Correlation of polymorphous variants (ApaI, TagI, BsmI) of the VDR receptor gene with the vitamin D level and liver fibrosis in children with autoimmune hepatitis

Матеріали та методи. Обстежили 51 дитину з автоімунним гепатитом у 2016–2018 рр. Усім дітям діагноз встановили відповідно до міжнародних рекомендацій із вивчення захворювань печінки (European Association for the Study of Liver (EASL) Clinical Practice Guidelines: Autoimmune hepatitis, 2015). У дітей, яких обстежили, виконали еластографію паренхіми печінки методом зсувної хвилі. У 42 дітей виконали пункційну біопсію печінки з гістологічним дослідженням біоптату. Активність захворювання визначали за допомогою гістологічного індексу активності (ІГА) за Knodell за результатами морфологічного дослідження біоптату печінки та біохімічними показниками. Стадію захворювання оцінювали за гістологічним індексом фіброзу METAVIR і напівкількісно за допомогою еластографії зсувної хвилі паренхіми печінки. У сироватці крові визначили рівень 25(ОН)D. Молекулярно-генетичним методом дослідили поліморфні локуси ApaI, TagI, BsmI гена рецептора вітаміну D (VDR). Оцінили асоціацію поліморфних варіантів гена VDR із рівнем забезпеченості вітаміном D і стадією фіброзу в дітей, які хворі на автоімунний гепатит.


Introduction
Genetically determined predisposition, linkage to the antigens of the human main histocompatibility complex (HLA), alleles 1-B8-DR3, DR4, DR7, DR17, leads to the development of a severe, progressive liver disease such as autoimmune hepatitis in individuals who have lost their immunological tolerance to liver antigens [28]. In recent years, one of the trigger factors in the development and progression of autoimmune hepatitis is recognized as a violation of the vitamin D metabolism in the body as a factor that can accelerate the progression of the pathological process in the liver and negatively affect the effectiveness of therapy [1,2]. Low levels of 25 (OH) D in patients with chronic liver disease (CLD) occur in 90 % of cases, including those in whom the severe deficiency is correlated predominantly with the histological changes, degree of the disease progression and liver fibrosis development as well as the response to treatment [2,3].
(rs2228570), BsmI (rs1544410), ApaI (rs7975232), TaqI (rs731236), EcoRV (rs4516035), Tru9I, Cdx2 (rs11568820). The association of the VDR gene polymorphism with the development of diabetes mellitus, urolithiasis, psoriasis, malignant neoplasms, cardiovascular, neurodegenerative and autoimmune diseases has been established. It is proved that polymorphic genetic variants are associated with different blood levels of 25-(OH)-D, while decreased levels of the vitamin, in turn, are associated with autoimmune diseases development [1,[14][15][16][17]. Studies by Vogel and co-authors suggest a possible association of vitamin D receptor gene polymorphism with an increased risk of developing autoimmune hepatitis. Thus, polymorphic variants of the VDR gene, in particular BsmI and TagI, were associated with autoimmune hepatitis development [18]. The role of FokI polymorphism is associated with an increased risk of autoimmune hepatitis in patients from China and Germany [16], and the CC genotype (rs7975232, ApaI polymorphism) was associated with progression of the disease and liver fibrosis development [19,20]. Reduced expression of the VDR gene in autoimmune hepatitis was correlated with progression of the disease and, accordingly, with an increased risk of cirrhosis and liver fibrosis [19].
Today, the causal link between the vitamin D deficiency and the course of autoimmune hepatitis is poorly understood. In the scientific literature, there are no papers relating to the study of the VDR gene allelic polymorphism in association with the progression of autoimmune hepatitis, depending on the level of vitamin D in children; this has defined the direction of our investigation.

The aim
To study the correlation of the frequency distribution of alleles, genotypes and their combinations for the allelic variants of ApaI, TagI, BsmI of the Vitamin D receptor gene (VDR) with the vitamin D levels and the stage of fibrosis in children with autoimmune hepatitis.

Materials and methods
The study was conducted at the Pediatric Hepatology Center, SI "Institute of Pediatrics, Obstetrics and Gynecology named after academician O. M. Lukianova of NAMS of Ukraine". A total of 51 children with autoimmune hepatitis, aged from 1 to 18 years were examined between 2016 and 2018. The study included children who had not received calcium and vitamin D supplements for 6 months.
In all children, the diagnosis was confirmed in accordance with the International Guidelines for the study of liver diseases (European Association for the Study of Liver (EASL)) Clinical Practice Guidelines: Autoimmune Hepatitis, 2015). For verifying the diagnosis, liver biopsy was performed in 42 patients with further morphological and immune-histochemical studies of the biopsy samples. The disease activity was determined using the histological activity index (HAI) by Knodell based on the results of a morphological study of the liver biopsy and biochemical parameters. The stage of the disease was evaluated by the histological index of fibrosis using the METAVIR scoring system and semi-quantitatively using the share wave elastography of liver parenchyma. In 9 patients, for whom the morphological examination of the liver biopsy was performed at the place of residence, the phase of fibrosis was evaluated by the parameters of the liver parenchyma stiffness by the shear wave elastography. Detection of the liver tissue stiffness was carried out in the SI "Institute of Nuclear Medicine of the National Academy of Medical Sciences of Ukraine", Kyiv using the scanner "Radmyr ULTIMA" on an area of the right intercostal spaces with convex (5 MHz) and linear array (10 MHz) abdominal transducers and a high-frequency transducer for the liver surface. The median index of the obtained measurements characterized the elasticity of the liver parenchyma; the result was expressed in kilopascal (kPa). To interpret the obtained parameters and determine the stage of fibrosis, we used the data of L. Castera and co-authors' investigation, according to which the values of elastography below 5.8 kPa, F1 ≥5.8-≤7.2 kPa corresponded to the level of F0 fibrosis (minimal changes), F2 -≥7.2-≤9.5 kPa (moderate changes), F3 -≥9.5-<12.5 kPa (significant changes) and F4 -more than 12.5 kPa (liver cirrhosis) [31].
In order to verify the diagnosis of vitamin D deficiency and insufficiency, the classification (2011) adopted by the International Institute of Medicine and the Endocrine Practice Guidelines Committee was used. According to this classification, vitamin D deficiency in children and adults is considered as a clinical syndrome due to low levels of 25(OH)D in blood serum (below 20 ng/ml or 50 nmol/l). The serum level of 25(OH)D from 21 ng/ml to 29 ng/ml (i.e., from 50.1 to 74.9 nmol/l) should be considered as vitamin D insufficiency. A normal level of vitamin D equals the blood serum concentration of 25(OH)D above 30 ng/ ml. The serum level of 25(OH)D was measured with an electrochemiluminescent method using the Elecsys 2010 system on the Cobas e 411 analyzer (Roche Diagnostics, Germany). The serum levels of total 25(OH)D that can be determined according to this method are in the range of 7.5-175.0 nmol/l, the variation coefficient is within 3.0 %.
The molecular-genetic study of polymorphous variants of the VDR gene was performed by the polymerase chain reaction (PCR) method. Initially, DNA was extracted from the peripheral blood using the commercial Quick-DNA Miniprep Plus Kit test system (manufactured by Zymo Research, USA). For the determination of the polymorphic variants of the BsmI G/A (rs1544410), TaqI T/C (rs731236) [30] and ApaI A/C (rs7975232) [31] of the VDR gene, the modified protocols with oligo-nucleotide primers and the restriction fragment length polymorphism (RFLP) analysis were used. The investigated genes were amplified using specific primers (produced by Metabion, Germany) and the commercial Dream Taq Green PCR Master Mix (manufactured by Thermo Scientific, USA). The test tubes with the final amplification mixture were transferred to the Flex Cycler BU amplifier (Analytic Jena, Germany) to provide an appropriate temperature regime.
The products of the DNA fragments amplification (amplicons) of the VDR gene were subjected to hydrolytic cleavage by restriction endonuclease BsmI, TaqI and ApaI (produced by Thermo Scientific, USA), respectively. For the restriction analysis, separate mixtures were prepared and transferred into the pre-labeled test tubes, and then amplicons were added. The proportional composition of the components in the template mixture is given in Table 1.
The reaction of the fragments restriction for BsmI G/A (rs1544410) and VDR ApaI A/C (rs7975232) of the VDR gene was carried out according to the manufacturer's recommendations in a solid-state microthermostat at 37 °C for 16 hours. The process was stopped by raising the temperature to 65 °C for 20 minutes. The restriction of TaqI T/C (rs731236) of the VDR gene was incubated at 65 °C for 16 hours without further thermo-inactivation of the enzyme (in accordance with the manufacturer's instructions). The state of the restriction fragments of the VDR gene was analyzed on 3 % agarose gel (agarose produced by Cleaver Scientific, UK) stained with ethidium bromide. Gene Ruler 50 bp DNA Ladder molecular weight marker (manufactured by Thermo Scientific, USA) was added to evaluate the fragments size ( Fig. 1-3).
As shown in Fig. 1, the VDR gene BsmI G/A (rs1544410) amplifiers were subjected to hydrolytic cleavage at an existing restriction site 5'-GAATGCN↓-3', resulting in fragmentation having a molecular weight of 644 bp and 179 bp -the GG genotype. The restriction site disappeared at the nucleotide substitution from G to A, so if the size of the amplified DNA fragments after interaction with the restriction nuclease remained unchanged (823 bp), the AA genotype was recorded. Accordingly, in the heterozygous genotype (GA), all three types of fragments were observed simultaneously: 823, 644 and 179 bp. Fig. 2 shows the electrophoregram of the VDR gene TaqI (rs731236) T/C restriction fragments. The amplicons were subjected to hydrolytic digestion by the TaqI restriction endonuclease at the specific restriction site 5'-T↓CGA-3'. On the amplified fragments of the VDR gene, one of these sites was always present that formed fragments of 496 and 249 bp under the action of TaqI endonuclease. By their presence, the TT genotype was determined. In response to the nucleotide substitution from T to C, an additional restriction site appeared. As a result, in the CC genotype, in addition to 249 bp, restriction fragments with a molecular weight of 295 bp and 201 bp were formed. In the heterozygous genotype (TC), fragments 496 bp, 295 bp, 249 bp and 201 bp, respectively, were observed.
The size of the amplified ApaI (rs7975232) A/C fragment of the VDR gene remained the same (501 bp) under the influence of the ApaI restriction endonuclease in the absence of a nucleotide substitution (the AA genotype), (Fig. 3). The presence of a polymorphic variant was determined by the appearance of the restriction site 5'-GGGCC↓C-3', resulting in formation of fragments 288 bp and 213 bp in size (the CC genotype). The heterozygous genotype (AC) was characterized by the all types of fragments presence: 501 bp, 288 bp, and 213 bp.
The obtained data were statistically analyzed using the Statistica 6.1 software package and SPSS17.0 (SPSS, Inc., Chicago, Illinois, USA). The general statistical analysis included median (Me) and interquartile intervals (UQ-LQ) calculations. Laboratory indices were presented in the form of arithmetic data (mean (M ± m), standard error of the mean (SEM)). For nominal variables, the correlation was calculated using the Pearson (χ2) criterion and Fisher's (two-tailed) criterion; those differences were considered statistically significant, for which a P value was <0.05.

Results
Among the surveyed children (n = 51), girls dominated and accounted for 61.0 %, and boys made up 39.0 %. Such a distribution corresponds to the clinical features of the disease, which is more often recorded among girls. Higher prevalence among females is associated with the HLA antigens system: HLA-1-B8-DR3 or DR4 [24][25][26]. Autoimmune hepatitis (AIH) belongs to orphan diseases with a small frequency in the general population that has caused a small number of children included in the study. While Verma et al. [26] believe that the prevalence of AIH in the world is between 2 and 17 per 100.000 children, and it may vary within ethnic groups. The disease is more often diagnosed at the age of 10-30 years. The average age of patients at the time of our study was 11 [8; 15] years, corresponding to the literature data [28].The average age for boys was 10 [8; 14] years, for girls 11.5 [9; 15]. 62.7% (n = 32) of patients were in the age group of 11-18 years. We did not find a significant difference between the frequency distributions of the of alleles and genotypes of the ApaI, TagI and BsmI polymorphic variants of the VDR gene in children with AIH by gender and age (P > 0.05), so further analysis of the studied gene variations effects was performed in the general group of patients. The results of the identified peculiarities of distribution in the general group are presented in Table 2. Distribution of genotypes based on the VDR ApaI, TagI, BsmI genetic variants was analyzed according to the Hardy-Weinberg Law. The concordance with the Hardy-Weinberg law was found for the ApaI and TagI genotype variants but not for the BsmI variant, that may be associated with a higher risk of multifactorial disease development such as autoimmune hepatitis.
According to our data, 72.0 % (n = 36) of the subjects had advanced liver fibrosis (F 3-4 METAVIR), and 34.0 % of them (n = 17) had signs of liver cirrhosis. Distribution of children by the stage of fibrosis is shown in Fig. 4.
The mean concentration of vitamin D in the examined children was 16.3 [10.9; 22.0] ng/ml corresponding to vitamin D deficiency. The vitamin D levels in children with AIH, depending on the stage of fibrosis, are given in Table 3. 12.0 % (n = 6) of children had optimal levels of vitamin D, deficiency was diagnosed in 15.5 % (n = 8), and deficiency was found in 72.5 % (n = 37) of the subjects. For computational convenience, we combined the subgroups of children with the stage of fibrosis F1 and F2, as well as F3 and F4.
According to our data, vitamin D levels in children depended on the stage of liver fibrosis. Patients with advanced fibrosis (F3-4 METAVIR) were significantly more likely to have vitamin D deficiency (χ 2 = 5.21; P = 0.022) compared to children with F1-2 fibrosis by the METAVIR score ( Table 5).
The evaluation of the AIH genetic risk was not the goal of our research. However we studied the effects of genetic variants on the complicated course of the disease associated with liver cirrhosis development by intra-group comparisons. We calculated genetic risk models including additive, recessive, dominant, multiplicative and co-dominant for complicated AIH development. The study of the association between genotypes of ApaI, BsmI and TagI polymorphic variants of the VDR gene with the stage of fibrosis in children with AIH demonstrated that children with advanced fibrosis (F3-4 METAVIR) were significantly more likely to have the CC genotype of the polymorphic variant TagI of the VDR gene (χ 2 = 3.953; P < 0.05) compared to children with the fibrosis stage F1-2 by the METAVIR score ( Table 4). The highest predictive ability was shown for recessive inheritance model.
In children of the general group, the association between the serum levels of vitamin D and the VDR gene polymorphism was investigated. There was no significant difference in the levels of vitamin D depending on the geno- Table 2. Frequency distribution of the alleles and genotypes of the AраI, TagI and BsmI polymorphisms of the VDR gene in children with autoimmune hepatitis (abs., %)    Table 5).

ApaI (A/C) TagI (T/C) BsmI (G/A)
According to the results of our study, the following combinations of the genotypes in the three allelic variants of the gene were not found among the examined patients: The results demonstrated that children with the AA/TC/ AA, AC/TT/GG and CC/TT/GG genotype combinations were significantly more likely to have higher vitamin D serum levels than children with the AC/TC/GA genotype combination of the polymorphic variants ApaI, TagI, BsmI of the VDR gene (P < 0.05). In general, all the examined children had varying degrees of vitamin D deficiency. Children with the AA/TC/AA genotype combination had the highest level of 25(OH)D -19.84 ± 8.29 ng/ml.
The correlation analysis of the VDR gene allelic variant combinations with the fibrosis stage showed that children with the genotype AA/CC/AA combination were significantly more likely to have advanced fibrosis F 3-4 by the METAVIR score (χ 2 = 3.953; P <0.05) ( Table 7).

Discussion
While searching for the scientific research results in PubMed, EMBASE and Cochrane Library we did not find any studies on the association between polymorphic variants of the VDR gene with the risk of AIH development or their peculiarities in the pediatric population, only some studies conducted among adults were revealed. Most studies were devoted to polymorphisms of the VDR gene in patients with primary biliary cirrhosis [20][21][22][23]29]. Thus, M. Vogel and co-authors, analyzing the effect of the VDR gene polymorphisms in AIH and primary biliary cirrhosis, revealed the association between BsmI and TaqI polymorphisms of the VDR and primary biliary cirrhosis in the German population, as well as the association between FokI and TaqI polymorphisms and AIH. The authors established a significant correlation between the GA (χ 2   19.230 ± 7.057* 12.442 ± 5.515 18.530 ± 7.805** *: the difference is significant (P < 0.05) between the GG and GA genotypes of the BsmI polymorphic variant; **: the difference is significant (P < 0.05) between the GA and AA genotypes of the BsmI polymorphic variant.       [21]. The information retrieval conducted by us has validated our study on the possibility of the VDR gene analysis application in the pediatric hepatology as a prognostic marker for the unfavorable course of AIH with progression to liver cirrhosis.
Analysis of the FokI polymorphic locus in our AIH patients was not performed. The distribution of allelic and genotype frequencies found in our study were similar to those found in patients from Germany.

Conclusions
Thus, according to our data, deficiency of vitamin D was found in 72.5 % of patients with AIH. The level of vitamin D depended on the stage of liver fibrosis. Children with advanced fibrosis F3-4 by the METAVIR score were significantly more likely to have vitamin D deficiency.
Vitamin D deficiency was associated with the genetic peculiarities of the patients: the GA genotype of the BsmI polymorphic variant presence and the AC/TC/GA genotype combination of the ApaI, TagI and BsmI polymorphic variants of the VDR gene.
The CC genotype of the TagI polymorphic variant and the AA/CC/AA genotype combination of the ApaI, TagI and BsmI polymorphic variants of the VDR gene were associated with advanced fibrosis F3-4 by the METAVIR score in children with AIH.