- Email: [email protected]
Determination of cytochromes in human liver contaminated with hemoglobin
Clinica Chimica Acta, 172 (1988) 93-100 Elsevier
93
CCA 04072
Determination of cytochromes in human liver contaminated with hemoglobin Akira
Tanaka
a, Atsuhiko Maki a, Yasuo Karniyama a, Yoshio Kazue Ozawa a and Yutaka Orii b
Yamaoka
a,
a Second Department of Surgery and b Department of Public Health, Faculty of Medicine, Kyoto University, Kyoto (Japan) (Received
14 July 1987; revision received 19 October
Key words: Human
liver mitochondria;
1987; accepted
Cytochrome;
after revision
Hemoglobin;
26 October
1987)
Liver cirrhosis
A spectrophotometric method was developed to determine cytochromes in human liver mitochondria contaminated with hemoglobin. The influence of hemoglobin on the measurement was canceled by keeping hemoglobin in the carbon monoxide bound form throughout determination. Mitochondria were solubilized by 2% sodium cholate, and cytochromes were reduced enzymatically with glutamate and succinate as substrates to the maximal extent. The amount of cytochromes determined spectrophotometrically was linearly correlated with mitochondrial protein at least up to the extent of 8 mg/ml.
Introduction It has been reported from our laboratory that the concentration of human liver cytochrome au3 increases in accordance with the severity of liver cirrhosis [l], and that the concentration of cytochrome au3 serves as a useful index in deciding the policy of surgical treatment of hepatocellular carcinoma with liver cirrhosis [2]. When human liver mitochondria are isolated from the surgical specimen, however, contamination of hemoglobin occurs which interferes with spectrophotometric determination of mitochondrial cytochromes. In this study, we have developed a method of completely canceling the disturbance caused by hemoglobin and correctly determining cytochromes.
Correspondence and requests for reprints to: Dr. A. Tanaka, Second Department of Surgery, Medicine, Kyoto University, 54, Kawaracho, Shogoin, Sakyoku, Kyoto, 606 Japan.
0009-8981/88/$03.50
0 1988 Elsevier Science Publishers
B.V. (Biomedical
Division)
Faculty
of
Methods Male rats of Wistar strain weighing 200-250 g were used. Immediately after decapitation, the rat liver was dissected, immersed in chilled isolation medium containing 0.3 mol/l mannitol and 0.1 mmol/l EDTA, cut into approximately 0.2-mm slices, and washed five times with the same medium to remove hemoglobin. Liver mitochondria were prepared by the methods previously reported by Ozawa et al [3]. The final mitochondrial pellet was suspended in the same isolation medium at a concentration of 60-80 mg protein/ml. Human liver materials weighing 1.0 g were obtained under informed consent from 10 hepatectomized patients for hepatocellular carcinoma, cholangiocarcinoma, metastatic carcinoma, or intrahepatic gall stone. The liver materials free from pathological conditions were selected from hepatectomized surgical specimens. Human liver mitochondria were prepared by the same methods applied to rats. Although hemoglobin was removed partially from mitochondria by washing with an isolation medium, the human mitochondria, unlike rat liver mitochondria, still contained a considerable amount of hemoglobin. Liver mitochondrial cytochromes were determined spectrophotometrically with a Shimadzu dual wavelength spectrophotometer (UV-3000). For determination of mitochondrial cytochromes, a mixture containing 0.3 mol/l mannitol, 0.01 mol/l KCl, 0.01 mol/l Tris-HCl (pH 7.4), 0.2 mmol/l EDTA was used; this will be referred to as the ‘mannitol medium’. Extinction coefficient differences were employed to determine cytochrome 4a3, cytochrome b, cytochrome c + ci as 33.0 mmol/l per cm at 605-630 nm, 17.9 mmol/l per cm at 562-575 nm, and 19.0 mmol/l per cm at 550-540 nm, respectively. Reduced hemoglobin was prepared according to the following procedures. A solution of bovine hemoglobin (Sigma type I) was dissolved in 0.1 mol/l phosphate buffer (pH 7.4), and was reduced by addition of sodium dithionite. The solution was then passed through a Sephadex G-25 (fine) column to remove dithionite. Reduced hemoglobin was treated with carbon monoxide and determined spectrophotometrically. The millimolar absorptivity of HbCO employed was 13.4 cm at 569 nm [4]. Mitochondrial protein was determined by the method of Lowry et al with bovine serum albumin as standard
[51. Results and discussion The influence of turbidity of mitochondria on measurement To investigate effects of turbidity on spectrophotometrical measurement of mitochondrial cytochromes, the measurements were compared between intact mitochondria and cholate-solubilized ones. In the former case, a given concentration of non-treated mitochondria was put in the ‘mannitol medium’ plus 5 mmol/l phosphate buffer (pH 7.4) and 1 mmol/l ADP; in the latter case, the mitochondrial suspension was solubilized by 2% sodium cholate to lessen error due to light scattering, and put in the ‘manmtol medium’. In both cases, after aeration, 6 ml of reaction mixture was divided into a sample and reference cuvette, and a base line
95
0.06
0 Mitochondrial
Protein
Concentration
2
4
6
(mg/ml)
Fig. 1. Influence of turbidity of mitochondrial suspension on cytochrome measurement. 0 and 0 show enzymatically induced and chemically reduced spectral differences of intact mitochondria, respectively; A and A show enzymatically induced and chemically reduced spectral differences of solubilized ones, respectively.
was recorded. Glutamate and succinate were added in concentrations of 5 mmol/l into the sample cuvette. As soon as the anaerobic state was attained, the difference spectrum was recorded. After a difference spectrum between aerobic and anaerobic state was recorded, 0.3 mg of sodium dithionite was added into the sample cuvette and the difference spectrum was again recorded. Mitochondrial concentrations were changed from 2 to 8 mg protein/ml. Absorbance differences in each case were plotted against mitochondrial protein as shown in Fig. 1. Figure 1 clearly demonstrates that without solubilization of mitochondria the values of spectral differences of cytochromes deviated from the true values due to light scattering effect, and that with solubilization the values of cytochromes were linearly correlated with mitochondrial protein at least up to the extent of 8 mg/rnl. With solubilized mitochondria there were little differences between dithionite-reduced state and enzymatically induced anaerobic state. Double beam spectrophotometry has been applied to determine cytochromes in turbid sample. Previous methods by Ozawa et al used intact mitochondria to determine flavoprotein and pyridine nucleotide in addition to cytochromes [3]. However, in order to determine the concentrations of cytochromes correctly, mitochondria should be solubilized to minimize the errors due to light scattering.
96
The influence of contaminated hemoglobin on measurement To assess the influence of contaminated hemoglobin on the spectrophotometrical measurement of mitochondrial cytochromes, cholate-solubilized mitochondria in of reduced hemoterms of 9 ~.mol/l cytochrome aa,, and a given concentration globin were put in the mannitol medium, and the difference spectra were recorded. After anaerobic state in the sample cuvette was attained by addition of glutamate and succinate, the difference spectrum was recorded. To convert hemoglobin in both sample and reference cuvettes into HbCO, the solutions in both cuvettes were bubbled with 100% CO for 1 min, and the reference cuvette was then aerated for a few seconds. By this procedure, hemoglobin in the reference cuvette remained in its CO bound form and cytochrome aa existed as a3+, ai+, while in the sample cuvette hemoglobin existed as HbCO and cytochrome aa as a2+, ai+ . CO, since hemoglobin has 200 times higher affinity to CO than cytochrome aa [6]. The spectral difference of hemoglobin was thus canceled, the difference spectrum of cytochrome aa had a shoulder at around 590 nm and the peak of cytochrome aa at 605 nm became lower by 0.80 times. Therefore, the millimolar extinction coefficient difference of cytochrome aa at 605-630 nm (reduced and CO treated)
0
1
I
I
I
I
I
I
2
4
6
8
10
15
[Hb)
/
1
(CC01
Fig. 2. Influence of contaminated hemoglobin on cytochrome measurement. 0 and 0 show spectral differences at 550-W nm (cytochrome c + q) in the absence and presence of CO treatment, respectively. A and A show those at 605-630 nm (cytochrome aos) in the absence and presence of CO treatment, respectively. CCO; cytochrome c oxidase.
91
employed was 26.4/cm. Molar ratios of hemoglobin to cytochrome aa3 were changed from 0 to 15. The spectral differences at wavelength pairs of 605-630 nm and 550-540 nm were plotted against the ratio as shown in Fig. 2. Figure 2 clearly shows that the influence of contaminated hemoglobin could be completely eliminated by the CO treatment at least up to 15 molar ratio of Hb to cytochrome 4a3. These results indicate that by this method cytochrome au3 and cytochrome c + ci could be determined directly and independently of contaminated hemoglobin. In the case of purified cytochromes or mitochondria which are obtained from experimental animals, the concentration of hemoglobin is negligible. However, human liver mitochondria prepared from surgical specimen contain a considerable amount of hemoglobin. Carbon monoxide binds to hemoglobin and cytochrome CILZ~ but not cytochromes b, c, and ci. Ozawa et al have reported a method which uses 10% CO and 90% 0, in an amytal treated mitochondrial suspension [3]. We have further developed this method to determine human liver mitochondrial cytochromes with hemoglobin contamination. Determination of human liver mitochondrial cytochromes Cytochromes of human liver mitochondria obtained from surgical specimen were determined spectrophotometrically. Mitochondria were solubilized with 2% cholate
I
I
I
I
500
550
600
650
wavelength
(nm)
Fig. 3. Measurement of human liver cytochromes. the difference spectra of human liver mitochondria antimycin A treatment, respectively.
A typical case is shown here. Curve a, b, and c show without CO treatment, after CO treatment and after
98 TABLE
I
Concentrations Patient no.
of cytochrome
caa,
b, and c + cr in human
liver mitochondria Cyt b
Sex
Age
Type of disease a
Type of surgery
1 2 3 4 5
F M M M M
62 63 58 56 57
MC HCC HCC HCC HCC
LC(+) LC(+) LC(+) LC(+) LC(+)
Lt Rt Rt Rt Rt
Lobectomy Lobectomy Lobectomy Lobectomy Lobectomy
0.072 0.069 0.064 0.052 0.043
0.068 0.058 0.071 0.065 0.059
0.114 0.096 0.103 0.103 0.078
6
M
59
HCC
LC( - )
0.057
0.078
0.123
7 8
F M
74 76
HCC CC
LC(-) LC(-)
0.055 0.052
0.083 0.062
0.096 0.117
9 10
F M
58 56
HCC IHG
LC(-) LC(-)
Lateral Segmentcctomy Rt Lobectomy Extended Rt Lobectomy Rt Lobectomy Rt Lobectomy
0.046 0.042
0.064 0.049
0.090 0.077
IHG,
intrahepatic
a HCC, Hepatocellular cholangiocarcinoma;
carcinoma; MC, LC, liver cirrhosis.
metastatic
Cyt aas (nmol/mg
carcinoma;
Cyt c + c,
protein)
gall
stone;
CC,
and put in the mannitol medium. After the base line was recorded, the difference spectrum between aerobic state and anaerobic state induced by 5 mmol/l glutamate and succinate was recorded as shown in curve a of Fig. 3. After CO bubbling of both cuvettes and aeration of the reference cuvette, the difference spectrum was recorded (curve b). After the sample cuvette was treated with 0.4 pg/ml antimycin A and aerated, the difference spectrum was recorded to determine the spectrum of cytochrome b alone (curve c). Figure 3 shows that without CO treatment a considerable amount of hemoglobin elevated the peaks in difference spectra of cytochrome c + ci and cytochrome aa and deepened the trough at around 575 nm. By adopting this CO treatment method cytochrome aas, b, and c + ci of human liver mitochondria from surgical specimens which contained hemoglobin could be correctly determined. Table I shows the amount of cytochromes in human liver mitochondria obtained from 10 hepatectomized patients. Among liver cirrhotics there were several cases in which the concentration of cytochrome aus was high, a finding supports our previous studies [1,2,7].
References 1 Ozawa K. Adaptive response of the liver mitochondria and its relation to hepatic functional reserve: evaluation by cytocbrome a( + a,) assay and glucose intolerance. Asian Med J 1980;23:499-528. 2 Uchida K, Jikko A, Yamato T, Kamiyama Y, Ozawa K. Relationship of glucose intolerance and indocyanine green clearance to respiratory enzyme levels in human cirrhotic liver. Am J Med Sci 1985;290:19-27. 3 Ozawa K, Kitamura 0, Mizukami T, et al. Human liver mitochondria. Clin Chim Acta 1972;38:385-393.
99 4 Antonini E, Brunori M. Hemoglobin and myoglobin in their reactions with ligands. Amsterdam: North-Holland Publ, 1971;18-19. 5 Lowry OH, Rosebrough NJ, Farr AJ. Protein measurement with the folin phenol reagent. J Biol Chem 1951;265-275. 6 Douglas CG, Haldane JS. The laws of combination of haemoglobin with carbon monoxide and oxygen. J Physiol 1912;44:275-304. 7 Sato M, Ida T, Ozawa K, et al. Adaptive increase of respiratory enzymes in the mitochondria from cirrhotic livers of patients and rats, and its relationship to glucose tolerance. Am J Med Sci 1977;273;4-20.
93
CCA 04072
Determination of cytochromes in human liver contaminated with hemoglobin Akira
Tanaka
a, Atsuhiko Maki a, Yasuo Karniyama a, Yoshio Kazue Ozawa a and Yutaka Orii b
Yamaoka
a,
a Second Department of Surgery and b Department of Public Health, Faculty of Medicine, Kyoto University, Kyoto (Japan) (Received
14 July 1987; revision received 19 October
Key words: Human
liver mitochondria;
1987; accepted
Cytochrome;
after revision
Hemoglobin;
26 October
1987)
Liver cirrhosis
A spectrophotometric method was developed to determine cytochromes in human liver mitochondria contaminated with hemoglobin. The influence of hemoglobin on the measurement was canceled by keeping hemoglobin in the carbon monoxide bound form throughout determination. Mitochondria were solubilized by 2% sodium cholate, and cytochromes were reduced enzymatically with glutamate and succinate as substrates to the maximal extent. The amount of cytochromes determined spectrophotometrically was linearly correlated with mitochondrial protein at least up to the extent of 8 mg/ml.
Introduction It has been reported from our laboratory that the concentration of human liver cytochrome au3 increases in accordance with the severity of liver cirrhosis [l], and that the concentration of cytochrome au3 serves as a useful index in deciding the policy of surgical treatment of hepatocellular carcinoma with liver cirrhosis [2]. When human liver mitochondria are isolated from the surgical specimen, however, contamination of hemoglobin occurs which interferes with spectrophotometric determination of mitochondrial cytochromes. In this study, we have developed a method of completely canceling the disturbance caused by hemoglobin and correctly determining cytochromes.
Correspondence and requests for reprints to: Dr. A. Tanaka, Second Department of Surgery, Medicine, Kyoto University, 54, Kawaracho, Shogoin, Sakyoku, Kyoto, 606 Japan.
0009-8981/88/$03.50
0 1988 Elsevier Science Publishers
B.V. (Biomedical
Division)
Faculty
of
Methods Male rats of Wistar strain weighing 200-250 g were used. Immediately after decapitation, the rat liver was dissected, immersed in chilled isolation medium containing 0.3 mol/l mannitol and 0.1 mmol/l EDTA, cut into approximately 0.2-mm slices, and washed five times with the same medium to remove hemoglobin. Liver mitochondria were prepared by the methods previously reported by Ozawa et al [3]. The final mitochondrial pellet was suspended in the same isolation medium at a concentration of 60-80 mg protein/ml. Human liver materials weighing 1.0 g were obtained under informed consent from 10 hepatectomized patients for hepatocellular carcinoma, cholangiocarcinoma, metastatic carcinoma, or intrahepatic gall stone. The liver materials free from pathological conditions were selected from hepatectomized surgical specimens. Human liver mitochondria were prepared by the same methods applied to rats. Although hemoglobin was removed partially from mitochondria by washing with an isolation medium, the human mitochondria, unlike rat liver mitochondria, still contained a considerable amount of hemoglobin. Liver mitochondrial cytochromes were determined spectrophotometrically with a Shimadzu dual wavelength spectrophotometer (UV-3000). For determination of mitochondrial cytochromes, a mixture containing 0.3 mol/l mannitol, 0.01 mol/l KCl, 0.01 mol/l Tris-HCl (pH 7.4), 0.2 mmol/l EDTA was used; this will be referred to as the ‘mannitol medium’. Extinction coefficient differences were employed to determine cytochrome 4a3, cytochrome b, cytochrome c + ci as 33.0 mmol/l per cm at 605-630 nm, 17.9 mmol/l per cm at 562-575 nm, and 19.0 mmol/l per cm at 550-540 nm, respectively. Reduced hemoglobin was prepared according to the following procedures. A solution of bovine hemoglobin (Sigma type I) was dissolved in 0.1 mol/l phosphate buffer (pH 7.4), and was reduced by addition of sodium dithionite. The solution was then passed through a Sephadex G-25 (fine) column to remove dithionite. Reduced hemoglobin was treated with carbon monoxide and determined spectrophotometrically. The millimolar absorptivity of HbCO employed was 13.4 cm at 569 nm [4]. Mitochondrial protein was determined by the method of Lowry et al with bovine serum albumin as standard
[51. Results and discussion The influence of turbidity of mitochondria on measurement To investigate effects of turbidity on spectrophotometrical measurement of mitochondrial cytochromes, the measurements were compared between intact mitochondria and cholate-solubilized ones. In the former case, a given concentration of non-treated mitochondria was put in the ‘mannitol medium’ plus 5 mmol/l phosphate buffer (pH 7.4) and 1 mmol/l ADP; in the latter case, the mitochondrial suspension was solubilized by 2% sodium cholate to lessen error due to light scattering, and put in the ‘manmtol medium’. In both cases, after aeration, 6 ml of reaction mixture was divided into a sample and reference cuvette, and a base line
95
0.06
0 Mitochondrial
Protein
Concentration
2
4
6
(mg/ml)
Fig. 1. Influence of turbidity of mitochondrial suspension on cytochrome measurement. 0 and 0 show enzymatically induced and chemically reduced spectral differences of intact mitochondria, respectively; A and A show enzymatically induced and chemically reduced spectral differences of solubilized ones, respectively.
was recorded. Glutamate and succinate were added in concentrations of 5 mmol/l into the sample cuvette. As soon as the anaerobic state was attained, the difference spectrum was recorded. After a difference spectrum between aerobic and anaerobic state was recorded, 0.3 mg of sodium dithionite was added into the sample cuvette and the difference spectrum was again recorded. Mitochondrial concentrations were changed from 2 to 8 mg protein/ml. Absorbance differences in each case were plotted against mitochondrial protein as shown in Fig. 1. Figure 1 clearly demonstrates that without solubilization of mitochondria the values of spectral differences of cytochromes deviated from the true values due to light scattering effect, and that with solubilization the values of cytochromes were linearly correlated with mitochondrial protein at least up to the extent of 8 mg/rnl. With solubilized mitochondria there were little differences between dithionite-reduced state and enzymatically induced anaerobic state. Double beam spectrophotometry has been applied to determine cytochromes in turbid sample. Previous methods by Ozawa et al used intact mitochondria to determine flavoprotein and pyridine nucleotide in addition to cytochromes [3]. However, in order to determine the concentrations of cytochromes correctly, mitochondria should be solubilized to minimize the errors due to light scattering.
96
The influence of contaminated hemoglobin on measurement To assess the influence of contaminated hemoglobin on the spectrophotometrical measurement of mitochondrial cytochromes, cholate-solubilized mitochondria in of reduced hemoterms of 9 ~.mol/l cytochrome aa,, and a given concentration globin were put in the mannitol medium, and the difference spectra were recorded. After anaerobic state in the sample cuvette was attained by addition of glutamate and succinate, the difference spectrum was recorded. To convert hemoglobin in both sample and reference cuvettes into HbCO, the solutions in both cuvettes were bubbled with 100% CO for 1 min, and the reference cuvette was then aerated for a few seconds. By this procedure, hemoglobin in the reference cuvette remained in its CO bound form and cytochrome aa existed as a3+, ai+, while in the sample cuvette hemoglobin existed as HbCO and cytochrome aa as a2+, ai+ . CO, since hemoglobin has 200 times higher affinity to CO than cytochrome aa [6]. The spectral difference of hemoglobin was thus canceled, the difference spectrum of cytochrome aa had a shoulder at around 590 nm and the peak of cytochrome aa at 605 nm became lower by 0.80 times. Therefore, the millimolar extinction coefficient difference of cytochrome aa at 605-630 nm (reduced and CO treated)
0
1
I
I
I
I
I
I
2
4
6
8
10
15
[Hb)
/
1
(CC01
Fig. 2. Influence of contaminated hemoglobin on cytochrome measurement. 0 and 0 show spectral differences at 550-W nm (cytochrome c + q) in the absence and presence of CO treatment, respectively. A and A show those at 605-630 nm (cytochrome aos) in the absence and presence of CO treatment, respectively. CCO; cytochrome c oxidase.
91
employed was 26.4/cm. Molar ratios of hemoglobin to cytochrome aa3 were changed from 0 to 15. The spectral differences at wavelength pairs of 605-630 nm and 550-540 nm were plotted against the ratio as shown in Fig. 2. Figure 2 clearly shows that the influence of contaminated hemoglobin could be completely eliminated by the CO treatment at least up to 15 molar ratio of Hb to cytochrome 4a3. These results indicate that by this method cytochrome au3 and cytochrome c + ci could be determined directly and independently of contaminated hemoglobin. In the case of purified cytochromes or mitochondria which are obtained from experimental animals, the concentration of hemoglobin is negligible. However, human liver mitochondria prepared from surgical specimen contain a considerable amount of hemoglobin. Carbon monoxide binds to hemoglobin and cytochrome CILZ~ but not cytochromes b, c, and ci. Ozawa et al have reported a method which uses 10% CO and 90% 0, in an amytal treated mitochondrial suspension [3]. We have further developed this method to determine human liver mitochondrial cytochromes with hemoglobin contamination. Determination of human liver mitochondrial cytochromes Cytochromes of human liver mitochondria obtained from surgical specimen were determined spectrophotometrically. Mitochondria were solubilized with 2% cholate
I
I
I
I
500
550
600
650
wavelength
(nm)
Fig. 3. Measurement of human liver cytochromes. the difference spectra of human liver mitochondria antimycin A treatment, respectively.
A typical case is shown here. Curve a, b, and c show without CO treatment, after CO treatment and after
98 TABLE
I
Concentrations Patient no.
of cytochrome
caa,
b, and c + cr in human
liver mitochondria Cyt b
Sex
Age
Type of disease a
Type of surgery
1 2 3 4 5
F M M M M
62 63 58 56 57
MC HCC HCC HCC HCC
LC(+) LC(+) LC(+) LC(+) LC(+)
Lt Rt Rt Rt Rt
Lobectomy Lobectomy Lobectomy Lobectomy Lobectomy
0.072 0.069 0.064 0.052 0.043
0.068 0.058 0.071 0.065 0.059
0.114 0.096 0.103 0.103 0.078
6
M
59
HCC
LC( - )
0.057
0.078
0.123
7 8
F M
74 76
HCC CC
LC(-) LC(-)
0.055 0.052
0.083 0.062
0.096 0.117
9 10
F M
58 56
HCC IHG
LC(-) LC(-)
Lateral Segmentcctomy Rt Lobectomy Extended Rt Lobectomy Rt Lobectomy Rt Lobectomy
0.046 0.042
0.064 0.049
0.090 0.077
IHG,
intrahepatic
a HCC, Hepatocellular cholangiocarcinoma;
carcinoma; MC, LC, liver cirrhosis.
metastatic
Cyt aas (nmol/mg
carcinoma;
Cyt c + c,
protein)
gall
stone;
CC,
and put in the mannitol medium. After the base line was recorded, the difference spectrum between aerobic state and anaerobic state induced by 5 mmol/l glutamate and succinate was recorded as shown in curve a of Fig. 3. After CO bubbling of both cuvettes and aeration of the reference cuvette, the difference spectrum was recorded (curve b). After the sample cuvette was treated with 0.4 pg/ml antimycin A and aerated, the difference spectrum was recorded to determine the spectrum of cytochrome b alone (curve c). Figure 3 shows that without CO treatment a considerable amount of hemoglobin elevated the peaks in difference spectra of cytochrome c + ci and cytochrome aa and deepened the trough at around 575 nm. By adopting this CO treatment method cytochrome aas, b, and c + ci of human liver mitochondria from surgical specimens which contained hemoglobin could be correctly determined. Table I shows the amount of cytochromes in human liver mitochondria obtained from 10 hepatectomized patients. Among liver cirrhotics there were several cases in which the concentration of cytochrome aus was high, a finding supports our previous studies [1,2,7].
References 1 Ozawa K. Adaptive response of the liver mitochondria and its relation to hepatic functional reserve: evaluation by cytocbrome a( + a,) assay and glucose intolerance. Asian Med J 1980;23:499-528. 2 Uchida K, Jikko A, Yamato T, Kamiyama Y, Ozawa K. Relationship of glucose intolerance and indocyanine green clearance to respiratory enzyme levels in human cirrhotic liver. Am J Med Sci 1985;290:19-27. 3 Ozawa K, Kitamura 0, Mizukami T, et al. Human liver mitochondria. Clin Chim Acta 1972;38:385-393.
99 4 Antonini E, Brunori M. Hemoglobin and myoglobin in their reactions with ligands. Amsterdam: North-Holland Publ, 1971;18-19. 5 Lowry OH, Rosebrough NJ, Farr AJ. Protein measurement with the folin phenol reagent. J Biol Chem 1951;265-275. 6 Douglas CG, Haldane JS. The laws of combination of haemoglobin with carbon monoxide and oxygen. J Physiol 1912;44:275-304. 7 Sato M, Ida T, Ozawa K, et al. Adaptive increase of respiratory enzymes in the mitochondria from cirrhotic livers of patients and rats, and its relationship to glucose tolerance. Am J Med Sci 1977;273;4-20.