The human liver is one of the target organs for estrogens, which play an important role as a mitogenic factor in both normal and diseased liver . Aromatase, an enzyme of the cytochrome P450 superfamily, is encoded by CYP19A1 located at chromosome 15q21.2 and plays a critical role in the biosynthesis of estrogens (C18 steroids) from androgens (C19 steroids) (Figure 1) , . 17β-Hydroxysteroid dehydrogenase (17β-HSD) is involved in interconversion of active and inactive sex steroids, including both estrogens and androgens. At least 11 distinct subtypes of 17β-HSD isozymes have been reported so far . Among these, 17β-HSD type 1 catalyzes the biologically inactive estrogen, estrone (E1), to active estrogen, estradiol (E2) (Figure 1); 17β-HSD type 2 has the opposite effect, and it catalyzes E2 to E1 (Figure 1). The ovary is the major source of circulating estrogens in premenopausal women, whereas in postmenopausal women and men, estrogen derives from peripheral aromatization of circulating androgen precursors in many extragonadal organs such as the skin, adipose tissues, liver, heart, osteoblasts and chondrocytes in the bone and brain , .
The established risk factors for hepatocellular carcinoma (HCC) are chronic infections with hepatitis B virus (HBV) or hepatitis C virus (HCV), alcohol abuse, or exposure to dietary aflatoxin in many parts of the world . Almost all patients with chronic HCV infection can achieve sustained virologic response (SVR) with direct-acting antivirals, which leads to significantly decreased rates of HCV-associated HCC . However, there remains a persistent risk for HCC, even among the patients without hepatitis and in those with SVR. For example, HCC is consistently 2–3 times higher in men than in women, suggesting that sex hormones such as androgens and estrogens could be both involved in HCC development and progression . Consistent with this hypothesis, long-term use of oral contraceptives and anabolic androgenic steroids is well known to be associated with an increased risk of development of liver tumors . In addition, animal model studies suggest that both androgens and estrogens could induce or promote HCC development  and that fadrozole, a nonsteroidal aromatase inhibitor, significantly reduces the incidence of spontaneous HCC in both male and female rats . Furthermore, aromatase expression is reported in several fibrolamellar HCCs in humans , , . Therefore, in this study, we immunolocalized and compared the immunoreactivity of aromatase, 17β-HSD type 1, and 17β-HSD type 2 among the normal human liver, fetal liver, chronic hepatitis (HBV or HCV), SVR (HBV-SVR or HCV-SVR), biliary atresia, alcoholic hepatitis, nonalcoholic steatohepatitis (NASH), primary sclerosing cholangitis (PSC) and primary biliary cholangitis (PBC).
Materials and methods
Tissue specimens and clinical data
We retrieved a total of 155 cases of clinical data and surgical pathology specimens at Tohoku University Hospital, Sendai, Japan (normal liver, n = 10; NASH, n = 18; PSC, n = 6, PBC, n = 13; biliary atresia, n = 18; alcoholic hepatitis, n = 11; HCV, n = 31; HCV-SVR, n = 10; HBV, n = 20; HBV-SVR, n = 8; infants, n = 10). The protocol for this study was approved by the Ethics Committee of Tohoku University School of Medicine, Sendai, Japan (2007-339, 2008-123, 2016-1-030). Informed consent was obtained from each patient. Of the total cases (except infants), we obtained 91 specimens from males and 54 from females (mean age, 53.3 years). We could not retrieve data regarding sex in fetal liver specimens.
We immunolocalized aromatase, 17β-HSD type 1 and 17β-HSD type 2. Details regarding immunohistochemical characterization and utilization were previously reported by the authors . The specimens were fixed in 10% formalin for 48–72 h at room temperature, embedded in paraffin, cut into 3-μm thick sections, and placed on glue-coated glass slides. The sections were deparaffinized in xylene and hydrated with graded alcohol solutions and distilled water. Endogenous peroxidase activity was blocked with 3% hydrogen peroxide for 10 min at room temperature. Antigen retrieval from tissue was performed using a microwave (500 W, 15 min) in citrate buffer (10 mmol, pH 6.0) for 17β-HSD type 1. We did not perform antigen retrieval for aromatase or 17β-HSD type 2. Sections were subsequently incubated for 30 min at room temperature in a blocking solution of 10% rabbit serum (Nichirei Bioscience, Tokyo, Japan) for aromatase and 17β-HSD type 1, and a blocking solution containing 10% goat serum (Nichirei Bioscience) was used for 17β-HSD type 2. The sections were incubated with primary antibodies for 16 h at 4 °C. Diluted monoclonal antibodies included anti-aromatase (Nichirei Bioscience; 1:500), anti-17β-HSD type 1 (Abnova, Heidelberg, Germany; 1:400) and 17β-HSD type 2 (Proteintech, Chicago, IL, USA; 1:200). Secondary antibody reactions were performed for 30 min at room temperature using diluted biotinylated rabbit anti-mouse antibodies (Nichirei Bioscience; 1:100), and peroxidase-conjugated avidin (Nichirei Bioscience) was added according to the manufacturer’s instructions. Reacted sections were visualized using 3,3′-diaminobenzidine-tetrachloride and 30% H2O2 in Tris buffer (0.05 mol, pH 7.6) and counterstained with hematoxylin. Placenta was used as a positive control for all of the antibodies.
Scoring of aromatase, 17β-HSD type 1 and 17β-HSD type 2 immunoreactivity
Immunoreactivity of aromatase, 17β-HSD type 1 and 17β-HSD type 2 was semiquantitatively evaluated using McCarty’s H-scoring system. Relative immunointensity was scored 0 (none), 1 (weak), 2 (moderate), or 3 (strong). The percentage of immunopositive cells and the relative immunointensity were multiplied, yielding a value ranging between 0 and 300 (Figure 2) .
Cell line and cytokine array
The HCC cell line (HepG2) was obtained from Cell Resource Center for Biomedical Research, Tohoku University, Sendai, Japan, and maintained in a RPMI-1640 medium (Sigma-Aldrich Co., St. Louis, MO, USA) with 10% fetal bovine serum. HepG2 cells were cultured in six well plates for 72 h, and HepG2 cell protein samples were extracted using Mammalian Protein Extraction Reagent (M-PER, Thermo Fisher Scientific, Waltham, MA, USA), with 1% Halt Protease Inhibitor Cocktail (Pierce Biotechnology, Rockford, IL, USA). HepG2 cytokines were analyzed using the RayBio Human Cytokine Antibody Array (#AAH-CYT-5, RayBiotech Inc., Norcross, GA, USA) according to the manufacturer’s instructions.
Mann-Whitney nonparametric tests were performed to compare continuous values among groups; p < 0.05 was considered significant. All analyses were performed using IBM SPSS Statistics version 24 for Windows (PASW Statistics for Windows, SPSS Inc., Chicago, IL, USA).
Immunohistochemical scores of aromatase in normal, diseased and fetal livers are summarized in Table 1. Aromatase immunoreactivity in normal liver was mainly detected in hepatocytic cytoplasm, adjacent to the portal area and central vein, and in some cases, weakly detected in cholangiocytes. The immunoreactivity score (mean ± standard deviation) in NASH (3.9 ± 9.8) was significantly lower than that in normal liver (26.0 ± 17.1) (p < 0.001). The immunoreactivity scores in HBV (59.5 ± 30.9, p = 0.006), HBV-SVR (68.1 ± 33.5, p = 0.001) and infants (100.5 ± 36.6, p = 0.001) were significantly higher than those in normal liver. No significant difference was observed between males and females in any etiology, and no significant difference was detected between non-cirrhotic and cirrhotic livers. We did not analyze the relationship between non-cirrhotic and cirrhotic livers in cases of PSC, PBC or biliary atresia because almost all the examined liver tissues were associated with decompensated cirrhosis and those liver tissues were derived from liver transplantation.
Immunohistochemistry of 17β-HSD type 1 and 17β-HSD type 2
17β-HSD type 1 and 17β-HSD type 2 immunohistochemistry scores in the normal, diseased, and fetal livers are summarized in Table 2 and Table 3. 17β-HSD type 1 immunoreactivity was detected in the cytoplasm of hepatocytes and Kupffer cells, and 17β-HSD type 2 in the cytoplasm of hepatocytes and some cholangiocytes. 17β-HSD type 1 immunoreactivity scores in any etiology other than HBV (116.3 ± 23.7, p = 0.350) and infants (120.0 ± 28.5, p = 0.912) were significantly lower than those in normal liver (122.5 ± 8.6) (Table 2). The immunoreactivity score of 17β-HSD type 2 in NASH (74.4 ± 36.6) was significantly lower than that in normal liver (128.0 ± 29.7) (p < 0.001) (Table 3). No statistical difference was detected between males and females or between non-cirrhotic and cirrhotic livers of any etiology in either 17β-HSD type 1 or type 2.
HepG2 cell lines secreted aromatase-inducible cytokines such as interleukin-1α (IL-1α), IL-β, IL-6, tumor necrosis factor (TNF)-α and oncostatin M (Figure 3).
Numerous studies have been conducted on the relationship between aromatase expression and breast carcinoma progression; however, only a few studies have reported a correlation between aromatase and HCC. In addition, the relationship between aromatase and hepatitis of any etiology has remained largely unknown. The human aromatase gene contains 10 exons, a number of alternative untranslated first exons (I) that are regulated by tissue-specific promoters and nine coding regions (exons II-X), with the ATG translational start site located in exon II . So far, 10 alternative tissue-specific promoters have been reported in humans, including promoters I.1, I.2 and I.2a in placenta; I.3 and I.4 in adipose tissue; I.5 in fetal tissue; I.6 in bone; I.7 in endothelial cells; I.f in the brain; and PII in gonads, adipose tissue and normal liver , , , . The promoter regions upstream of each of the first exons have different cohorts of response elements; therefore, regulation of aromatase expression in each tissue is different. Promoter II-mediated aromatase expression is stimulated by prostaglandin E2 (PGE2) , and promoter I.4 is regulated by IL-6, IL-11, oncostatin M and TNF-α . Normal breast adipose tissue maintains low levels of aromatase expression primarily via promoter I.4 ; however, in breast cancer tissue, PGE2 and other factors secreted from breast cancer cells result in alternative promoter use or promoter switching for aromatase expression, and the activities of the more potent promoters I.3 and II are markedly increased. Therefore, the total aromatase mRNA levels in breast cancer tissue are enormously higher than those in normal breast tissue , , .
With regard to aromatase expression in the liver, promoters I.3, I.4 and II are reported to be used in HCC and promoter I.4, I.5 and I.6 in fetal liver , , , . In normal adult liver, very little or low-level aromatase expression is driven by promoter II and is not necessarily regulated by the same factors involved in fetal liver and HCC . Recently, Koh et al.  reported an association between non-viral hepatitis-related HCC risk and a functional polymorphism in the aromatase promoter upstream of exon I.6. They revealed that transcriptional activity of C allele of the CYP19 I.6 promoter was 60% higher than in the A allele. Among individuals with absence of both HBV and HCV serologic markers, HCC risk was significantly higher in the subjects homozygous for the C allele than those homozygous for the A allele . Zhang et al. revealed that 17β-HSD type 1 (rs676387) and 17β-HSD type 2 (rs8191246) polymorphisms were associated with the risk of HCC development . With the epidemiological evidence of a male predominance in HCC, all of the experimental evidence supports the association of sex steroid hormones with HCC development and progression.
Harada et al.  reported locally increased aromatase protein expression in non-neoplastic hepatocytes around HCC and metastatic colorectal cancer. They described that local factors synthesized by liver primary or metastatic tumors or other cell elements might be secreted into the external environment, inducing aromatase expression in adjacent liver tissues. Aromatase expression was also reported in HCC too , , . Yabuuchi et al. reported a correlation between aromatase activity and the degree of histologic differentiation of HCC, or significantly higher activity in patients with Edmondson Grade II HCC than in those with Grade III . Aromatase expression is regulated by distinct sets of stimulating factors such as cytokines or prostaglandins. Hata et al.  reported that breast adenocarcinoma cell line (MCF-7) and colon adenocarcinoma cell line (DLD-1) as well as HepG2 secreted aromatase-inducible cytokines such as IL-1 and IL-6. Bauer et al.  revealed TNF-α expression in HepG2 cell line, and Lu et al. detected the expression of IL-1α and IL-1β in human HCC tissue , all of which constitute the aromatase-inducible cytokines. In the present study, we revealed that HepG2 secreted aromatase-stimulating factors such as IL-1α, IL-1β, IL-6, TNF-α and oncostatin M (Figure 3). These results all suggest that aromatase expression may be induced not only in non-neoplastic hepatocytes around HCC or metastatic liver carcinoma in a paracrine fashion but also in HCC too in an autocrine fashion (Figure 4). Castagnetta et al. reported that aromatase activity could be detected only in HCC epithelium, and not in stromal cells . In the present study, we did not detect aromatase immunoreactivity in Kupffer cells or stromal cells including fibroblasts, a big difference from that in breast cancer (Figure 4). We revealed that hepatocytes in HBV patients as well as HBV-SVR patients expressed significantly higher aromatase scores than in the normal liver as well as relatively high 17β-HSD type 1 scores. Granata et al.  proposed three major metabolic pathways of androgen metabolism in human liver tissue: estrogenic, androgenic and mixed. They demonstrated that aromatase-driven estrogen formation is consistently prevalent in HCC and that bioactive androgens, originating from the 5α-reductase pathway, are dominant in normal and HCV-cirrhotic liver tissues .
This study admittedly contains some limited values to allow a consolidated conclusion, because the size of the cohort is not sufficiently large and the methodology is confined. However, high aromatase scores in HBV and HBV-SVR tissues and relatively high 17β-HSD type 1 scores in HBV tissue do suggest a possible association between HBV and estrogen. Conversely, estrogen may play an important role in preventing HBV progression and HBV-related HCC development . Therefore, further investigations are required to clarify the relationships among aromatase expression, chronic hepatitis, particularly HBV-related hepatitis, and HCC occurrence, including how the mechanism is related to aromatase expression, what kind of aromatase-stimulating factors are involved, and which cells secrete aromatase-stimulating factors (hepatocytes, fibroblasts, or other cells in an autocrine or paracrine manner) (Figure 4).
We thank Ms. Yayoi Aoyama in the Tohoku University, Sendai, Japan for her excellent technical assistance.
Harada N, Ota H, Yoshimura N, Katsuyama T, Takagi Y. Localized aberrant expression of cytochrome P450 aromatase in primary and metastatic malignant tumors of human liver. J Clin Endocrinol Metab. 1998;83:697–702. PubMedGoogle Scholar
Chen Z, Wang O, Nie M, Elison K, Zhou D, Li M, et al. Aromatase deficiency in a Chinese adult man caused by novel compound heterozygous CYP19A1 mutations: effects of estrogen replacement therapy on the bone, lipid, liver and glucose metabolism. Mol Cell Endocrinol. 2015;399:32–42. CrossrefGoogle Scholar
Chen SA, Besman MJ, Sparkes RS, Zollman S, Klisak I, Mohandas T, et al. Human aromatase: cDNA cloning, Southern blot analysis, and assignment of the gene to chromosome 15. DNA. 1988;7:27–38. CrossrefPubMedGoogle Scholar
Peltoketo H, Luu-The V, Simard J, Adamski J. 17Beta-hydroxysteroid dehydrogenase (HSD)/17-ketosteroid reductase (KSR) family; nomenclature and main characteristics of the 17HSD/KSR enzymes. J Mol Endocrinol. 1999;23:1–11. PubMedCrossrefGoogle Scholar
Bulun SE, Lin Z, Imir G, Amin S, Demura M, Yilmaz B, et al. Regulation of aromatase expression in estrogen-responsive breast and uterine disease: from bench to treatment. Pharmacol Rev. 2005;57:359–83. CrossrefPubMedGoogle Scholar
Ioannou GN, Beste LA, Chang MF, Green PK, Lowy E, Tsui JI, et al. Effectiveness of sofosbuvir, ledipasvir/sofosbuvir, or paritaprevir/ritonavir/ombitasvir and dasabuvir regimens for treatment of patients with hepatitis C in the Veterans Affairs national health care system. Gastroenterology. 2016;151:457–71. PubMedCrossrefGoogle Scholar
Hany MA, Betts DR, Schmugge M, Schönle E, Niggli FK, Zachmann M, et al. A childhood fibrolamellar hepatocellular carcinoma with increased aromatase activity and a near triploid karyotype. Med Pediatr Oncol. 1997;28:136–8. CrossrefGoogle Scholar
Muramori K, Taguchi S, Taguchi T, Kohashi K, Furuya K, Tokuda K, et al. High aromatase activity and overexpression of epidermal growth factor receptor in fibrolamellar hepatocellular carcinoma in a child. J Pediatr Hematol Oncol. 2011;33:e195–7. CrossrefGoogle Scholar
Agarwal VR, Takayama K, Van Wyk JJ, Sasano H, Simpson ER, Bulun SE. Molecular basis of severe gynecomastia associated with aromatase expression in a fibrolamellar hepatocellular carcinoma. J Clin Endocrinol Metab. 1998;83:1797–800. Google Scholar
Miki Y, Suzuki T, Tazawa C, Yamaguchi Y, Kitada K, Honma S, et al. Aromatase localization in human breast cancer tissues: possible interactions between intratumoral stromal and parenchymal cells. Cancer Res. 2007;67:3945–54. PubMedCrossrefGoogle Scholar
McCarty KS Jr, Miller LS, Cox EB, Konrath J, McCarty KS Sr. Estrogen receptor analyses. Correlation of biochemical and immunohistochemical methods using monoclonal antireceptor antibodies. Arch Pathol Lab Med. 1985;109:716–21. PubMedGoogle Scholar
Koh WP, Yuan JM, Wang R, Govindarajan S, Oppenheimer R, Zhang ZQ, et al. Aromatase (CYP19) promoter gene polymorphism and risk of nonviral hepatitis-related hepatocellular carcinoma. Cancer. 2011;117:3383–92. PubMedCrossrefGoogle Scholar
Shozu M, Zhao Y, Bulun SE, Simpson ER. Multiple splicing events involved in regulation of human aromatase expression by a novel promoter, I.6. Endocrinology. 1998;139:1610–7. PubMedCrossrefGoogle Scholar
Zhao Y, Agarwal VR, Mendelson CR, Simpson ER. Estrogen biosynthesis proximal to a breast tumor is stimulated by PGE2 via cyclic AMP, leading to activation of promoter II of the CYP19 (aromatase) gene. Endocrinology. 1996;137:5739–42. PubMedCrossrefGoogle Scholar
Zhou J, Gurates B, Yang S, Sebastian S, Bulun SE. Malignant breast epithelial cells stimulate aromatase expression via promoter II in human adipose fibroblasts: an epithelial-stromal interaction in breast tumors mediated by CCAAT/enhancer binding protein beta. Cancer Res. 2001;61:2328–34. PubMedGoogle Scholar
Yamamoto T, Sakai C, Yamaki J, Takamori K, Yoshiji S, Kitawaki J, et al. Estrogen biosynthesis in human liver-a comparison of aromatase activity for C-19 steroids in fetal liver, adult liver and hepatoma tissues of human subjects. Endocrinol Jpn. 1984;31:277–81. CrossrefPubMedGoogle Scholar
Zhang LS, Yuan F, Guan X, Li J, Liu XL, Sun J, et al. Association of genetic polymorphisms in HSD17B1, HSD17B2 and SHBG genes with hepatocellular carcinoma risk. Pathol Oncol Res. 2014;20:661–6. PubMedCrossrefGoogle Scholar
Castagnetta LA, Agostara B, Montalto G, Polito L, Campisi I, Saetta A, et al. Local estrogen formation by nontumoral, cirrhotic, and malignant human liver tissues and cells. Cancer Res. 2003;63:5041–5. PubMedGoogle Scholar
Granata OM, Cocciadifero L, Campisi I, Miceli V, Montalto G, Polito LM, et al. Androgen metabolism and biotransformation in nontumoral and malignant human liver tissues and cells. J Steroid Biochem Mol Biol. 2009;113:290–5. CrossrefPubMedGoogle Scholar
Yabuuchi I, Kawata S, Tamura S, Ito N, Matsuda Y, Nishioka M, et al. Aromatase activity in human hepatocellular carcinoma. Relationship with the degree of histologic differentiation. Cancer. 1993;71:56–61. PubMedCrossrefGoogle Scholar
Bauer I, Al Sarraj J, Vinson C, Larsen R, Thiel G. Interleukin-1β and tetradecanoylphorbol acetate-induced biosynthesis of tumor necrosis factor α in human hepatoma cells involves the transcription factors ATF2 and c-Jun and stress-activated protein kinases. J Cell Biochem. 2007;100:242–55. CrossrefPubMedGoogle Scholar
Lu P, Nakamoto Y, Nemoto-Sasaki Y, Fujii C, Wang H, Hashii M, et al. Potential interaction between CCR1 and its ligand, CCL3, induced by endogenously produced interleukin-1 in human hepatomas. Am J Pathol. 2003;162:1249–58. CrossrefPubMedGoogle Scholar
About the article
Published Online: 2018-02-28
Research funding: Authors state no funding involved.
Conflict of interest: The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Informed consent: Informed consent is not applicable.
Ethical approval: The conducted research is not related to either human or animals use.