0.01 heart 0.28+0.13 0.03+0.02 lung 0.39+0.07 0.21+0.03 spleen 0.27+0.05 0.15+0.02 stomach 0.13+0.03 0.17+0.02 muscle 0.12+0.06 0.12+0.05
ethoxycarbonylPROXYL 0.05+0.03 0.27+0.07 0.02+0.00 0.03+0.01 0.05+0.02 0.11+0.11 0.16+0.09 0.12+0.05 0.10i-_0.04
doxylbutane 0.02+0.02 0.12+0.01 0.00+0.00 0.01+0.00 0.01+0.00 0.06+0.01 0.08+0.02 0.05+0.02 0.07+0.05
(b) sum of intact nitroxyl and its reduced form (Ixmol spin/g-tissue) brain 0.02+0.00 0.47+0.04 0.28+0.07 0.57+0.11 blood 1.27+0.08 0.40i-_0.03 0.30!-_0.08 0.38+0.08 fiver 1.15+0.28 0.99+0.12 0.72+0.25 0.72+0.29 kidney 1.65+0.27 0.55+0.05 0.54+0.07 0.57+0.03 heart 0.72+0.04 0.63+0.15 0.36+0.11 0.47+0.07 lung 0.42+0.07 0.30i-_0.04 0.31+0.09 0.27+0.04 spleen 0.47+0.04 0.34+0.08 0.22+0.07 0.37+0.08 stomach 0.16+0.02 0.18+0.02 0.14+0.05 0.06+0.01 muscle 0.18+0.06 0.27+0.05 0.18+0.06 0.26+0.08 Blood was obtained by cutting carotid of mice 3 min after injection of 100 p.1 of nitroxylprobes (150 mM). Brain, liver, kidney, heart, lung, spleen, stomach and femoral muscle were rapidly dissected and homogenized in phosphate-buffered saline. The spinrecovery was determined from peak area of ESR signal with 1,1-diphenyl-2picrylhydrazyl as a standard. The reduced form of nitroxyl-probes was oxidized by potassium ferricyanide.
645
Vol. 42, No. 3, 1997
BIOCHEMISTRYond MOLECULAR BIOLOGY INTERNATIONAL
recoveries in blood or tissue homogenates were expressed as the ratio of spin molarity in samples to g-blood or tissue. Nitroxyl radicals are easily reduced to the corresponding hydroxylamine in living mice after inlravenous injection (6). The reduced forms involving hydroxylamine are known to be reoxidized to the corresponding nitroxyl radicals by addition of ferricyanide (19). Thus, the intact probe and sum of intact plus its reduced forms were estimated from spin-recoveries before and after reoxidation by ferricyanide, respectively. Table 2a and b show spin recoveries of intact nitroxyl nitroxyl-probo and sum of intact plus its reduced forms in blood and tissues of mice 3 min after intravenous injection. The spin-recovery of intact one in brain was several-folds higher in lipophilic nitroxyl-probes than that of carboxy-PROXYL, while carboxy-PROXYL was the highest in blood. The sum of intact plus its reduced form was also 10-20 times higher in lipophilic probes than that in hydrophilic one. These data strongly suggest that methoxycarbonyl-, ethoxycarbonyl-PROXYL and doxyl-butane pass the blood-brain barrier. Nitroxyl radicals are very useful probes for in vivo ESR measurements of biological and pathological phenomena relating oxidative stress and active oxygen species (22). This indicates that the blood-brain-barrier permeable nitroxyl-probes, which we have for the first time synthesized, should offer new valuable information about in vivo free radical reactions in brain of living animals. Especially, methoxycarbonyl- and ethoxycarbonyl-PROXYL have large advantage that these compounds may be hydrolyzed and remain in brain. These newly synthesized nitroxyl-probes must become useful probes and contrast agents for in vivo ESR measurements and ESR-CT imaging. REFERENCES 1. Chart, P.H. (1996) Stroke 27, 1124-1129. 2. Cobbs, C.S., Levi, D.S., Aldape, K., and Israel, M.A. (1996) Cancer Res. 56, 3192-3195. 3. Caeney, J.M., Starke-Reed. P.E., Oliver, C.N., l_andum, R.W., Cheng, M.S., Wu, J.F., and Floyd. R.A. (1991)Proc. Natl. Acad. Sci. USA 88, 3633-3636. 4. Rose, D.R., Siddique, T., Patterson, D., Figlewicz, D.Z., Sapp, P., Hentati, A., Donaldson, D., Goto, J., O'Regan, J.P., Deng, H.-Q., Rahmani, Z., Krizus, A., McKenna-Yasek, D., Cayabyab, A., Gaston, S.M., Berger, R., Tanzi, R.E., Halperin, J.J., Herzfeld, B., van den Berg, R., Hung, W.-Y., Bird, T., Deng, G., Mulder, D.W., Smyth, C., Lang, N.G., Soriano, E., Perocak-Vance, M.A., Haines, J., Rouleau, G.A., Gusella, J.S., Horritz, H.R., and Brown R.H. Jr. (1993) Nature 362, 59-62. 5. Evans, P.H. (1993) Br. Med. Bull. 49, 577-587. 6. Utsumi, H., Muto, E., Masuda, S., and Hamada, A. (1990) Biochem. Biophys. Res. Commun. 172, 1342-1348. 7. Glockner, J.F., Chang, H.-C., and Swartz, H.M. (1991) Magn. Reson. Med. 20, 123-133. 8. Quaresima, V., Alecci, M., Ferrari, M., and Sotgiu, A. (1992) Biochem. Biophys. Res. Commun. 183, 829-835. 9. Komarov, A.M., Joseph, J., and Lai, C.-S. (1994) Biochem. Biophys. Res. Commun. 201, 1035-1042. 10. Kuppusamy, P., Chzhan, M., Vij, K., Shteynbuk, M., Lefer, D.J., Giannella, E., and Zweier, J.L. (1994) Proc. Natl. Acad. Sci. USA 91, 3388-3392. 11. Gomi, F., Utsumi, H., Hamada, A., and Matsuo, M. (1993) Life Sci. 52, 2027-2033. 12. Miura, Y., Utsumi, H., and Hamada, A. (1992) Biochem. Biophys. Res. Commun. 182, 1108-1114.
646
Vol. 42, No. 3, 1997
BIOCHEMISTRY and MOLECULAR BIOLOGY INTERNATIONAL
13. Utsumi, H., Takeshita, K., Miura, Y., Masuda, S., and Hamada, A. (1993) Free Rad. Res.Commun. 19, $219-$225. 14. Utsumi, H., Ichikawa, K., and Takeshita, K. (1995) Toxicol. Lett. 82/83, 561-565. 15. Miura,Y., Hamada, A., and Utsumi, H. (1995) Free Rad. Res. 22, 209-214. 16. Irie, T., Fukushi, K., Akimoto, Y., Tamagami, H., and Nozaki, T. (1994) Nucl. Med. Biol. 21, 801-808. 17. Keana, J.F.W., Keana, S.B., and Beetham, D. (1967) J. Amer. Chem. Soc. 89, 30553056. 18. Takeshita, K., Utsumi, H., and Hamada, A. (1991) Biochem. Biophys. Res. Commun. 177, 874-880. 19. Liebmann, J., Bourg, J., Krishna, C.M., Glass, J., Cook, J.A., and Mitchell, J.B. (1994) Life Sci. 5 4, PL503-509. 20. Shimshick, E.J., and McConnell, H.M. (1973) Biochemistry 12, 2351-2360. 21. Tsien, R.Y. (1981) Nature 290, 527-528. 22. Utsumi, H., and Takeshita, K. (1995) in Bioradicals Detected by ESR Spectroscopy (OhyaNishiguchi, H., and Packer, L., Eds.), pp. 321-334, Birkhauser Verlag Basel, Switzerland.
647
