- Email: [email protected]
Gln27Glu β2-adrenergic receptor variant is associated with hypertriglyceridemia and the development of fatty liver
Clinica Chimica Acta 314 Ž2001. 85–91 www.elsevier.comrlocaterclinchim
Gln27Glu b2-adrenergic receptor variant is associated with hypertriglyceridemia and the development of fatty liver Noriyuki Iwamoto ) , Yoko Ogawa, Susumu Kajihara, Akitaka Hisatomi, Tsutomu Yasutake, Toru Yoshimura, Toshihiko Mizuta, Toshiya Hara, Iwata Ozaki, Kyosuke Yamamoto DiÕision of Metabolism and Endocrinology, Department of Internal Medicine, Saga Medical School, 5-1-1 Nabeshima, Saga 849, Japan Received 4 April 2001; received in revised form 18 July 2001; accepted 23 July 2001
Abstract Background: Nonalcoholic steatohepatitis ŽNASH. is associated with the metabolism of lipid, glucose and energy. Beta-adrenergic receptors play an important role in the regulation of energy expenditure, in part, by stimulating lipid mobilization through lipolysis. Methods: To assess whether it is common for the b2-adrenergic receptor ŽB2AR. gene polymorphisms in codons 16 and 27 to play a role in the development of fatty liver, we investigated 251 unrelated healthy Japanese males who were drug-free and showed no signs of heavy drinking. Results: The allelic frequency of B2AR gene mutation in codons 16 and 27 did not differ between obese subjects ŽBMI ) 25.0 kgrm2 , n s 151. and non-obese subjects ŽBMI F 25.0 kgrm2 , n s 100.. The Gly16 homozygotes had a lower high-density lipoprotein cholesterol ŽHDL-C. level than the Arg16 homozygotes Ž1.50 " 0.4 vs. 1.32 " 0.3 mmolrl, p s 0.014.. However, no significant association with fatty liver was observed in the Gly16 allele frequency. The Gln27Glu27 heterozygotes showed higher concentrations of serum triglycerides ŽTG. than the Gln27Gln27 homozygotes Ž1.62 " 0.93 vs. 2.21 " 1.67 mmolrl, p s 0.013.. This correlation was also observed in all subjects regardless of weight classification. Univariate analysis indicated that subjects with the heterozygous Gln27Glu mutant alleles had a significantly higher prevalence of fatty liver vs. those without the mutation ŽGlu27 allele frequency, 0.07 vs. 0.12, p s 0.047; odds ratio, 1.92; 95% confidence interval, 1.01–3.68.. However, multivariate logistic regression models showed the prevalence of fatty liver to be significantly related to the homeostasis model assessment ŽHOMA. index, BMI, triglyceride and HDL-cholesterol. Conclusions: These results suggest that the amino-terminal polymorphisms of the b2-adrenergic receptor gene in codon 27 were associated with hypertryglyceridemia and independent of obesity, and thereby could be involved in the molecular pathogenesis of fatty liver. q 2001 Elsevier Science B.V. All rights reserved. Keywords: Beta2-adrenergic receptor gene; Polymorphism; Nonalcoholic steatohepatitis Žfatty liver.; Triglyceride; Obesity
1. Introduction
)
Corresponding author. Tel.: q81-952-31-6511; fax: q81-95234-2017. E-mail address: [email protected] ŽN. Iwamoto..
Fatty liver is one of the most common liver diseases found in clinical medicine. Generally, fatty liver change consists of some disruption of the normal physiology of triglyceride synthesis and secre-
0009-8981r01r$ - see front matter q 2001 Elsevier Science B.V. All rights reserved. PII: S 0 0 0 9 - 8 9 8 1 Ž 0 1 . 0 0 6 3 3 - 7
86
N. Iwamoto et al.r Clinica Chimica Acta 314 (2001) 85–91
tion, but the precise pathogenetic mechanisms are incompletely defined. Depending on the exact cause, multiple factors are likely responsible. It has been long recognized that fatty liver Žhepatic steatosis. occurs frequently in heavy alcohol drinkers. Although alcoholic liver disease is the most common cause, nonalcoholic hepatitis ŽNASH. is increasingly recognized as a significant form of liver disease. It is for this reason that NASH has been an important precursor to the development of fibrosis in diverse conditions such as hepatitis C, and alcoholic and nonalcoholic liver diseases w1,2x. It has been reported that more than 50% of obesity patients w3x and individuals with diabetes mellitus andror hyperlipidemia have developed fatty liver w4x. Catecholamines via b2–3 adrenergic receptors regulate body fat accumulation and energy expenditure. This regulation is in part affected by the stimulation of lipid mobilization through lipolysis in fat cells, which has been associated with obesity w5,6x. In the b3-adrenergic receptor ŽB3AR. gene, a missense mutation replacing Trp with Arg at codon 64 was reported to be associated with obesity-related metabolism disorders, such as insulin resistance and fatty liver w7–11x. Several polymorphisms have been found in the human B2AR gene w12x. These occur in the amino terminus of the receptor at amino acids 16 ŽArg or Gly. and 27 ŽGln or Glu. and in the fourth transmembrane-spanning domain at amino acid 164 ŽThr or Ile.. The amino acid substitutions at position 27 were shown to be associated with obesity in Swedish men and women w13,14x. In Japanese subjects, receptor variants at codon 27 were associated with insulin resistance w15,16x. However, it is unknown whether these genetic variants are associated with fatty liver. In the present study, we studied the common missense mutation of the B2AR gene in 251 Japanese male subjects to assess the potential association of gene mutation with nonalcoholic fatty liver.
S.D.. and 119 similarly unrelated subjects with nonfatty subjects aged 36–50 years Ž43.0 " 4.8 years. from 1111 patients who attended the Saga Health Care Center for health screening. The criteria for exclusion were as follows: heavy drinkers Ževeryday alcohol intake., treatment with any drug for diseases, known chronic liver disease and cancer, and females. We selected men only because fatty liver is found more often in males, and could be gender-related w17x. The diagnosis of fatty liver was based on the finding of liver–kidney contrast of abdominal ultrasonography. Liver–kidney contrast was held to be present in the case of significantly increased echogenicity of the liver compared with the right kidney by a gray scale real-time scanner w17–20x. Ultrasonography was performed in all cases by a single observer in a double-blind fashion. Informed consent was obtained from the subjects. Diagnosis of obesity ŽBMI ) 25 kgrm2 . was based on BMI according to the 1999 Japan Society for the Study of Obesity guideline.
2. Subjects and methods
2.3. Restriction fragment length polymorphism– polymerase chain reaction (RFLP–PCR)
2.2. Methods After overnight fasting, blood samples were taken from an antecubital vein, centrifuged Ž5000 rpm, 10 min. and stored at y20 8C prior to analysis. Serum concentrations of aspartate aminotransferase ŽAST., alanine aminotransferase ŽALT., P-glutamylglutamyltransferase ŽGGT., total cholesterol ŽT-cho., triglycerides ŽTG., high-density lipoprotein cholesterol ŽHDL-C. and fasting plasma glucose ŽFPG. were measured by the Saga Health Care Center’s routine chemistry laboratory ŽHitachi 7600.. Serumfree fatty acid ŽFFA. was measured by enzymatic assay, and plasma insulin level was analyzed by radioimmunoassay ŽSRL, Saga, Japan.. Blood pressure was measured in the right arm with the patient at rest and seated.
2.1. Subjects We randomly selected 132 subjects with fatty liver, aged 35–50 years Ž42.2 " 4.6 years, means"
Genomic DNA was extracted from peripheral leukocytes by digestion with proteinase K followed by phenolrchloroform extraction. The B2AR gene
N. Iwamoto et al.r Clinica Chimica Acta 314 (2001) 85–91
sequences were done by PCR in a volume of 10 l containing 0.1 U LATaq DNA polymerase ŽTakara, Japan., 1.5 mmolrl MgCl 2 , 100 molrl dNTPs, 10 mmolrl Tris–HCl pH 8.3 and 50 mmolrl of KCl. We used the following primers: the forward primers were 5X-CTTCTTGCTGGCACGCAAT-3X and 5XCCAGTGAAG-TGATGAAGTAGTTGG-3X for codon 16; and 5X-GGCCCATGACCAGATCAGCA3X and 5X-GAATGAGGCTTCCAGGCGTC-3X for codon 27. DNA was amplified by an initial denaturation step at 94 8C for 5 min, followed by 35 cycles of denaturation Ž98 8C, 20 s., annealing Ž55 8C for codon 16 or 65 8C for codon 27, 20 s. and extension Ž72 8C, 1 min., and a final extension at 72 8C for 10 min. Aliquots Ž3 l. of PCR products were digested with the addition of BsrD1 Žcodon 16, 56 8C, 1 U. or Ita1 Žcodon 27, 37 8C, 1 U.. After overnight incubation, the digested samples were separated by electrophoresis on a 4% ultra-pure DNA agarose gel using Tris–borate EDTA Ž8.9 mmolrl Tris–borate, 0.2 mmolrl EDTA. buffer and visualized under ultraviolet illumination after staining with ethidium bromide w14x.
2.4. Statistical analysis The results are presented as means" S.D. Homeostasis model assessment ŽHOMA. index indicating insulin resistance was calculated following the methods of Matthews et al. w21x. All analyses were performed according to the intention-to-treat principle. We used the chi-square test and Fisher’s exact test for the categorical comparisons of the data. Differences in the means of continuous measurements were tested by the Student’s t test and checked by the Mann–Whitney U test. Significant predictors identified by univariate analysis were then included in a forward, stepwise multiple logistic regression model to identify the most important factors affecting the incidence of fatty liver. Relative risk was estimated by the odds ratios ŽOR. and their 95% confidence intervals Ž95% CI.. A p - 0.05 was considered statistically significant; all tests were twotailed. All statistical analyses were performed on a personal computer using the statistical package SPSS for Windows ŽVersion 7.5.1 SPSS, Japan..
87
3. Results 3.1. Allelic frequency of the B2AR Õariants according to BMI in Japanese men We detected 63 Ž25%. Arg16 homozygotes, 123 Ž49%. Arg16rGly16 heterozygotes, 65 Ž26%. Gly16rGly16 homozygotes and 202 Ž81%. Gln27 homozygotes, 49 Ž19%. Gln27Glu27 heterozygotes and no Glu27 homozygotes in a group of 251 studied subjects. The Glu27 allele was slightly higher in obese subjects, but the frequency between obese subjects ŽBMI ) 25 kgrm2 , n s 151. and non-obese subjects ŽBMI F 25 kgrm2 , n s 100. did not statistically differ in the Gly16 allele Ž0.51 vs. 0.50. and in the Glu27 allele Ž0.11 vs. 0.08. ŽTable 1.. 3.2. Phenotypic characteristics of different polymorphisms in codons 16 and 27 The phenotypic characteristics of polymorphisms in codon 16 are shown in Table 2. The Gly16 homozygotes had a lower HDL-cholesterol level than the Arg16 homozygotes Ž1.50 " 0.4 vs. 1.32 " 0.3 mmolrl, p - 0.05.. Serum triglyceride level showed higher values Ž1.67 " 1.0 vs. 1.91 " 1.5 mmolrl., but did not statistically differ. Table 3 shows the phenotypes of the codon 27 polymorphisms. Individuals with the Glu27 allele mutation showed a higher concentration of serum triglyceride than those without this mutation Ž2.21 " Table 1 Allelic frequency of the B2AR variant according to obesity Codon 16
ArgrArg Žwild.
ArgrGly Žhetero.
GlyrGly Žhomo.
Gly16 allele frequency
Non-obese Obese
26 Ž26%. 37 Ž25%.
48 Ž48%. 75 Ž49%.
26 Ž26%. 39 Ž26%.
0.50 0.51)
Codon 27
GlnrGln Žwild.
GlnrGlu Žhetero.
Glu27 allele frequency
Non-obese Obese
84 Ž84%. 118 Ž78%.
16 Ž16%. 33 Ž22%.
0.08 0.11) )
Non-obese Ž F 25.0 kgrm2 . and obese ŽBMI) 25.0 kgrm2 . according to the 1999 Japan Society for the Study of Obesity guideline. ) P s NS vs. non-obese subjects. )) ps NS vs. non-obese subjects.
N. Iwamoto et al.r Clinica Chimica Acta 314 (2001) 85–91
88
Table 2 Baseline characteristics of subjects with B2AR gene polymorphisms in codon 16
n Age Žyears. BMI Žkgrm2 . AST ŽIUrl. ALT ŽIUrl. GGT ŽIUrl. Systolic BP Žmm Hg. Diastolic BP Žmm Hg. FPG ŽmM. Insulin ŽUrml. HOMA index T-cho Žmmolrl. HDL-C Žmmolrl. TG Žmmolrl. FFA ŽmEqrml.
Arg16Arg
Arg16Gly
Gly16Gly
p value
ŽA.
ŽB.
ŽC.
A vs. B
A vs. C
63 Ž25.1%. 43.1 " 4.9 26.1 " 3.5 27.0 " 12.5 34.8 " 23.0 38.6 " 23.8 127.1 " 18.5 77.6 " 12.7 5.35 " 0.69 10.7 " 4.0 2.54 " 1.0 5.41 " 0.9 1.50 " 0.4 1.67 " 1.0 0.54 " 0.3
123 Ž49.0%. 42.3 " 4.6 26.1 " 3.6 26.8 " 11.8 36.2 " 26.4 38.1 " 37.3 126.0 " 16.0 74.8 " 11.9 5.48 " 1.46 10.2 " 4.1 2.52 " 1.0 5.26 " 0.9 1.45 " 0.4 1.68 " 1.0 0.49 " 0.2
65 Ž25.9%. 42.6 " 4.6 26.3 " 3.4 29.2 " 14.0 35.7 " 27.5 36.6 " 29.1 125.0 " 13.0 74.2 " 10.5 5.24 " 0.99 10.7 " 6.5 2.57 " 2.0 5.17 " 0.8 1.32 " 0.3 1.91 " 1.5 0.50 " 0.2
NS NS NS NS NS NS NS NS NS NS NS NS NS NS
NS NS NS NS NS NS NS NS NS NS NS - 0.05 NS NS
Data are means" S.D.; n, number of subjects; BP, blood pressure.
1.67 vs. 1.62 " 0.93 mmolrl, p - 0.05.. Moreover, this correlation was also observed in all subjects regardless of weight classification Ž1.42 " 0.92 vs. Table 3 Baseline characteristics of subjects with B2AR gene polymorphisms in codon 27
n Age Žyears. BMI Žkgrm2 . AST ŽIUrl. ALT ŽIUrl. GGT ŽIUrl. Systolic BP Žmm Hg. Diastolic BP Žmm Hg. FPG ŽmM. Insulin ŽUrml. HOMA index T-cho Žmmolrl. HDL-C Žmmolrl. TG Žmmolrl. overall Ž ns 251. obese Ž ns151. non-obese Ž ns100. FFA ŽmEqrml.
Gln27rGln27 Žwild.
Gln27rGlu27 Žmutant.
202 Ž80.5%. 42.5"4.7 26.1"3.4 27.6"12.9 36.0"27.3 38.0"30.8 126.1"16.2 75.8"12.0 5.32"0.99 10.4"5.0 2.51"1.5 5.28"0.86 1.45"0.43 1.62"0.93
49 Ž19.5%. 42.8"4.8 26.4"3.8 26.9"11.6 34.5"18.8 37.2"37.7 125.4"14.9 73.4"10.8 5.63"1.78 10.5"4.0 2.65"1.3 5.24"0.85 1.36"0.33 2.21"1.67
1.78"0.91
2.31"1.84
1.42"0.92
1.98"1.27
0.51"0.2
0.49"0.1
Data are means"S.D.; n, number of subjects; BP, blood pressure. No difference in any group.
1.98 " 1.27 mmolrl, p - 0.05, non-obese subjects. Ž1.78 " 0.91 vs. 2.31 " 1.84 mmolrl, p - 0.05, obese subjects. ŽTable 3.. However, both groups were comparable with respect to age, BMI, AST, ALT, GGT, blood pressure, insulin, FPG, HOMA index, total cholesterol, HDL-cholesterol and FFA ŽTable 3.. 3.3. UniÕariate analysis of fatty liÕer, B2AR genotype and clinical Õariables Subjects with fatty liver showed significantly higher concentrations of liver enzymes ŽTable 4.. The allelic frequency of the Glu27 mutation significantly differed between fatty liver subjects and nonfatty liver subjects Ž p - 0.05, OR 1.92, 95% CI:
Table 4 Liver-related enzymes of subjects with fatty liver and controls
n Age Žyears. AST ŽIUrl. ALT ŽIUrl. GGT ŽIUrl.
Fatty liver
Controls
132 42.2"4.9 Ž35–50. 29.5"12.7 43.8"26.7 44.5"35.5
119 43.0"4.8 Ž36–50. 25.2"12.2 26.8"21.7 30.526.5
Data are means"S.D.; n, number of subjects.
p value NS - 0.01 - 0.001 0.001
N. Iwamoto et al.r Clinica Chimica Acta 314 (2001) 85–91
89
Table 5 Association between polymorphisms in the B2AR gene and fatty liver
Table 7 Multivariate logistic regression analysis of fatty liver on clinical variables
Codon 16
Variables
OR Ž95% CI. a
HOMA index BMI Žkgrm2 . Triglyceride Žmmolrl. HDL cholesterol Žmmolrl.
2.03 Ž1.37–3.02. 1.22 Ž1.09–1.38. 1.78 Ž1.17–2.71.
0.710 0.202 0.576
0.30 Ž0.11–0.83.
y1.196
ArgrArg
Fatty liver 34 Ž26%. Non-fatty liver 29 Ž24%. Codon 27
GlnrGln
ArgrGly
GlyrGly
66 Ž50%. 32 Ž24%. 57 Ž48%. 33 Ž26%. GlnrGlu
0.49 ) 0.51
Glu27 allele frequency
Fatty liver 100 Ž76%. 32 Ž24%. 0.12 Non-fatty liver 102 Ž86%. 17 Ž14%. 0.07 )
Gly16 allele frequency
.
p value - 0.0005 - 0.001 0.01 - 0.05
., Standardized regression coefficient. a Odds ratios ŽOR. were calculated for a 1 U increase in given variables.
))
ps 0.59 vs. non-fatty liver subjects. p- 0.05 vs. non-fatty liver subjects.
))
1.01–3.68., but the Gly16 mutation did not differ ŽTable 5.. In fatty liver subjects, BMI Ž p - 0.001, OR 1.38, 95% CI: 1.25–1.53., fasting blood glucose Ž p - 0.05, OR 1.22, 95% CI: 0.94–1.58., insulin Ž p - 0.001, OR 1.39, 95% CI: 1.25–1.54., HOMA index Ž p - 0.001, OR 2.91, 95% CI: 2.05–4.13., total cholesterol Ž p s 0.003, OR 1.61, 95% CI: 1.17–2.22. and triglyceride Ž p - 0.001, OR 2.91, 95% CI: 1.97–4.30. levels were significantly higher than in non-fatty liver subjects. Moreover, serum concentrations of HDL-C were significantly lower than in non-fatty liver subjects Ž p - 0.001, OR 0.11, 95% CI: 0.05–0.25.. However, the differences in serum-free fatty acid levels were not statistically significant Ž p s 0.429, OR 0.94, 95% CI: 0.29–2.98. ŽTable 6..
3.4. MultiÕariate logistic regression analysis of fatty liÕer Significant predictors of fatty liver identified by the univariate analysis ŽBMI, HOMA index, HDLcholesterol, triglyceride, total cholesterol and B2AR genotype in codon 27. were included in a forward, stepwise multiple logistic regression model. This analysis demonstrated that the HOMA index Ž p 0.001, OR 2.03, 95% CI: 1.37–3.02., BMI Ž p 0.001, OR 1.22, 95% CI: 1.09–1.38. and triglyceride Ž p s 0.01, OR 1.78, 95% CI: 1.17–2.71. were retained as independent variables, and HDL-cholesterol as an independent preventive predictor Ž p - 0.05, OR 0.30, 95% CI: 0.11–0.83.. These models explained 34.5% of the variation in fatty liver incidence ŽTable 7.. B2AR genotype in codon 27 was
Table 6 Univariate analysis of fatty liver on clinical variables Variables
Controls Ž n s 119.
Fatty liver Ž n s 132.
OR Ž95% CI. a
.
p value
BMI Žkgrm2 . FPG Žmmolrl. Insulin ŽUrml. HOMA index HDL-C Žmmolrl. TG Žmmolrl. T-cho Žmmolrl. Glu27 allele frequency FFA ŽmEqrml.
24.5 " 3.2 5.26 " 1.0 8.38 " 2.8 1.97 " 0.8 1.59 " 0.5 1.31 " 0.7 5.10 " 0.8 0.07 0.51 " 0.2
27.6 " 3.2 5.50 " 1.3 12.3 " 5.5 3.05 " 1.7 1.29 " 0.3 2.11 " 1.3 5.43 " 0.9 0.12 0.51 " 0.2
1.38 Ž1.25–1.53. 1.22 Ž0.94–1.58. 1.39 Ž1.25–1.54. 291 Ž2.05–4.13. 0.11 Ž0.05–0.25. 2.91 Ž1.97–4.30. 1.61 Ž1.17–2.22. 1.92 Ž1.01–3.68. 0.94 Ž0.29–2.98.
0.324 0.200 0.330 1.067 y2.225 1.069 0.478 0.652 y0.066
- 0.001 - 0.005 - 0.001 - 0.001 - 0.001 - 0.001 - 0.005 - 0.05 NS
., Standardized regression coefficient. Data are means" S.D. a Odds ratios ŽOR. were calculated for a 1 U increase in the given variables.
90
N. Iwamoto et al.r Clinica Chimica Acta 314 (2001) 85–91
not identified as an independent predictor of fatty liver.
4. Discussion Large et al. w14x found that Gln27 to Glu was markedly associated with obesity. However, no significant association with obesity was observed in our study. Since obesity is not under a single genetic influence, this discrepancy could be explained by a variety of genetic background or environmental factors w22,23x. The polymorphisms in codons 16 and 27 were in a strong linkage disequilibrium w24x. Our study showed that all of the Glu27 allele carrying subjects had one or two Gly16 allele Ždata not shown.. The Gly16 homozygotes had a significantly lower serum HDL-cholesterol level than the Arg16 homozygotes, but in non-Glu27 carriers, the codon 16 variant was more weakly associated with HDL-cholesterol than in overall subjects Ždata not shown.. This might suggest that the codon 27 variant reflects the HDLcholesterol level. In the P3-adrenergic receptor gene, the Trp64Arg mutation was not only associated with visceral obesity but also with lower serum triglyceride levels. The reason is that B3AR is almost expressed in the adipose tissue and lipolysis could be decreased w25x. In contrast to the b2-adrenergic gene, we observed significant differences in serum triglyceride levels in codon 27 of the beta2-adrenergic receptor genotypes. Green et al. w26x reported that the Glu27 variant of B2AR in recombinant cells is more resistant to agonist-promoted down-regulation than wild-type. We speculated that the reduction in receptor down-regulation might lead to the release of non-esterified fatty acids from the adipose tissue, which results in hypertriglyceridemia. Therefore, we additionally measured serum-free fatty acid levels, but did not detect any differences in concentration as a factor of B2AR genotype. The b2-adrenergic receptor is expressed not only in the adipose tissue, but also in other tissues, such as pancreatic beta cells, skeletal muscle and hepatocytes that regulate triglyceride secretion and assembly. Therefore, the association of the polymorphisms with hypertriglyceridemia might be attributable to beta2-adrenergic receptors in tissues
other than adipose tissue. Recently, polymorphisms of the 5X-leader cistron of the human B2AR have been shown to regulate receptor expression, and linkage disequilibrium has been observed between these polymorphisms and the B2AR-coding block polymorphisms Arg16 and Gln27 w27,28x. The alterations in the B2AR gene expression levels may also lead to influence triglyceride metabolism. Theoretically, fatty liver could accumulate fat through several mechanisms. Dietary fat and fatty acids made by lipolysis in the adipose tissue may be released in the circulation and taken up by the liver. Triglyceride production may be augmented by increased mitochondrial synthesis of fatty acids or by reduced oxidation w29,30x. In our study, univariate analysis showed that the prevalence of fatty liver was significantly higher in the Glu27 mutant allele subjects than in the wild-type, although multiple logistic regression analysis did not identify the allele as a significant prognostic variable. These results suggest that B2AR polymorphisms may indirectly induce fatty liver through their relationship with triglycerides. It has not yet been explicated that the Gln27 to Glu mutation induces hypertriglyceridemia. Further studies are required to reveal the functional differences of this variant.
Acknowledgements We thank the staff of the Metabolism and Endocrinology Division in Saga Medical School, who did the DNA extraction and abdominal ultrasonography.
References w1x McCullough AJ, Falck-Ytter Y. Body consumption and hepatic steatosis as precursors for fibrotic liver disease. Hepatology 1999;29:1328–30. w2x Day CP, Yeaman SJ. The biochemistry of alcoholic fatty liver. Biochim Biophys Acta 1994;1215:33–48. w3x Braillon A, Capron JP, Herve MA, Degott C, Quenum C. Liver in obesity. Gut 1985;26:133–9. w4x Bacon BR, Farahvash MJ, Janney CG, Neuschwander-Tetri BA. Nonalcoholic steatohepatitis: an expanded clinical entity. Gastroenterology 1994;107:1103–9.
N. Iwamoto et al.r Clinica Chimica Acta 314 (2001) 85–91 w5x Enocksson S, Shimizu M, Lonnqvist F, Nordenstorm J, Arner P. Demonstration of an in vivo functional beta 3adrenoceptor in man. J Clin Invest 1995;95Ž5.:2239–45. w6x Lonnqvist F, Wahrenberg H, Hellstorm L, Reynisdottir S, Arner P. Lipolytic cathecholamine resistance due to decreased beta 2-adrenoceptor expression in fat cells. J Clin Invest 1992;90Ž6.:2175–86. w7x Clement K, Vaisse C, Manning BS, Basdevant A, Guy-Grand B, Ruiz J, et al. Genetic variation in the P3-adrenergic receptor and an increased capacity to gain weight in patients with morbid obesity. N Engl J Med 1995;333:352–4. w8x Walston J, Silver K, Bogardus G, Knowler WC, Ceil FS, Austin S, et al. Time of onset of non-insulin-dependent diabetes mellitus and genetic variation in the P3-adrenergic receptor gene. N Engl J Med 1995;333:343–7. w9x Widen E, Lehto M, Kanninenn T, Walston J, Shuldiner AR, Groop LC. Association of a polymorphism in the P3-adrenergic receptor gene with features of the insulin resistance syndrome in Finns. N Engl J Med 1995;333:348–51. w10x Kadowaki H, Yasuda K, Iwamoto K, Otabe S, Shimokawa K, Silver K, et al. A mutation in the P3-adrenergic receptor gene is associated with obesity and hyperinsulinemia in Japanese subjects. Biochem Biophys Res Commun 1995; 215:555–60. w11x Shima Y, Tsukada T, Nakanishi K, Ohta H. Association of the Trp64Arg mutation of the P3-adrenergic receptor with fatty liver and mild glucose intolerance in Japanese subjects. Clin Chim Acta 1998;274:167–76. w12x Reihsaus E, Innis M, MacIntyre N, Liggett SB. Mutations in the gene encoding for the beta2-adrenergic receptor in normal and asthmatic subjects. Am Rev Respir Cell Mol Biol 1993;8:334–9. w13x Ehrenborg E, Skogsberg J, Ruotolo G, Large V, Eriksson P, Aner P, et al. The QrE27 polymorphism in the beta2-adrenoceptor gene is associated with increased body weight and dyslipoproteinaemia involving triglyceride-rich lipoprotein. J Intern Med 2000;247Ž6.:651–6. w14x Large VL, Hellstorm S, Reynisdottir F, Reynisdottir F, Lonnqvist P, Eriksson L, et al. Human beta-2 adrenoceptor gene polymorphisms are highly frequent in obesity and associated with altered adipocyte beta-2 adrenoceptor function. J Clin Invest 1997;100:3005–13. w15x Mori Y, Kim-Motoyama H, Ito Y, Katakura T, Yasuda K, Ishiyama-Shigemoto S, et al. The Gln27GluP2-adrenergic receptor variant is associated with obesity due to subcutaneous fat accumulation in Japanese men. Biochem Biophys Res Commun 1999;258:138–40. w16x Ishiyama-Shigemoto S, Yamada K, Yuan X, Ichikawa F, Nonaka K. Association of polymorphisms in the P2P adrenergic receptor gene with obesity, hypertriglyceridemia, and diabetes mellitus. Diabetologia 1999;42:98–101. w17x Lonardo A, Trande P. Are there any sex differences in fatty
w18x
w19x
w20x
w21x
w22x
w23x
w24x
w25x
w26x
w27x
w28x
w29x
w30x
91
liver? A study of glucose metabolism and body fat distribution. J Gastroenterol Hepatol 2000;15:775–82. Cigolini M, Taragher G, Agostino G, Tonili M, Muggeo M, Sandre G. Liver steatosis and its relation to plasma haemostatic factors in apparently healthy men: role of the metabolic syndrome. Thromb Haemostasis 1996;76:69–73. Lonaaardo A, Bellini M, Tondelli E. The bright liver syndrome: prevalence and determinants of a ‘bright’ liver echopattern. Ital J Gastroenterol Hepatol 1998;29:351–6. Perry IJ, Wannamethee SG, Shaper AG. Prospective study of serum P-glutamyltransferase and risk of NIDDM. Diabetes Care 1998;21:732–7. Matthews DR, Hosker JP, Rudenski AS, Naylor RA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and P-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1998;28:412–9. Hayakawa T, Nagai Y, Kahara T, Yamashita H, Takamura T, Abe T, et al. Gln27Glu and Arg16Gly polymorphisms of the beta2-adrenergic receptor gene are not associated with obesity in Japanese men. Metabolism 2000;49Ž9.:1215–8. Kawamura T, Egusa G, Fujikawa R, Okubo M. Gln27Glu variant of the beta2-adrenergic receptor gene is not associated with obesity and diabetes in Japanese-Americans. Metabolism 2000;50Ž4.:443–6. Drysdale CM, McGraw DW, Stack CB, Stephens JC, Judson RS, Nandabalan K, et al. Complex promoter and coding region P2-adrenergic receptor haplotypes alter receptor expression and predict in vivo responsiveness. Proc Natl Acad Sci U S A 2000;97Ž19.:10483–8. Kim-Motoyama H, Yasuda K, Yamaguchi T, Yamada N, Katakura T, Shuldiner AR. A mutation of the beta 3-adrenergic receptor is associated with visceral obesity but decreased serum triglyceride. Diabetologia 1997;40:469–72. Green SA, Turki J, Innis M, Liggett SB. Amino-terminal polymorphisms of the human P2-adrenergic receptor impart distinct agonist-promoted regulatory properties. Biochemistry 1994;33:9414–9. McGraw DW, Forbes SL, Kramer LA, Liggett SB. PolymorX phisms of the 5 leader cistron of the human P2-adrenergic receptor regulate receptor expression. J Clin Invest 1998; 102:1927–32. Yamada K, Ishiyama-Shigemoto S, Ichikawa F, Yuan X, X Koyanagi A, Koyama W, et al. Polymorphisms in the 5 leader cistron of the P2-adrenergic receptor gene associated with obesity and type 2 diabetes. J Clin Endocrinol Metab 1999;84:1754–7. Oliver FW, Christopher PD. Non-alcoholic steatohepatitis ŽNASH.: a disease of emerging identity and importance. J Hepatol 1998;29:495–501. Geoffrey FG, Khwaja I, Richard JP. Mobilisation of triglycerol stores. Biochim Biophys Acta 2000;1483:37–57.
Gln27Glu b2-adrenergic receptor variant is associated with hypertriglyceridemia and the development of fatty liver Noriyuki Iwamoto ) , Yoko Ogawa, Susumu Kajihara, Akitaka Hisatomi, Tsutomu Yasutake, Toru Yoshimura, Toshihiko Mizuta, Toshiya Hara, Iwata Ozaki, Kyosuke Yamamoto DiÕision of Metabolism and Endocrinology, Department of Internal Medicine, Saga Medical School, 5-1-1 Nabeshima, Saga 849, Japan Received 4 April 2001; received in revised form 18 July 2001; accepted 23 July 2001
Abstract Background: Nonalcoholic steatohepatitis ŽNASH. is associated with the metabolism of lipid, glucose and energy. Beta-adrenergic receptors play an important role in the regulation of energy expenditure, in part, by stimulating lipid mobilization through lipolysis. Methods: To assess whether it is common for the b2-adrenergic receptor ŽB2AR. gene polymorphisms in codons 16 and 27 to play a role in the development of fatty liver, we investigated 251 unrelated healthy Japanese males who were drug-free and showed no signs of heavy drinking. Results: The allelic frequency of B2AR gene mutation in codons 16 and 27 did not differ between obese subjects ŽBMI ) 25.0 kgrm2 , n s 151. and non-obese subjects ŽBMI F 25.0 kgrm2 , n s 100.. The Gly16 homozygotes had a lower high-density lipoprotein cholesterol ŽHDL-C. level than the Arg16 homozygotes Ž1.50 " 0.4 vs. 1.32 " 0.3 mmolrl, p s 0.014.. However, no significant association with fatty liver was observed in the Gly16 allele frequency. The Gln27Glu27 heterozygotes showed higher concentrations of serum triglycerides ŽTG. than the Gln27Gln27 homozygotes Ž1.62 " 0.93 vs. 2.21 " 1.67 mmolrl, p s 0.013.. This correlation was also observed in all subjects regardless of weight classification. Univariate analysis indicated that subjects with the heterozygous Gln27Glu mutant alleles had a significantly higher prevalence of fatty liver vs. those without the mutation ŽGlu27 allele frequency, 0.07 vs. 0.12, p s 0.047; odds ratio, 1.92; 95% confidence interval, 1.01–3.68.. However, multivariate logistic regression models showed the prevalence of fatty liver to be significantly related to the homeostasis model assessment ŽHOMA. index, BMI, triglyceride and HDL-cholesterol. Conclusions: These results suggest that the amino-terminal polymorphisms of the b2-adrenergic receptor gene in codon 27 were associated with hypertryglyceridemia and independent of obesity, and thereby could be involved in the molecular pathogenesis of fatty liver. q 2001 Elsevier Science B.V. All rights reserved. Keywords: Beta2-adrenergic receptor gene; Polymorphism; Nonalcoholic steatohepatitis Žfatty liver.; Triglyceride; Obesity
1. Introduction
)
Corresponding author. Tel.: q81-952-31-6511; fax: q81-95234-2017. E-mail address: [email protected] ŽN. Iwamoto..
Fatty liver is one of the most common liver diseases found in clinical medicine. Generally, fatty liver change consists of some disruption of the normal physiology of triglyceride synthesis and secre-
0009-8981r01r$ - see front matter q 2001 Elsevier Science B.V. All rights reserved. PII: S 0 0 0 9 - 8 9 8 1 Ž 0 1 . 0 0 6 3 3 - 7
86
N. Iwamoto et al.r Clinica Chimica Acta 314 (2001) 85–91
tion, but the precise pathogenetic mechanisms are incompletely defined. Depending on the exact cause, multiple factors are likely responsible. It has been long recognized that fatty liver Žhepatic steatosis. occurs frequently in heavy alcohol drinkers. Although alcoholic liver disease is the most common cause, nonalcoholic hepatitis ŽNASH. is increasingly recognized as a significant form of liver disease. It is for this reason that NASH has been an important precursor to the development of fibrosis in diverse conditions such as hepatitis C, and alcoholic and nonalcoholic liver diseases w1,2x. It has been reported that more than 50% of obesity patients w3x and individuals with diabetes mellitus andror hyperlipidemia have developed fatty liver w4x. Catecholamines via b2–3 adrenergic receptors regulate body fat accumulation and energy expenditure. This regulation is in part affected by the stimulation of lipid mobilization through lipolysis in fat cells, which has been associated with obesity w5,6x. In the b3-adrenergic receptor ŽB3AR. gene, a missense mutation replacing Trp with Arg at codon 64 was reported to be associated with obesity-related metabolism disorders, such as insulin resistance and fatty liver w7–11x. Several polymorphisms have been found in the human B2AR gene w12x. These occur in the amino terminus of the receptor at amino acids 16 ŽArg or Gly. and 27 ŽGln or Glu. and in the fourth transmembrane-spanning domain at amino acid 164 ŽThr or Ile.. The amino acid substitutions at position 27 were shown to be associated with obesity in Swedish men and women w13,14x. In Japanese subjects, receptor variants at codon 27 were associated with insulin resistance w15,16x. However, it is unknown whether these genetic variants are associated with fatty liver. In the present study, we studied the common missense mutation of the B2AR gene in 251 Japanese male subjects to assess the potential association of gene mutation with nonalcoholic fatty liver.
S.D.. and 119 similarly unrelated subjects with nonfatty subjects aged 36–50 years Ž43.0 " 4.8 years. from 1111 patients who attended the Saga Health Care Center for health screening. The criteria for exclusion were as follows: heavy drinkers Ževeryday alcohol intake., treatment with any drug for diseases, known chronic liver disease and cancer, and females. We selected men only because fatty liver is found more often in males, and could be gender-related w17x. The diagnosis of fatty liver was based on the finding of liver–kidney contrast of abdominal ultrasonography. Liver–kidney contrast was held to be present in the case of significantly increased echogenicity of the liver compared with the right kidney by a gray scale real-time scanner w17–20x. Ultrasonography was performed in all cases by a single observer in a double-blind fashion. Informed consent was obtained from the subjects. Diagnosis of obesity ŽBMI ) 25 kgrm2 . was based on BMI according to the 1999 Japan Society for the Study of Obesity guideline.
2. Subjects and methods
2.3. Restriction fragment length polymorphism– polymerase chain reaction (RFLP–PCR)
2.2. Methods After overnight fasting, blood samples were taken from an antecubital vein, centrifuged Ž5000 rpm, 10 min. and stored at y20 8C prior to analysis. Serum concentrations of aspartate aminotransferase ŽAST., alanine aminotransferase ŽALT., P-glutamylglutamyltransferase ŽGGT., total cholesterol ŽT-cho., triglycerides ŽTG., high-density lipoprotein cholesterol ŽHDL-C. and fasting plasma glucose ŽFPG. were measured by the Saga Health Care Center’s routine chemistry laboratory ŽHitachi 7600.. Serumfree fatty acid ŽFFA. was measured by enzymatic assay, and plasma insulin level was analyzed by radioimmunoassay ŽSRL, Saga, Japan.. Blood pressure was measured in the right arm with the patient at rest and seated.
2.1. Subjects We randomly selected 132 subjects with fatty liver, aged 35–50 years Ž42.2 " 4.6 years, means"
Genomic DNA was extracted from peripheral leukocytes by digestion with proteinase K followed by phenolrchloroform extraction. The B2AR gene
N. Iwamoto et al.r Clinica Chimica Acta 314 (2001) 85–91
sequences were done by PCR in a volume of 10 l containing 0.1 U LATaq DNA polymerase ŽTakara, Japan., 1.5 mmolrl MgCl 2 , 100 molrl dNTPs, 10 mmolrl Tris–HCl pH 8.3 and 50 mmolrl of KCl. We used the following primers: the forward primers were 5X-CTTCTTGCTGGCACGCAAT-3X and 5XCCAGTGAAG-TGATGAAGTAGTTGG-3X for codon 16; and 5X-GGCCCATGACCAGATCAGCA3X and 5X-GAATGAGGCTTCCAGGCGTC-3X for codon 27. DNA was amplified by an initial denaturation step at 94 8C for 5 min, followed by 35 cycles of denaturation Ž98 8C, 20 s., annealing Ž55 8C for codon 16 or 65 8C for codon 27, 20 s. and extension Ž72 8C, 1 min., and a final extension at 72 8C for 10 min. Aliquots Ž3 l. of PCR products were digested with the addition of BsrD1 Žcodon 16, 56 8C, 1 U. or Ita1 Žcodon 27, 37 8C, 1 U.. After overnight incubation, the digested samples were separated by electrophoresis on a 4% ultra-pure DNA agarose gel using Tris–borate EDTA Ž8.9 mmolrl Tris–borate, 0.2 mmolrl EDTA. buffer and visualized under ultraviolet illumination after staining with ethidium bromide w14x.
2.4. Statistical analysis The results are presented as means" S.D. Homeostasis model assessment ŽHOMA. index indicating insulin resistance was calculated following the methods of Matthews et al. w21x. All analyses were performed according to the intention-to-treat principle. We used the chi-square test and Fisher’s exact test for the categorical comparisons of the data. Differences in the means of continuous measurements were tested by the Student’s t test and checked by the Mann–Whitney U test. Significant predictors identified by univariate analysis were then included in a forward, stepwise multiple logistic regression model to identify the most important factors affecting the incidence of fatty liver. Relative risk was estimated by the odds ratios ŽOR. and their 95% confidence intervals Ž95% CI.. A p - 0.05 was considered statistically significant; all tests were twotailed. All statistical analyses were performed on a personal computer using the statistical package SPSS for Windows ŽVersion 7.5.1 SPSS, Japan..
87
3. Results 3.1. Allelic frequency of the B2AR Õariants according to BMI in Japanese men We detected 63 Ž25%. Arg16 homozygotes, 123 Ž49%. Arg16rGly16 heterozygotes, 65 Ž26%. Gly16rGly16 homozygotes and 202 Ž81%. Gln27 homozygotes, 49 Ž19%. Gln27Glu27 heterozygotes and no Glu27 homozygotes in a group of 251 studied subjects. The Glu27 allele was slightly higher in obese subjects, but the frequency between obese subjects ŽBMI ) 25 kgrm2 , n s 151. and non-obese subjects ŽBMI F 25 kgrm2 , n s 100. did not statistically differ in the Gly16 allele Ž0.51 vs. 0.50. and in the Glu27 allele Ž0.11 vs. 0.08. ŽTable 1.. 3.2. Phenotypic characteristics of different polymorphisms in codons 16 and 27 The phenotypic characteristics of polymorphisms in codon 16 are shown in Table 2. The Gly16 homozygotes had a lower HDL-cholesterol level than the Arg16 homozygotes Ž1.50 " 0.4 vs. 1.32 " 0.3 mmolrl, p - 0.05.. Serum triglyceride level showed higher values Ž1.67 " 1.0 vs. 1.91 " 1.5 mmolrl., but did not statistically differ. Table 3 shows the phenotypes of the codon 27 polymorphisms. Individuals with the Glu27 allele mutation showed a higher concentration of serum triglyceride than those without this mutation Ž2.21 " Table 1 Allelic frequency of the B2AR variant according to obesity Codon 16
ArgrArg Žwild.
ArgrGly Žhetero.
GlyrGly Žhomo.
Gly16 allele frequency
Non-obese Obese
26 Ž26%. 37 Ž25%.
48 Ž48%. 75 Ž49%.
26 Ž26%. 39 Ž26%.
0.50 0.51)
Codon 27
GlnrGln Žwild.
GlnrGlu Žhetero.
Glu27 allele frequency
Non-obese Obese
84 Ž84%. 118 Ž78%.
16 Ž16%. 33 Ž22%.
0.08 0.11) )
Non-obese Ž F 25.0 kgrm2 . and obese ŽBMI) 25.0 kgrm2 . according to the 1999 Japan Society for the Study of Obesity guideline. ) P s NS vs. non-obese subjects. )) ps NS vs. non-obese subjects.
N. Iwamoto et al.r Clinica Chimica Acta 314 (2001) 85–91
88
Table 2 Baseline characteristics of subjects with B2AR gene polymorphisms in codon 16
n Age Žyears. BMI Žkgrm2 . AST ŽIUrl. ALT ŽIUrl. GGT ŽIUrl. Systolic BP Žmm Hg. Diastolic BP Žmm Hg. FPG ŽmM. Insulin ŽUrml. HOMA index T-cho Žmmolrl. HDL-C Žmmolrl. TG Žmmolrl. FFA ŽmEqrml.
Arg16Arg
Arg16Gly
Gly16Gly
p value
ŽA.
ŽB.
ŽC.
A vs. B
A vs. C
63 Ž25.1%. 43.1 " 4.9 26.1 " 3.5 27.0 " 12.5 34.8 " 23.0 38.6 " 23.8 127.1 " 18.5 77.6 " 12.7 5.35 " 0.69 10.7 " 4.0 2.54 " 1.0 5.41 " 0.9 1.50 " 0.4 1.67 " 1.0 0.54 " 0.3
123 Ž49.0%. 42.3 " 4.6 26.1 " 3.6 26.8 " 11.8 36.2 " 26.4 38.1 " 37.3 126.0 " 16.0 74.8 " 11.9 5.48 " 1.46 10.2 " 4.1 2.52 " 1.0 5.26 " 0.9 1.45 " 0.4 1.68 " 1.0 0.49 " 0.2
65 Ž25.9%. 42.6 " 4.6 26.3 " 3.4 29.2 " 14.0 35.7 " 27.5 36.6 " 29.1 125.0 " 13.0 74.2 " 10.5 5.24 " 0.99 10.7 " 6.5 2.57 " 2.0 5.17 " 0.8 1.32 " 0.3 1.91 " 1.5 0.50 " 0.2
NS NS NS NS NS NS NS NS NS NS NS NS NS NS
NS NS NS NS NS NS NS NS NS NS NS - 0.05 NS NS
Data are means" S.D.; n, number of subjects; BP, blood pressure.
1.67 vs. 1.62 " 0.93 mmolrl, p - 0.05.. Moreover, this correlation was also observed in all subjects regardless of weight classification Ž1.42 " 0.92 vs. Table 3 Baseline characteristics of subjects with B2AR gene polymorphisms in codon 27
n Age Žyears. BMI Žkgrm2 . AST ŽIUrl. ALT ŽIUrl. GGT ŽIUrl. Systolic BP Žmm Hg. Diastolic BP Žmm Hg. FPG ŽmM. Insulin ŽUrml. HOMA index T-cho Žmmolrl. HDL-C Žmmolrl. TG Žmmolrl. overall Ž ns 251. obese Ž ns151. non-obese Ž ns100. FFA ŽmEqrml.
Gln27rGln27 Žwild.
Gln27rGlu27 Žmutant.
202 Ž80.5%. 42.5"4.7 26.1"3.4 27.6"12.9 36.0"27.3 38.0"30.8 126.1"16.2 75.8"12.0 5.32"0.99 10.4"5.0 2.51"1.5 5.28"0.86 1.45"0.43 1.62"0.93
49 Ž19.5%. 42.8"4.8 26.4"3.8 26.9"11.6 34.5"18.8 37.2"37.7 125.4"14.9 73.4"10.8 5.63"1.78 10.5"4.0 2.65"1.3 5.24"0.85 1.36"0.33 2.21"1.67
1.78"0.91
2.31"1.84
1.42"0.92
1.98"1.27
0.51"0.2
0.49"0.1
Data are means"S.D.; n, number of subjects; BP, blood pressure. No difference in any group.
1.98 " 1.27 mmolrl, p - 0.05, non-obese subjects. Ž1.78 " 0.91 vs. 2.31 " 1.84 mmolrl, p - 0.05, obese subjects. ŽTable 3.. However, both groups were comparable with respect to age, BMI, AST, ALT, GGT, blood pressure, insulin, FPG, HOMA index, total cholesterol, HDL-cholesterol and FFA ŽTable 3.. 3.3. UniÕariate analysis of fatty liÕer, B2AR genotype and clinical Õariables Subjects with fatty liver showed significantly higher concentrations of liver enzymes ŽTable 4.. The allelic frequency of the Glu27 mutation significantly differed between fatty liver subjects and nonfatty liver subjects Ž p - 0.05, OR 1.92, 95% CI:
Table 4 Liver-related enzymes of subjects with fatty liver and controls
n Age Žyears. AST ŽIUrl. ALT ŽIUrl. GGT ŽIUrl.
Fatty liver
Controls
132 42.2"4.9 Ž35–50. 29.5"12.7 43.8"26.7 44.5"35.5
119 43.0"4.8 Ž36–50. 25.2"12.2 26.8"21.7 30.526.5
Data are means"S.D.; n, number of subjects.
p value NS - 0.01 - 0.001 0.001
N. Iwamoto et al.r Clinica Chimica Acta 314 (2001) 85–91
89
Table 5 Association between polymorphisms in the B2AR gene and fatty liver
Table 7 Multivariate logistic regression analysis of fatty liver on clinical variables
Codon 16
Variables
OR Ž95% CI. a
HOMA index BMI Žkgrm2 . Triglyceride Žmmolrl. HDL cholesterol Žmmolrl.
2.03 Ž1.37–3.02. 1.22 Ž1.09–1.38. 1.78 Ž1.17–2.71.
0.710 0.202 0.576
0.30 Ž0.11–0.83.
y1.196
ArgrArg
Fatty liver 34 Ž26%. Non-fatty liver 29 Ž24%. Codon 27
GlnrGln
ArgrGly
GlyrGly
66 Ž50%. 32 Ž24%. 57 Ž48%. 33 Ž26%. GlnrGlu
0.49 ) 0.51
Glu27 allele frequency
Fatty liver 100 Ž76%. 32 Ž24%. 0.12 Non-fatty liver 102 Ž86%. 17 Ž14%. 0.07 )
Gly16 allele frequency
.
p value - 0.0005 - 0.001 0.01 - 0.05
., Standardized regression coefficient. a Odds ratios ŽOR. were calculated for a 1 U increase in given variables.
))
ps 0.59 vs. non-fatty liver subjects. p- 0.05 vs. non-fatty liver subjects.
))
1.01–3.68., but the Gly16 mutation did not differ ŽTable 5.. In fatty liver subjects, BMI Ž p - 0.001, OR 1.38, 95% CI: 1.25–1.53., fasting blood glucose Ž p - 0.05, OR 1.22, 95% CI: 0.94–1.58., insulin Ž p - 0.001, OR 1.39, 95% CI: 1.25–1.54., HOMA index Ž p - 0.001, OR 2.91, 95% CI: 2.05–4.13., total cholesterol Ž p s 0.003, OR 1.61, 95% CI: 1.17–2.22. and triglyceride Ž p - 0.001, OR 2.91, 95% CI: 1.97–4.30. levels were significantly higher than in non-fatty liver subjects. Moreover, serum concentrations of HDL-C were significantly lower than in non-fatty liver subjects Ž p - 0.001, OR 0.11, 95% CI: 0.05–0.25.. However, the differences in serum-free fatty acid levels were not statistically significant Ž p s 0.429, OR 0.94, 95% CI: 0.29–2.98. ŽTable 6..
3.4. MultiÕariate logistic regression analysis of fatty liÕer Significant predictors of fatty liver identified by the univariate analysis ŽBMI, HOMA index, HDLcholesterol, triglyceride, total cholesterol and B2AR genotype in codon 27. were included in a forward, stepwise multiple logistic regression model. This analysis demonstrated that the HOMA index Ž p 0.001, OR 2.03, 95% CI: 1.37–3.02., BMI Ž p 0.001, OR 1.22, 95% CI: 1.09–1.38. and triglyceride Ž p s 0.01, OR 1.78, 95% CI: 1.17–2.71. were retained as independent variables, and HDL-cholesterol as an independent preventive predictor Ž p - 0.05, OR 0.30, 95% CI: 0.11–0.83.. These models explained 34.5% of the variation in fatty liver incidence ŽTable 7.. B2AR genotype in codon 27 was
Table 6 Univariate analysis of fatty liver on clinical variables Variables
Controls Ž n s 119.
Fatty liver Ž n s 132.
OR Ž95% CI. a
.
p value
BMI Žkgrm2 . FPG Žmmolrl. Insulin ŽUrml. HOMA index HDL-C Žmmolrl. TG Žmmolrl. T-cho Žmmolrl. Glu27 allele frequency FFA ŽmEqrml.
24.5 " 3.2 5.26 " 1.0 8.38 " 2.8 1.97 " 0.8 1.59 " 0.5 1.31 " 0.7 5.10 " 0.8 0.07 0.51 " 0.2
27.6 " 3.2 5.50 " 1.3 12.3 " 5.5 3.05 " 1.7 1.29 " 0.3 2.11 " 1.3 5.43 " 0.9 0.12 0.51 " 0.2
1.38 Ž1.25–1.53. 1.22 Ž0.94–1.58. 1.39 Ž1.25–1.54. 291 Ž2.05–4.13. 0.11 Ž0.05–0.25. 2.91 Ž1.97–4.30. 1.61 Ž1.17–2.22. 1.92 Ž1.01–3.68. 0.94 Ž0.29–2.98.
0.324 0.200 0.330 1.067 y2.225 1.069 0.478 0.652 y0.066
- 0.001 - 0.005 - 0.001 - 0.001 - 0.001 - 0.001 - 0.005 - 0.05 NS
., Standardized regression coefficient. Data are means" S.D. a Odds ratios ŽOR. were calculated for a 1 U increase in the given variables.
90
N. Iwamoto et al.r Clinica Chimica Acta 314 (2001) 85–91
not identified as an independent predictor of fatty liver.
4. Discussion Large et al. w14x found that Gln27 to Glu was markedly associated with obesity. However, no significant association with obesity was observed in our study. Since obesity is not under a single genetic influence, this discrepancy could be explained by a variety of genetic background or environmental factors w22,23x. The polymorphisms in codons 16 and 27 were in a strong linkage disequilibrium w24x. Our study showed that all of the Glu27 allele carrying subjects had one or two Gly16 allele Ždata not shown.. The Gly16 homozygotes had a significantly lower serum HDL-cholesterol level than the Arg16 homozygotes, but in non-Glu27 carriers, the codon 16 variant was more weakly associated with HDL-cholesterol than in overall subjects Ždata not shown.. This might suggest that the codon 27 variant reflects the HDLcholesterol level. In the P3-adrenergic receptor gene, the Trp64Arg mutation was not only associated with visceral obesity but also with lower serum triglyceride levels. The reason is that B3AR is almost expressed in the adipose tissue and lipolysis could be decreased w25x. In contrast to the b2-adrenergic gene, we observed significant differences in serum triglyceride levels in codon 27 of the beta2-adrenergic receptor genotypes. Green et al. w26x reported that the Glu27 variant of B2AR in recombinant cells is more resistant to agonist-promoted down-regulation than wild-type. We speculated that the reduction in receptor down-regulation might lead to the release of non-esterified fatty acids from the adipose tissue, which results in hypertriglyceridemia. Therefore, we additionally measured serum-free fatty acid levels, but did not detect any differences in concentration as a factor of B2AR genotype. The b2-adrenergic receptor is expressed not only in the adipose tissue, but also in other tissues, such as pancreatic beta cells, skeletal muscle and hepatocytes that regulate triglyceride secretion and assembly. Therefore, the association of the polymorphisms with hypertriglyceridemia might be attributable to beta2-adrenergic receptors in tissues
other than adipose tissue. Recently, polymorphisms of the 5X-leader cistron of the human B2AR have been shown to regulate receptor expression, and linkage disequilibrium has been observed between these polymorphisms and the B2AR-coding block polymorphisms Arg16 and Gln27 w27,28x. The alterations in the B2AR gene expression levels may also lead to influence triglyceride metabolism. Theoretically, fatty liver could accumulate fat through several mechanisms. Dietary fat and fatty acids made by lipolysis in the adipose tissue may be released in the circulation and taken up by the liver. Triglyceride production may be augmented by increased mitochondrial synthesis of fatty acids or by reduced oxidation w29,30x. In our study, univariate analysis showed that the prevalence of fatty liver was significantly higher in the Glu27 mutant allele subjects than in the wild-type, although multiple logistic regression analysis did not identify the allele as a significant prognostic variable. These results suggest that B2AR polymorphisms may indirectly induce fatty liver through their relationship with triglycerides. It has not yet been explicated that the Gln27 to Glu mutation induces hypertriglyceridemia. Further studies are required to reveal the functional differences of this variant.
Acknowledgements We thank the staff of the Metabolism and Endocrinology Division in Saga Medical School, who did the DNA extraction and abdominal ultrasonography.
References w1x McCullough AJ, Falck-Ytter Y. Body consumption and hepatic steatosis as precursors for fibrotic liver disease. Hepatology 1999;29:1328–30. w2x Day CP, Yeaman SJ. The biochemistry of alcoholic fatty liver. Biochim Biophys Acta 1994;1215:33–48. w3x Braillon A, Capron JP, Herve MA, Degott C, Quenum C. Liver in obesity. Gut 1985;26:133–9. w4x Bacon BR, Farahvash MJ, Janney CG, Neuschwander-Tetri BA. Nonalcoholic steatohepatitis: an expanded clinical entity. Gastroenterology 1994;107:1103–9.
N. Iwamoto et al.r Clinica Chimica Acta 314 (2001) 85–91 w5x Enocksson S, Shimizu M, Lonnqvist F, Nordenstorm J, Arner P. Demonstration of an in vivo functional beta 3adrenoceptor in man. J Clin Invest 1995;95Ž5.:2239–45. w6x Lonnqvist F, Wahrenberg H, Hellstorm L, Reynisdottir S, Arner P. Lipolytic cathecholamine resistance due to decreased beta 2-adrenoceptor expression in fat cells. J Clin Invest 1992;90Ž6.:2175–86. w7x Clement K, Vaisse C, Manning BS, Basdevant A, Guy-Grand B, Ruiz J, et al. Genetic variation in the P3-adrenergic receptor and an increased capacity to gain weight in patients with morbid obesity. N Engl J Med 1995;333:352–4. w8x Walston J, Silver K, Bogardus G, Knowler WC, Ceil FS, Austin S, et al. Time of onset of non-insulin-dependent diabetes mellitus and genetic variation in the P3-adrenergic receptor gene. N Engl J Med 1995;333:343–7. w9x Widen E, Lehto M, Kanninenn T, Walston J, Shuldiner AR, Groop LC. Association of a polymorphism in the P3-adrenergic receptor gene with features of the insulin resistance syndrome in Finns. N Engl J Med 1995;333:348–51. w10x Kadowaki H, Yasuda K, Iwamoto K, Otabe S, Shimokawa K, Silver K, et al. A mutation in the P3-adrenergic receptor gene is associated with obesity and hyperinsulinemia in Japanese subjects. Biochem Biophys Res Commun 1995; 215:555–60. w11x Shima Y, Tsukada T, Nakanishi K, Ohta H. Association of the Trp64Arg mutation of the P3-adrenergic receptor with fatty liver and mild glucose intolerance in Japanese subjects. Clin Chim Acta 1998;274:167–76. w12x Reihsaus E, Innis M, MacIntyre N, Liggett SB. Mutations in the gene encoding for the beta2-adrenergic receptor in normal and asthmatic subjects. Am Rev Respir Cell Mol Biol 1993;8:334–9. w13x Ehrenborg E, Skogsberg J, Ruotolo G, Large V, Eriksson P, Aner P, et al. The QrE27 polymorphism in the beta2-adrenoceptor gene is associated with increased body weight and dyslipoproteinaemia involving triglyceride-rich lipoprotein. J Intern Med 2000;247Ž6.:651–6. w14x Large VL, Hellstorm S, Reynisdottir F, Reynisdottir F, Lonnqvist P, Eriksson L, et al. Human beta-2 adrenoceptor gene polymorphisms are highly frequent in obesity and associated with altered adipocyte beta-2 adrenoceptor function. J Clin Invest 1997;100:3005–13. w15x Mori Y, Kim-Motoyama H, Ito Y, Katakura T, Yasuda K, Ishiyama-Shigemoto S, et al. The Gln27GluP2-adrenergic receptor variant is associated with obesity due to subcutaneous fat accumulation in Japanese men. Biochem Biophys Res Commun 1999;258:138–40. w16x Ishiyama-Shigemoto S, Yamada K, Yuan X, Ichikawa F, Nonaka K. Association of polymorphisms in the P2P adrenergic receptor gene with obesity, hypertriglyceridemia, and diabetes mellitus. Diabetologia 1999;42:98–101. w17x Lonardo A, Trande P. Are there any sex differences in fatty
w18x
w19x
w20x
w21x
w22x
w23x
w24x
w25x
w26x
w27x
w28x
w29x
w30x
91
liver? A study of glucose metabolism and body fat distribution. J Gastroenterol Hepatol 2000;15:775–82. Cigolini M, Taragher G, Agostino G, Tonili M, Muggeo M, Sandre G. Liver steatosis and its relation to plasma haemostatic factors in apparently healthy men: role of the metabolic syndrome. Thromb Haemostasis 1996;76:69–73. Lonaaardo A, Bellini M, Tondelli E. The bright liver syndrome: prevalence and determinants of a ‘bright’ liver echopattern. Ital J Gastroenterol Hepatol 1998;29:351–6. Perry IJ, Wannamethee SG, Shaper AG. Prospective study of serum P-glutamyltransferase and risk of NIDDM. Diabetes Care 1998;21:732–7. Matthews DR, Hosker JP, Rudenski AS, Naylor RA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and P-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1998;28:412–9. Hayakawa T, Nagai Y, Kahara T, Yamashita H, Takamura T, Abe T, et al. Gln27Glu and Arg16Gly polymorphisms of the beta2-adrenergic receptor gene are not associated with obesity in Japanese men. Metabolism 2000;49Ž9.:1215–8. Kawamura T, Egusa G, Fujikawa R, Okubo M. Gln27Glu variant of the beta2-adrenergic receptor gene is not associated with obesity and diabetes in Japanese-Americans. Metabolism 2000;50Ž4.:443–6. Drysdale CM, McGraw DW, Stack CB, Stephens JC, Judson RS, Nandabalan K, et al. Complex promoter and coding region P2-adrenergic receptor haplotypes alter receptor expression and predict in vivo responsiveness. Proc Natl Acad Sci U S A 2000;97Ž19.:10483–8. Kim-Motoyama H, Yasuda K, Yamaguchi T, Yamada N, Katakura T, Shuldiner AR. A mutation of the beta 3-adrenergic receptor is associated with visceral obesity but decreased serum triglyceride. Diabetologia 1997;40:469–72. Green SA, Turki J, Innis M, Liggett SB. Amino-terminal polymorphisms of the human P2-adrenergic receptor impart distinct agonist-promoted regulatory properties. Biochemistry 1994;33:9414–9. McGraw DW, Forbes SL, Kramer LA, Liggett SB. PolymorX phisms of the 5 leader cistron of the human P2-adrenergic receptor regulate receptor expression. J Clin Invest 1998; 102:1927–32. Yamada K, Ishiyama-Shigemoto S, Ichikawa F, Yuan X, X Koyanagi A, Koyama W, et al. Polymorphisms in the 5 leader cistron of the P2-adrenergic receptor gene associated with obesity and type 2 diabetes. J Clin Endocrinol Metab 1999;84:1754–7. Oliver FW, Christopher PD. Non-alcoholic steatohepatitis ŽNASH.: a disease of emerging identity and importance. J Hepatol 1998;29:495–501. Geoffrey FG, Khwaja I, Richard JP. Mobilisation of triglycerol stores. Biochim Biophys Acta 2000;1483:37–57.