The importance of the UGT1A1 variants in the development of osteopenia and osteoporosis in postmenopausal women

The UDP-glucuronosyltransferase 1A1 (UGT1A1) is involved in the process of estrogen conjugation and elimination. The aim of the study was to analyze whether the UGT1A1 genetic variants are associated with the development of osteopenia and osteoporosis in postmenopausal women. The analysis of the rs4148323 (UGT1A1*6) and rs3064744 (UGT1A1*28) variants in the UGT1A1 gene was conducted using real-time PCR. A significant correlation was observed between the genotypes of the rs3064744 (UGT1A1*28) sequence variant and body mass in women with osteoporosis. The analysis of the Z-score values revealed that women with osteoporosis and carrying the 6/6 variant had the lowest Z-score values as compared to women with the 6/7 and the 7/7 variants (− 1.966 ± 0.242 vs. − 1.577 ± 0.125 and − 1.839 ± 0.233). In addition, the odds ratio for the investigated genotypes (6/6, 6/7, 7/7) indicated an increased risk for osteopenia and osteoporosis in women with the 7/7 homozygous genotype. The analysis of the frequencies of the GG, GA and AA genotypes of the rs4148323 UGT1A1 gene showed no statistically significant differences between the groups. Our analysis revealed that the UGT1A1 rs3064744 variant may affect the risk of developing osteoporosis in postmenopausal Polish women. The UGT1A1 rs4148323 variant is not directly associated with the development of osteopenia and osteoporosis.


Scientific Reports
| (2021) 11:17385 | https://doi.org/10.1038/s41598-021-96429-x www.nature.com/scientificreports/ A better understanding of the molecular mechanisms underlying osteoporosis is vital for the diagnosis and treatment, not to mention the earliest possible identification of the factors predisposing to the development of the disease 5 . So far, numerous molecular analyses were performed to investigate the possible role of the genetic factors in the etiology of osteoporosis 6 . The risk of osteoporosis and osteopenia has been linked to genetic variants, especially the COL1A1, VDR, BMP2, TLR genes, as well as the LRP5 gene involved in the Wnt/β-catenin signaling pathway [7][8][9][10][11][12] .
The search for new genes which play an important role in the regulation of bone mass and the development of osteoporosis continues. Previous GWAS studies demonstrated a link between polymorphisms of the genes related to estrogen metabolism and osteoporosis and the risk of bone fracture 12 . Recently, much attention has been paid to the potential biomarkers, including UDP-glucuronosyltransferases (UGTs).
UDP-glucuronosyltransferases comprise a superfamily of membrane-bound conjugating enzymes involved in the inactivation and elimination of numerous endogenous and exogenous compounds. UGTs catalyze the glucuronidation reaction, which is associated with the metabolism of bilirubin, bile acids, fatty acids, steroid hormones, thyroid hormones, and fat-soluble vitamins 12,13 . Glucuronidation is also one of the most important phase II biotransformation reactions 14 . UGTs are expressed in various tissues: brain, liver, kidneys, small intestine, colon, stomach, lungs, epithelium, ovaries, testes, mammary glands, and prostate [13][14][15] . UGT1A1 is expressed in the uterus and is involved in the conjugation and elimination of estrogens. Studies indicate that permanent estrogen deficiency after menopause is the main cause of osteoporosis in older women 16 . However, the relationship between osteoporosis and the UGT1A1 gene variant in Caucasian postmenopausal women remains to be fully elucidated. Therefore, researchers are constantly looking for new genetic variants that could affect the risk of developing osteoporosis. Conducted scientific studies analyze the influence of genetic variants of the UGT1A1, including UGT1A1*6 (211G>A, rs4148323), UGT1A1*27 (686C>A, rs35350960), UGT1A1*60 (− 3263T>A, rs4124874), and TA repeat variation of UGT1A1*28 (A(TA) 7 TAA, rs3064744) on the risk of developing osteoporosis or other pathological entities, e.g. Gilbert's Syndrome. Moreover, it has been shown that the UGT1A1*28 variant influences glucuronidation of bazedoxifene used for the prevention and treatment of osteoporosis.
The aim of the study was to investigate whether the UGT1A1 rs3064744 (UGT1A1*28) and the rs4148323 (UGT1A1*6) genetic variants are associated with the development of osteopenia and osteoporosis in postmenopausal women.

Methods
Patients. The study included 675 Polish postmenopausal women (109 with osteopenia, 333 with osteoporosis and 233 healthy controls). BMD measurements were performed at the Laboratory of Densitometry, Clinical Hospital No. 1, Pomeranian Medical University in Szczecin. BMD was measured in the lumbar spine, from L2 to L4 vertebrae, using DEXA (Dual Energy X-ray Absorptiometry). Densitometry was performed using the LUNAR DPX 100 camera (Lunar Corp., Madison, USA). Normal BMD value using DEXA is between one standard deviation from the mean in relation to the age of peak bone mass (− 1 < T-score > 1). Based on these measurements, the women were classified into the following groups: osteopenia (− 2.5 < T-score < − 1), osteoporosis (T-score < − 2.5), and normal T-score-controls (T-score > − 1). The ratio of mean BMD in relation to mean value for young adults (YA) and in comparison to age (age-matched, AM), was also evaluated. Furthermore, height and weight were measured, and the body mass index (BMI) was calculated. Data on disease manifestation, drug use, age at first and last menstruation, gravidity and parity, and birth weight were collected. The inclusion criteria for the study were as follows: menopause at least 1 year before, no hormone replacement therapy (HRT) or drugs affecting bone mass (selective estrogen receptor modulators SERMs, calcitonin, bisphosphonates, heparin, steroids, thyroid hormones, antiepileptic drugs, GnRH analogues, tibolone). Patients with endocrine and metabolic disorders, hematological diseases, kidney disease, cancers, autoimmune and connective tissue diseases, and after bilateral ovariectomy were excluded from the analysis. Additionally, women who did not smoke were qualified for the study because tobacco smoking may increase the risk of osteoporosis. Moreover, women were not selected in terms of physical activity. The study procedures were approved by the Bioethics Committee of Poznan University of Medical Sciences, Poland (no. 1415/03 (158/06)). The Ethics statement was approved according to Helsinki Declaration. Written informed consent was obtained from all participants.
Analysis of the rs4148323 (UGT1A1*6) and the rs3064744 (UGT1A1*28) variants in the UGT1A1 gene. Blood samples were collected at the Department of Orthopedics and Traumatology, Pomeranian Medical University. The analysis of the UGT1A1 gene variants was conducted at the Department of Stem Cell and Regenerative Medicine, Institute of Natural Fibers and Medicinal Plants, Poznan. Genomic DNA was extracted from peripheral blood using QIAamp Blood Kit (Qiagen GmbH, Hilden, Germany), in accordance with the manufacturer's protocol. DNA concentration was measured using DeNovix DS-11 Spectrophotometer (DeNovix Inc., USA). LightCycler FastStart DNA Master HybProbe (Roche Diagnostics) assay and Light-Cycler®480 instrument for the UGT1A1 gene genotyping were used. Determination of the rs4148323 and the rs3064744 variants of the UGT1A1 gene was performed using LightSNiP (TIBMolbiol), which contained the primers and probes specific for the amplified fragment. PCR was performed in 10 μl reaction mixture according to the manufacturer's protocol under the following conditions: initial denaturation at 95 °C for 10 min, and 35 cycles as follows: denaturation at 95 °C for 10 s, annealing at 60 °C for 10 s, elongation for 15 s at 72 °C, and melting for 30 s at 95 °C and 40 °C for 120 s. The UGT1A1 sequence variants were observed as different melting curves of the PCR products. The UGT1A1 promoter region generally contains six TA repeats, but alleles containing seven repeats lead to reduced gene expression (UGT1A1*28 variant, rs3064744). All genotyping data obtained were double-assessed to minimize error. A duplicate plate was entered to check the quality of genotyp- Statistical analysis. Data analysis was performed using SPSS Statistics 17.0 for Windows. The observed frequencies were compared with the expected frequencies and tested for the Hardy-Weinberg equilibrium. The expected results are presented with 95% confidence intervals (CI). The odds ratio (OR) for the genotypes and the alleles was calculated. Then, the effect of the UGT1A1 genetic variants on T-score, Z-score, L2L4AM (bone mineral density compared with an age-matched), L2L4YA (bone mineral density in young adult), L2L4BMD (bone mineral density between lumbar vertebrae L2-L4), BMI (body mass index), and other clinical parameters was evaluated. Correlation analysis between the genotypes and the clinical parameters was conducted using one-way ANOVA. The p-value of < 0.05 was considered as statistically significant.
All methods were carried out in accordance with relevant guidelines and regulations.

Results
The analysis of the rs4148323 (UGT1A1*6) and the rs3064744 (UGT1A1*28) variants in the UGT1A1 gene was based on different melting curves of the PCR products. Table 1 presents the clinical parameters of postmenopausal women classified into the groups with osteoporosis, osteopenia and controls. The association of the UGT1A1 on the risk of developing osteopenia and osteoporosis was evaluated, which was then correlated with the clinical parameters, including bone parameters. Analyzing the obtained results in the women with osteoporosis, we observed that the body mass was lower in carriers of genotypes 6/6 (60.379 ± 1.265 kg) and 6 Interestingly, women with osteopenia and osteoporosis had lower birth weight as compared to the control group. Analyzing the Z-score values, we also determined that women with osteoporosis and carrying the 6/6 variant had the lowest Z-score value as compared to women with the 6/7 and 7/7 variants (− 1.966 ± 0.242 vs. − 1.577 ± 0.125 and − 1.839 ± 0.233, p = 0.096). For the T-score values in relation to the genotypes for the UGT1A1 variant, no differences were observed between the studied groups. In addition, the effect of the UGT1A1 genetic variants on the duration of a woman's reproductive years was analyzed. No statistically significant differences among the groups were found, because the reproductive years of an average woman were between the ages of 12 and 52 in all groups. The frequency of homozygous 6/6 genotype of the UTG1A1*28 variant (rs3064744) did not differ in the group of women with osteopenia and postmenopausal controls ( Table 2). A slightly higher 6/7 genotype frequency was demonstrated in the control group, whereas the frequency of the 7/7 genotype was higher in the group with osteopenia as compared to controls ( Table 2). In addition, heterozygous 6/7 genotype frequency was slightly lower in the group of women with osteoporosis as compared to controls (46.2% vs. 51.5%, p = 0.146, OR = 0.81, 95% CI 0.57-1.15) ( Table 3). A higher frequency of the 7/7 genotype was observed in the osteoporosis group as well as osteopenia as compared to controls (15.0% vs. 10.7%, p = 0.146, OR = 1.47, 95% CI 0.86-2.56; 18.3% vs. 10.7%, p = 0.049, OR = 1.87, 95% CI 0.93-3.70, respectively). In addition, the odds ratio for the investigated genotypes (6/6, 6/7, 7/7) indicated a higher risk for osteopenia and osteoporosis in women with the 7/7 homozygous genotype (Tables 2 and 3).
The analysis of the frequencies of the GG, GA and AA genotypes in the rs4148323 polymorphism of the UGT1A1 gene showed no statistically significant differences between the investigated groups ( Table 4). The GG genotype was dominant among the women with osteopenia, osteoporosis and controls, while the GA genotype was sporadic in the control group and women with osteopenia. The AA genotype was not found in any of the groups. As far as frequency of the rs4148323 polymorphism and the clinical parameters were concerned, no statistically significant differences were observed (Table 4).

Discussion
In this study, the UGT1A1 genetic variant (UGT1A1*28) was used as a complementary marker of bone mass loss. Early identification and detection of the factors predisposing to the development of osteopenia or osteoporosis allow to implement appropriate prophylaxis and, if necessary, initiate pharmacotherapy. Changes in the parameters such as bone density and bone mass affect predominantly postmenopausal women 1 , which is the consequence of the changes in their hormonal profile. In postmenopausal women, estrogens are synthesized almost exclusively from the androstenedione formed in the adrenal glands, which is converted into estrone in extraglandular tissues 17 . With the increase in body weight and fat content, which is observed in postmenopausal women, the number of estrogen sources increases, while bone turnover decreases, with simultaneous increase in bone mass loss, which seems to be the dominant mechanism of bone tissue changes [18][19][20][21][22][23] .
UGT1A1 is involved in the process of estrogen conjugation and elimination 16 . In the present study, the frequency of the UGT1A1*28 variant among Caucasian women was assessed. The search for a connection and a possible correlation between the variants of the analyzed gene and various diseases has so far been reported in the literature for neonatal jaundice 24 , and tumors 25,26 , among others. In this study, a comparison of the homozygous 6/6 genotype frequency of the UGT1A1*28 variant (rs3064744) between the women with osteopenia and postmenopausal controls revealed no differences. However, the frequency of the 6/7 genotype was higher in the control group, while the 7/7 genotype seemed to be more common in people with osteopenia and osteoporosis as compared to controls. In addition, the heterozygous 6/7 genotype was found to be slightly less common in women with osteoporosis. The frequencies of the GG, GA and AA genotypes were also analyzed, but no statistically www.nature.com/scientificreports/ significant differences between the groups were found. Nevertheless, it can be concluded that among the three analyzed genotypes, the GG genotype was dominant, and the AA genotype was not found. www.nature.com/scientificreports/ When comparing the frequency of the analyzed genetic variants and the clinical parameters, a correlation between the genotypes of UGT1A1*28 and body mass was observed in the group of women with osteoporosis. No statistically significant differences were found for other clinical parameters. It seems, therefore, that the limited number of parameters between which correlation was found is a favorable phenomenon in the context of the diagnostic process and the use of research on the genetic variants on the development of osteoporosis. It eliminates the need to search for connections with other clinical parameters and, consequently, allows for a more accurate prediction of the actual impact of the polymorphisms on the development of osteoporosis.
The race of the study population is a vital issue in the analysis of genetic variants in terms of their frequency [27][28][29] . Since the UGT1A1*28 allele occurs mainly in Caucasian and African American 28,29 populations, while the UGT1A1*6 allele is widely described in Asian 30,31 populations, taking into account the race parameter seems to be well-justified. The study of the UGT1A1 variants is not only important in the context of the metabolism of anticancer drugs, but also, bearing in mind the hormonal associations with osteoporosis, because it seems that the UGT1A1*28 genetic variant may affect the rate of estrogen metabolism 32 . Thus, changes in the nucleotide sequence of the UGT1A1 gene might affect the severity and progression rate of osteoporosis. It is also possible that the described genetic variants may be related to the rate of bone mass loss, thereby affecting the rate of symptom onset. Our results showed the UGT1A1 rs3064744 (UGT1A1*28) genetic variant may affect the risk of developing osteopenia and osteoporosis in postmenopausal women, especially in the presence of homozygous genotypes containing two mutated alleles. Studies by Trontelj et al. showed that patients with the UGT1A1*28 genetic variant may affect bone mineral density in women with osteoporosis taking raloxifene. They indicated that women with the *28/*28 (7/7) genotype had an increased BMD compared to patients with the *1/*1 (6/6) and *1/*28 (6/7) genotypes 33 .
Our findings regarding lack of an association between the UGT1A1*6 variant with osteoporosis are consistent with the observations made in the population of postmenopausal Japanese women. Yokota et al., also did not report a correlation between the UGT1A1 variant and osteoporosis 16 . Nevertheless, it should be taken into account that the absence of statistically significant differences between the compared groups may have resulted from the cross-sectional nature of the study. Therefore, when analyzing all persons included in a given group as a whole, statistical significance may not be observed, however, individual variability should not be forgotten 32,34 . In addition, the results obtained in this study revealed a statistically significant correlation between the analyzed genotypes and body weight. Lower body mass was observed in women with osteoporosis as compared to postmenopausal controls. Low body weight is a predisposing factor for developing osteoporosis, although it remains debatable whether obesity can be a protective factor against bone mass loss 35 , although higher body weight in healthy controls may support the hypothesis. Moreover, taking into account the function of the UGT1A1 protein, one of the main proteins involved in glucuronidation of drugs and other compounds 14 , as well as elimination of estrogens and the consequent reduction of their circulating pool 16 , lower weight may be expected in women with osteoporosis as compared to their postmenopausal healthy peers. The observed values of body mass parameters may be related to the fact that an increase in body weight is accompanied by a corresponding increase in insulin resistance, which attempts to be compensated by elevated secretion of insulin, whose receptors are located on the surface of the osteoblasts. In addition, in women with insulin resistance, increased production of the ovarian hormones and a decreased concentration of sex hormone-binding proteins are observed, which translates Table 2. The frequency of the specific alleles and genotypes of the UGT1A1*28 variant in the group of women with osteopenia and controls. The expected value was calculated in accordance with the Hardy-Weinberg equilibrium (HWE). HWE equilibrium test was used to obtain the exact p value. OR odds ratio.  www.nature.com/scientificreports/ into enhanced bioavailability of the estrogen pool, which in turn increase bone mass 36,37 . Therefore, it can be assumed that the UGT1A1*28 genetic variant may be related to the transcriptional activity of the gene followed www.nature.com/scientificreports/ by the level of protein expression. People with the 6/7 genotype are characterized by a 1/3 reduction in UDPglucuronosyltransferase activity 38 . Molecular analysis performed in this study also showed that the heterozygous 6/7 genotype of the UGT1A1*28 variant was slightly less common in women with osteoporosis (46.2%) as compared to healthy controls (51.5%). It was also observed that the 7/7 genotype was more common in women with osteoporosis and osteopenia as compared to the control group (15.0% vs. 10.7%, p = 0.146, OR = 1.47, 95% CI 0.86-2.56; 18.3% vs. 10.7%, p = 0.049, OR = 1.87, 95% CI 0.93-3.70, respectively). Hence, it seems safe to conclude that, as far as the Polish population is concerned, low frequency of the 6/7 genotype and high frequency of the 7/7 genotype are characteristic for pathological conditions, e.g., Gilbert's Syndrome, osteopenia, osteoporosis 39 . Molecular analysis of osteoporosis is one of the most dynamically developing areas of research related to bone biology. Therefore, studies focusing on the analysis of genetic variants of the "candidate genes" to be used as complementary molecular markers of bone mass disorders are constantly gaining importance.
In conclusion, the results of the present study indicate that the UGT1A1 rs3064744 (UGT1A1*28) genetic variant may affect the risk of developing osteopenia and osteoporosis in postmenopausal women, especially in the presence of homozygous genotypes containing two mutated alleles. The analysis of the frequencies of the GG, GA and AA genotypes of the rs4148323 UGT1A1 gene showed no statistically significant differences between the groups. The UGT1A1 rs4148323 (UGT1A1*6) genetic variant is not directly associated with the development of osteopenia and osteoporosis in postmenopausal Polish women.