Expression of genes involved in retinoic acid biosynthesis in human ...

11 downloads 131 Views 590KB Size Report
Expression of genes involved in retinoic acid biosynthesis in human gastric cancer. Authors; Authors and affiliations. E. S. Kropotova; O. L. Zinov'eva; A. F.
ISSN 00268933, Molecular Biology, 2013, Vol. 47, No. 2, pp. 280–292. © Pleiades Publishing, Inc., 2013. Original Russian Text © E.S. Kropotova, O.L. Zinov’eva, A.F. Zyryanova, E.L. Choinzonov, S.G. Afanas’ev, N.V. Cherdyntseva, S.F. Beresten, N.Yu. Oparina, T.D. Mashkova, 2013, published in Molekulyarnaya Biologiya, 2013, Vol. 47, No. 2, pp. 317–330.

CELL MOLECULAR BIOLOGY UDC 577.21

Expression of Genes Involved in Retinoic Acid Biosynthesis in Human Gastric Cancer E. S. Kropotovaa, O. L. Zinov’evaa, A. F. Zyryanovaa, E. L. Choinzonovb, S. G. Afanas’evb, N. V. Cherdyntsevab, S. F. Beresten’a, N. Yu. Oparinaa, and T. D. Mashkovaa a

Engelhardt Institute of Molecular Biology, Russian Academy of Sciences, Moscow, 119991 Russia; email: [email protected] b Tomsk Cancer Research Institute, Siberian Branch, Russian Academy of Medical Sciences, Tomsk, 634009 Russia Received October 2, 2012; in final form, November 1, 2012

Abstract—Alltransretinoic acid (ATRA) is the main biologically active metabolite of retinol (vitamin A) that is required for the regulation of processes such as embryogenesis, tissue differentiation, proliferation, and others. Multiple alcohol, retinol, and retinaldehyde dehydrogenases (ADHs, RDHs, and RALDHs), as well as aldoketo reductases (AKRs) catalyze the biosynthesis of retinoic acid in humans. For many normal and neoplastic tissues, the key ATRAsynthesizing enzymes remain unknown. We identified ATRAgenerating genes that are expressed in normal and malignant gastric tissues using the transcriptomic database analysis. Quantitative changes in the expression levels of these genes in gastric cancer were determined by semiquan titative RTPCR and realtime PCR. Significant decreases in the mRNA levels of genes that encode the enzymes that catalyze the reversible oxidation/reduction of retinol and retinaldehyde (ADH4, ADH1B, ADH1C, RDHL, AKR1B10, AKR1B1, and RDH12), as well as the oxidation of retinaldehyde (RALDH1) were revealed in most tumor samples. A sharp reduction in the expression levels of genes encoding the key enzymes that convert retinol and retinaldehyde to retinoic acid could lead to a significant decrease in the content of ATRA, the transcriptional regulator of many genes, which can in turn lead to the dysregulation of cell prolif eration/differentiation and initiate the development of cancer. DOI: 10.1134/S0026893313020076 Keywords: alltransretinoic acid, biosynthesis of retinoic acid, retinol, retinal(dehyde), gene expression, gas tric cancer

Natural and synthetic retinoids (derivates of vita min A (retinol)) are important for the prevention and therapy of several types of cancer. Major natural bio logically active retinoid, alltransretinoic acid (ATRA), is essential for embryogenesis, proliferation, tissue differentiation, and carcinogenesis. The biolog ical effect of ATRA is based on its interaction with nuclear retinoid receptors (RAR and RXR), which regulates the expression of genes that are responsible for cell differentiation, proliferation, and apoptosis. Over 500 target genes regulated by retinoic receptors have been identified [1]. The source of retinoic acid in human body is repre sented mostly by retinoic esters in fat and plant caroti noids (mostly βcarotene) from food. Retinolbind ing proteins transport retinol from its major depot in Abbreviations: ADH, alcohol dehydrogenase; AKR, aldoketo reductase; ATRA, alltransretinoic acid; CRBP1, Cellular retinolbinding protein1; CYP26A1, B1, C1 are proteins of CYP26 family (cytochrome P450dependent monooxigenases); RALDH, retinaldehyde dehydrogenase; RDH, retinol dehydro genase; RefExA, Reference database for gene Expression Analy sis; PCR, polymerase chain reaction; RTPCR, reverse tran scription PCR.

liver to different tissues, where it is transformed into ATRA [2]. ATRA biosynthesis is a complex process that impli cates multiple enzymes [3, 4] and includes the revers ible oxidation of alltransretinol into alltransreti naldehyde (alltransretinal) and its subsequent oxida tion into retinoic acid. ATRAsynthesizing enzymes vary in their catalytic activity, as well as tissue and substratespecificity. Retinol transformation in retinal implicates cytosolic alcohol dehydrogenases (ADH), as well as microsomal NAD+dependent retinol dehy drogenases (RDH). ADH can oxidize alcohols, including ethanol and retinol, and reduce various aldehydes, but maximal activity is observed in the case of oxidation. Seven ADHencoding genes located within a short region of the 4q21–23 chromosome have been identified in humans. They are divided into five classes (I–V), i.e., ADH1 (class I, ADH1A, ADH1B, and ADH1C), ADH2 (class II), ADH3 (class III), ADH4 (class IV), and ADH6 (class V) [5, 6]. ADH1, ADH3, and ADH4 can synthesize ATRA in vivo [4, 7, 8]. ADH1, ADH2, and, in particular, ADH4 exhibit pronounced retinoloxidizing activity [9, 10], whereas ADH3 exhibits rather low activity [8].

280

EXPRESSION OF GENES INVOLVED IN RETINOIC ACID BIOSYNTHESIS

The most active NAD+dependent retinoloxidiz ing enzymes in humans [10–12] include alltransret inolspecific RDH10 [13] and RoDH4 [9]. The reduction of retinal to retinol requires the activity of NADP+dependent RDH, aldoketo reductases (AKR), and ADH, although the efficiency of the latter is less pronounced compared to retinol oxidation reac tion. The most active NADP+dependent RDH are represented by RDH12 and RDH11 [10, 14]. AKRs have only recently been considered as highly efficient retinalreducing enzymes. The most interesting AKRs include two human AKRs from the AKR1B family of aldose reductases, i.e., AKR1B1 and AKR1B10 [9, 15]. In humans, the oxidation of retinal to ATRA is catalyzed by retinoidactive RALDH1 (ALDH1A1), RALDH2 (ALDH1A2), and RALDH3 (ALDH1A3). In mice, ATRA biosynthesis implicates orthologous enzymes Raldh1, 2, and 3 [4]. The amount of ATRA is regulated both by ATRAsynthesizing enzymes and humanspecific ATRAdegrading cytochromes CYP26A1, CYP26B1, and CYP26C1 [16]. Disturbed retinoltoATRA conversion was observed in several types of cancer, including colorec tal [17], prostate [18], breast [19], ovarian [20], and gastric [21] cancer. Gastric cancer is one of the most frequent types of cancer in Russia and other countries [22, 23]. ATRA, which is synthesized from alltrans retinol [24], is required to maintain normal morphol ogy of gastric mucosa [21]. The risk of developing gas tric cancer is increased in patients with elevated levels of retinol precursors (α and β carotenes) in blood [25], whereas a high level of provitamin and vitamin A can markedly decrease the risk of the disease [26]. ATRA biosynthesis in various tissues includes dif ferent enzyme sets, and their precise content has not been elucidated for many tissues, including normal and tumor gastric mucosa. This work is aimed at iden tifying the genes that encode enzymes that are active in gastric tissues using bioinformatics analysis, as well as studying the expression of these genes in normal and malignant gastric tissues using semiquantitative RT PCR and realtime PCR. MATERIALS AND METHODS Clinical Samples Thirtyeight pairs (normal and tumor tissues) of postsurgery samples obtained from patients with gas tric cancer (25 males (65.8%) and 13 females (34.2%) aged 27–82, with average age 56) were analyzed. Patients over 50 years old constituted 68.4%. Histolog ically normal tissues adjacent to the tumor were con sidered as normal. The samples were obtained from the collection of the Hospital of Tomsk Cancer Research Institute, Siberian Branch of Russian Acad emy of Medical Sciences. Immediately after the oper ation, the samples were frozen at –70°C and stored at –70°C. MOLECULAR BIOLOGY

Vol. 47

No. 2

2013

281

All tumor samples were characterized according to the international clinicalmorphological tumor classi fication system (TNM). Five patients had stageI tumors, ten patients had stageII tumors, fifteen had stageIII tumors, and eight patients had stageIV tumors. Twentytwo patients had metastases in regional lymph nodes. Moderate differentiation was observed in 11 patients, low differentiation (including mucosa) was observed in 24 patients, mixed differenti ation was observed in 3 patients. Samples containing over 80% of tumor cells were selected using histologi cal analysis. The samples were collected in accordance with the protocol validated by a special committee. The work was performed in keeping with the prin ciples of patients' voluntary involvement and confi dentiality in accordance with the Fundamentals of the Russian Federation Legislation on Health Care. Approval by the Ethics Committee attached to the Cancer Research Institute of the Siberian Division, Russian Academy of Medical Sciences and the patient’s informed consent were obtained. Analysis of Transcriptome Databases Based on Microarray Hybridization (RefExA and ONCOMINE) The level of expression of candidate genes involved in ATRA biosynthesis in normal gastric mucosa was estimated using the standardized RefExA database (Reference Database for Gene Expression Analysis) (http://157.82.78.238/refexa/main_search.jsp) that contains the results of the hybridization of cDNA from normal tissues obtained on Affymetrix microarrays (HGU133A and HG133B). The differential expres sion of genes in the case of gastric adenocarcinoma was estimated using ONCOMINE databases (http:// www.oncomine.org). RNA Purification from Tissue Samples and Synthesis of FirstStrand cDNA Total RNA was isolated from samples of normal and tumor gastric tissues frozen and ground in liquid nitrogen. The samples were homogenized using MicroDismembrator U (Sartorius). RNA was puri fied using RNeasy Mini kit (Qiagen) according to the manufacturer’s recommendations and treated with DNase I. The quality of RNA was verified with elec trophoresis in 1% agarose gel with ethidium bromide. The RNA concentration was determined using an ND 1000 spectrophotometer (NanoDrop Technolo gies, Inc). The RNA template was used for cDNA syn thesis. Firststrand cDNA was synthesized using 1 μg of total RNA treated with reverse transcriptase Super ScriptTM III (Invitrogen) and random hexanucleotide primers (Syntol, Moscow). Semiquantitative RTPCR. The level of mRNA, encoding ATRAsynthesizing enzymes was prelimi narily determined with RTPCR. The primers for dif

282

KROPOTOVA et al.

ferent exons were designed so that at least one of the primers overlapped the exon boundary. In this case, the contamination of genomic DNA did not interfere with the reaction. The gene that encodes βactin (ACTB) was used as a reference control, since its expression was stable in both normal and tumor gastric mucosa. The reaction mixture for amplification (25 μL) contained 67 mM TrisHCl, pH 8.8, 16.6 mM (NH4)2SO4, 0.1% Tween20, 2.5 mM MgCl2, 0.2 mM of each dNTP, 0.1–0.01 μg cDNA, 0.4 μM of each primer, 2 a.u. SmarTaqDNApolymerase (Dialat LTD, Moscow). Amplification was performed under the following conditions: primary denaturation 94°C for 3 min; 25–40 cycles (depending on the gene) at 94°C for30 s, 60°C for 30 s, and 72°C for 30 s; one cycle at 72°C for 5 min. PCR was performed using MasterCycler (Eppendorf). Amplification products were analyzed in 1.8% agarose gel with 0.5 μg/mL ethidium bromide. All amplification reactions were repeated three times. As a result, the optimal condi tions of RTPCR, which ensured the linear depen dence of the concentration of the PCR product, and the number of cycles were selected. After the electrophoresis of PCR products, the fluorescence intensity of the bands was quantified using GeneProfiler software for image densitometry (http://www/scanalytics.com) normalized to the ref erence gene and expressed as the relative nor mal/tumor (N/T) intensity. RealTime PCR The quantification of mRNA that encodes ATRA associated genes in 38 paired samples (normal and tumor tissues) was performed using realtime PCR on an Applied Biosystems 7500 Fast RealTime PCR System amplificator. The ACTB gene was used as a ref erence control. The reactions (25 μL) were carried out in a standard MicroAmp™ Fast Optical 96Well Reac tion Plates and repeated three times. The reaction mixture contained intercalating dye EvaGreen (Biotium, Inc.), primers, template cDNA, ROX refer ence dye, thermostable reaction mixture RialityTM [28] with TaqDNApolymerase, dNTP, MgCl2, and PCR buffer. EvaGreen, a newgeneration intercalat ing dye, exhibits a less pronounced inhibitory effect on the PCR reaction compared to widely used SYBR Green dyes; it is more stable and has no mutagenic effect. All genes except RDH10 were amplified with the same primers as were used for RTPCR, i.e., RDH10, F2 and RDH10, R2. Each plate contained three negative controls (samples without cDNA). The reaction was held under the following conditions: 95°C for 10 min, 40 cycles at 95°C for 15 s, and 60°С for 1 min. The specificity of PCR products was verified by the melting curve and electrophoresis in 1.8% aga rose gel. The relative quantitative analysis was per formed using LinRegPCR (12.0) software.

Statistical Analysis The statistical significance of the alteration of tar get mRNA level was estimated by the nonparametric Wilcoxon test. The results were considered to be sig nificant at P < 0.05. RESULTS Identification of Candidate Genes Involved in ATRA Biosynthesis in Normal and Tumor Gastric Tissues Using Analysis of Transcriptomic Databases RefExA and ONCOMINE An analysis of the published data enabled us to select the genes that were involved in ATRA biosyn thesis and exhibited tissuespecific pattern of expres sion. The primary set of 32 genes included seven ADH genes (ADH1A, B, and C, ADH2, ADH3, ADH4, ADH6), 16 RDH (RDH10, RDHL, RDH5, RoDH, RoDH4, RDHE2, XOR, RDH8, RDH1114, RDH17, DHRS4, DHRS4L2, SDRO]), two AKR genes (AKR1B1 and B10), three RALDH (RALDH1, 2 and 3), three CYP26 (CYP26A1, B1 and C1), and the gene that encodes CRBP1. To identify the candidate genes involved in ATRA biosynthesis in normal and tumor gastric mucosa, we considered the results of a microarray analysis represented in RefExA and ONCOMINE transcriptomic databases. In normal gastric mucosa, we identified the mRNA of only 11 ATRAsynthesizing proteins that were interesting for further study (RefExA), eight of which were differ entially expressed in normal and tumor samples (ONCOMINE) (Table 1). In normal gastric mucosa retinoloxidizing stage of ATRA biosynthesis is associated mostly with the expression of ADH1B, ADH1C, ADH3, and ADH4 genes. The content of ADH1B, ADH1C, and ADH4 mRNAs in tumor samples is strongly decreased, whereas the level of ADH3 mRNA is unaltered. According to RefExA data, the levels of ADH1A and ADH6 are relatively lower in normal gastric mucosa, while the level of ADH2 mRNA is much lower. In normal gastric mucosa, the level of mRNAs encoding retinoloxidizing RDH10 and RDHL was rather high. In tumor samples, the expression of RDHL was reduced, while the expression of RDH10 was insignificantly increased (approximately twofold). In normal gastric tissues, mRNAs encoding reti nalreducing enzymes were represented by RDH11, RDH12, AKR1B1, and AKR1B10. Maximal hybridiza tion signal corresponded to AKR1B10 gene. The level RDH12 and AKR1B10 mRNAs was dramatically decreased in tumor samples, while the level of RDH11 and AKR1B1 remained unaltered. The high level of mRNAs that encode retinaloxidizing enzymes was only observed in the case of RALDH1 and, in tumor tissues, its level was decreased. Simultaneous analysis of two different transrip tomic databases enabled us to identify a set of candi MOLECULAR BIOLOGY

Vol. 47

No. 2

2013

EXPRESSION OF GENES INVOLVED IN RETINOIC ACID BIOSYNTHESIS

283

Table 1. Altered expression of genes involved in ATRA biosynthesis in gastric cancer (bioinformatics analysis) Gene symbol commonly used* ADH1A (class I) ADH1B (class I) ADH1C (class I) ADH2 (class II) ADH3 (class III) ADH4 (class IV) RDH10 RDHL RDH5 RoDH4 RDH11 RDH12 AKR1B1 AKR1B10 RALDH1 RALDH2 RALDH3 CYP26A1 CRBP1 Notes:

official

Level of gene expression in normal tissues (RefExA), a.u.**

ADH1A ADH1B ADH1C ADH4 ADH5 ADH7 RDH10 DHRS9 RDH5 RDH16 RDH11 RDH12 AKR1B1 AKR1B10 ALDH1A1 ALDH1A2 ALDH1A3 CYP26A1 RBP1

44 296 312 4 303 136 170 65 15 7 54 139 404 3021 1352 10 65 34 70

Alteration of gene expression in tumor (ONCOMINE)*** –2.39↓ (0.01) –4.34↓ (7.40E18) –8.7↓ (1.80E06) –2.15↓ (0.01) Unaltered –5.9↓ (8.05E12) 2.19↑ (5.70E04) –3.99↓ (1.26E04) Unaltered Unaltered Unaltered –7.8↓ (1.29E11) Unaltered –8.15↓ (4.58E08) –2.36↓ (4.43E05) Unaltered 2.34↑ (3.10E05) Unaltered Unaltered

* Most common symbols are used. ** Level of expression was determined with regard to the intensity of hybridization signal (a.u.). *** Ratio of changes in expression in normal–tumor pairs was measured, ↓ indicates decreased level of expression, ↑ indicates increased level of expression. Values of p are shown in brackets.

date genes involved in ATRA biosynthesis in normal and tumor gastric tissues, and define the expression pattern of these genes in gastric cancer. The results of bioinformatics transcriptomic analysis were verified by semiquantitative RTPCR and realtime PCR. For experimental verification, we selected genes that were expressed in normal and/or tumor gastric tissues at high and medium levels (ADH1B, ADH1C, ADH3, ADH4, RDH10, RHDL, RDH11, RDH12, AKR1B1, AKR1B10, and RALDH1). We also included genes that were expressed at a low level or were not expressed, but rather encoded highly active retinoidassociated enzymes (ADH1A, ADH2, RoDH4, RALDH2, RALDH3, and CYP26A1) and proteins associated with gastrointestinal cancer (RDH5). We also analyzed the expression of the CRBP1 gene, which is implicated in ATRA biosynthesis. Altered Expression of Genes That Encode RetinolOxidizing Enzymes The level of mRNA of 19 candidate genes (ADH1A, ADH1B, ADH1C, ADH2, ADH3, ADH4, RDH10, MOLECULAR BIOLOGY

Vol. 47

No. 2

2013

RoDH4, RDHL, RDH5, RDH11, RDH12, AKR1B1, AKR1B10, RALDH1, RALDH2, RALDH3, CYP26A1, and CRBP1), which are involved in ATRA biosynthe sis, was determined by semiquantitative RTPCR in 30 paired samples (normal/tumor) obtained from patients with gastric cancer. Nucleotide sequences of primers for all genes are represented in Table 2. Ret inoloxidizing genes expressed in normal gastric mucosa were represented by ADH1B, ADH1C, ADH4, ADH3, RDH10, RDHL, and RDH5 (Fig. 1). The level of corresponding mRNAs was decreased in the raw ADH1B = ADH1C > RDH10 ≥ ADH4 > ADH3 > RDHL > RDH5. mRNAs of ADH1A, ADH2, and RoDH4 were not found in either normal or tumor gas tric tissues. The results of RTPCR analysis of altered gene expression in gastric tumor samples are represented in Fig. 1 and Table 3. The level of RDH10, ADH3, and RDH5 mRNA in most tumor samples was almost the same as in normal tissues (77, 67, and 53%, respec tively). The level of RDHL mRNA was decreased in 60% of tumor samples. The most prominent alteration of the expression was observed in the case of ADH4,

284

KROPOTOVA et al.

Table 2. Primers for RTPCR and realtime PCR Gene

Primer nucleotide sequence (5' → 3')

PCRproduct, bp

NCBI Refseq mRNA ID

ADH1A

F GGGTGACTACAGTCAAACCAGGTGA R GGTCCCCTGAGGATTGCTTACATCG

131

NM_000667.3

ADH1B

F TGGCACCAGCACCTTCTCCC R GAGCCTGGGGTGACCTTGGC

156

NM_000668.4

ADH1C

F AGACAGAATCAATATGAGCACAGCAGG R ACCAGGTTGCCACTAACCACATGC

190

NM_000669.3

ADH2

F TGCAATCAACAATGCCAAGGTCACCC R CAGTCAGTGGCTCCCAGGGC

174

NM_000670.3

ADH3

F GCTGGTGCTTCCCGGATCATTGG R GCCTCAAGTGCTGCTCTCATGACC

203

NM_000671.3

ADH4

F GACCAGCGTGGTTGTAGGAGTTCC R TCCATGTGCGTCCAGTGAAGAGC

83

RDH10

F1 GCCATGCACACTTCTGGACCACT R1 ACTCCGGCAGTACTGAACAATCCCA F2 AATAATGCTGGTGTGGTCTCTGG R2 CTGGCACAGTAATCCTCAACTCC

112 (1)

NM_001166504.1 NM_172037.3

209 (2)

RDHL

F TGGAAACTTGGCAGCCAGAA R CCAGAGACCTTTCTCCCCAA

199

NM_001142271.1

RDH5

F TTCTCTGACAGCCTGAGGCG R TGCGCTGTTGCATTTTCAGG

202

NM_001199771.1

RoDH4

F GACCGGTCCAGTCCAGAGGTC R TAGCGAGTACGGGGGTGGCAG

161

NM_003708.3

RDH11

F CACTTGGGTCACTTCCTCCTAAC R AACGCCAGAGCCTTTTAGTCTC

228

NM_016026.3

RDH12

F TGGAACGATGCTGGTCACCTTGG R GATCACCACTACCTTGCCAGGAAGC

139

NM_152443.2

AKR1B1

F TGGATGAAGGGCTGGTGAA R GGTGGCACTCAATCTGGTTA

121

NM_001628.2

AKR1B10 F GGACCTGTTCATCGTCAGCAA R CCCCAGACTTGAATCCCTGTG

145

NM_020299.4

RALDH1 F AACTCCTCTCACTGCTCTCCACG R GTCACCCTCTTCAGATTGCTTTTCC

210

NM_000689.4

RALDH2 F CAAGATGTCTGGAAATGGGAGAG R CTTTAAGTAAGGACCGTGGCTCA

210

NM_001206897.1

RALDH3 F TCAACTGCTACAACGCCCTCTAT R CGCCGTCCGATGTTTGAG

184

NM_000693.2

CYP26A1 F GCTGTACCGGGGCATGAAGGC R CAGCCGCTCTCCCCTCTCC

130

NM_057157.2

CRBP1

F TAGAGATGAGAGTGGAAGGTGTGGT R GGGGTGGCTGGACATTTTTG

153

NM_002899.3

ACTB

F CCTTCCTGGGCATGGAGTC R CTTGATCTTCATTGTGCTGGGT

191

NM_001101.3

Note: Primers RDH10, F1 and RDH10, R1 were used for RTPCR, primers RDH10, F2 and RDH10, R2 were used for realtime PCR. MOLECULAR BIOLOGY

Vol. 47

No. 2

2013

EXPRESSION OF GENES INVOLVED IN RETINOIC ACID BIOSYNTHESIS

1

2

3

Normal 4 5

6

7

1

2

3

Tumor 4 5

6

285

7

ADH4 ADH1B ADH1C ADH3 RDHL RDH5 RDH10 RDH11 RDH12 AKR1B1 AKR1B10 RALDH1 RALDH2 RALDH3 CYP26A1 CRBP1 ACTB

Fig. 1. RTPCR analysis of mRNA of genes involved in ATRA biosynthesis in gastric cancer compared to the adjacent normal tissues. RALDH1 and AKR1B10 after 28 cycles; ADH1C, ADH1B after 30 cycles; ADH4, RDH10, RDH11, RDH12 after 32 cycles, ADH4 and AKR1B1 after 33 cycles; RDHL and RALDH3 after 34 cycles; RDH5 and CRBP1 after 35 cycles; RALDH2 after 36 cycles, and CYP26A1 after 39 cycles. βActin gene (ACTB) was used as an internal reference control (28 cycles). Electro phoretic separation of each amplification product was performed in 1.8% agarose gel.

ADH1B, and ADH1C genes. The results of realtime PCR performed on an extended panel of samples (38 pairs) revealed dramatic (31, 15 and 23fold) decrease in the mRNA level in 84, 53, and 58% of tumor samples, respectively (Table 3), and 92% of samples exhibited the decreased expression of at least one of these genes (Fig. 2a). It should be noted that the mRNA content of these genes was significantly decreased, even at early stages of gastric cancer. Therefore, in most tumor samples, we detected dra matic decrease of mRNA of genes encoding retinol oxidating enzymes that are actively expressed in nor mal gastric tissues. Altered Expression of Genes That Encode RetinalReducing Enzymes Retinal reduction to retinol implicates the activity of NADP+dependent RDH (RDH12 and RDH11), MOLECULAR BIOLOGY

Vol. 47

No. 2

2013

AKR (AKR1В1 and AKR1В10), as well as ADH (ADH4, ADH1B, and ADH1C), although their reducing activity is less than oxidative. The level of mRNAs that encodes these enzymes was decreased in the raw AKR1B10 > ADH1B = ADH1C > ADH4 = RDH12 = RDH11 > AKR1B1. The level of AKR1B1 mRNA was reduced in 53% of tumor samples, whereas the level of RDH11 mRNA was only decreased in 33% and unaltered in 44% of samples. The most prominent alteration was observed for AKR1B10 and RDH12 mRNA. The results of realtime PCR performed on an extended panel of samples revealed a dramatic (20, and 23fold) decrease in mRNA in 74 and 90% of tumor samples, respectively (Table 3 and Fig. 2), and the expression of the corre sponding genes was significantly decreased, even at early stages of gastric cancer. It should be noted that 92% of samples exhibited the decreased expression of at least one of these two genes (AKR1B10, RDH12),

286

KROPOTOVA et al.

Table 3. Altered expression of genes involved in ATRA biosynthesis in gastric cancer (RTPCR and realtime PCR data) Altered mRNA level in gastric cancer Gene decrease

increase 3% (1/38) 16.7

unaltered

ADH4

84% (32/38) 31 (2.3–1790) p < 0.0001

13% (5/38)

ADH1B

53% (20/38) 15 (2.1–1538) p = 0.0007

10% (4/38) 6.1 (3.0–15.3)

37% (14/38)

ADH1C

58% (22/38) 23 (2.1–744) p = 0.0007

18% (7/38) 5.3 (2.2–12.7)

24% (9/38)

ADH3

23% (7/30) 3.5 (2.1–8.0)

13% (3/30) 4.8 (2.1–15.0)

64% (20/30)

RDHL

60% (18/30) 4.9 (2.5–10.0) p = 0.0008

20% (6/30) 4.6 (4.0–5.5)

20% (6/30)

RDH5

40% (12/30) 11.9 (2.5–10.0)

7% (2/30) 9.2 (6.4–12.0)

53% (16/30)

RDH10

17% (5/30) 3.7 (3.0–4.5)

7% (2/30) 3.2 (2.1–5.2)

76% (23/30)

RDH11

33% (10/30) 3.3 (2.1–5.5)

23% (7/30) 3.6 (2.1–10.2)

44% (13/30)

RDH12

84% (32/38) 37.8 (2.4–1999) p < 0.0001

8% (3/38) 2.2 (2.1–2.4)

8% (3/38)

AKR1B1

53% (16/30) 4.6 (2.5–10.1) p = 0.0007

17% (5/30) 2.9 (2.1–4.0)

30% (9/30)

AKR1B10

74% (28/38) 20.4 (2.1–393) p < 0.0001

13% (5/38) 6.1(2.1–15)

13% (5/38)

RALDH1

57% (17/30) 4.8 (2.5–10.0) p = 0.0001

10% (3/30) 3.1 (2.5–4.0)

33% (10/30)

RALDH2

20% (6/30) 6.1 (3.0–10.0)

7% (2/30) 2.6 (2.2–3.0)

73% (22/30)

RALDH3

30% (9/30) 4.5 (2.1–4.3)

20% (6/30) 3.5 (2.5–5.5)

50% (15/30)

CYP26A1

13% (4/30) 2.7 (2.1–4.5)

53% (16/30) 2.6 (2.1–4.0) p < 0.0001

33% (10/30)

CRBP1

20% (6/30) 2.3 (2.1–5.5)

23% (7/30) 2.3 (3.5–10.0)

57% (17/30)

Note: The frequency of changes was estimated in percentages. The number of samples with altered or unaltered mRNA/the total number of samples are shown in brackets. Mean geometrical value of mRNA level change (times) is shown below: interval of measurements is shown in brackets. p value was calculated in the case of dominant alteration in mRNA level (shaded).

and 100% of samples exhibited the decreased expres sion of at least one out of three of the most active reti nalreducing enzymes (AKR1B10, RDH12 and ADH4) (Fig. 2). Therefore, in most tumor samples, we detected a dramatic decrease in the expression of genes that encode retinalreducing enzymes that are actively expressed in normal gastric tissues.

Expression of Genes That Encode RetinalOxidizing Enzymes RALDH1, 2, and 3 and ATRADegrading CYP26A1 and CRBP1 Three RALDH enzymes (1, 2, and 3) can be involved in retinaltoATRA oxidation. RALDH1 is expressed most actively in normal gastric mucosa (RALDH1 > RALDH3 > RALDH2). In gastric cancer, MOLECULAR BIOLOGY

Vol. 47

No. 2

2013

EXPRESSION OF GENES INVOLVED IN RETINOIC ACID BIOSYNTHESIS

287 (a)

Relative expression level, N/T

10

1

ADH4

0.1

ADH1C ADH1B 0.01

0.001

0.0001 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38

stage of cancer

I

II

III

IV (b)

Relative expression level, N/T

10

1

RDH12

0.1

AKR1B10 0.01

0.001

0.0001 stage of cancer

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38

I

III

II

IV

Fig. 2. The relative expression level of ADH1C, ADH1B, ADH4 (a), and AKR1B10, RDH12 (b) genes in gastric cancer, determined with realtime PCR in 38 paired samples normal/tumor. ACTB gene was used as reference control. I–IV are stages of gastric cancer.

the level of mRNAs that correspond to RALDH2 and RALDH3 genes is altered insignificantly, while the level of RALDH1 mRNA is decreased in most (57%) sam ples of gastric cancer (Table 3). MOLECULAR BIOLOGY

Vol. 47

No. 2

2013

The level of CYP26A1 mRNA in normal gastric mucosa was insignificant. A negligible increase (on average 2.6fold) was detected in 53% of tumor sam ples. The level of CRBP1 mRNA remained unaltered

288

KROPOTOVA et al.

in most tumor samples (Table 3). Therefore, the expression patterns of genes involved in ATRA biosyn thesis, which were obtained as a result of bioinformat ics analysis, RTPCR, and realtime PCR coincide in general. No significant correlations between mRNA content, age, and gender, as well as the stage of disease and metastases, have been identified. DISCUSSION Retinoic acid plays an important role in embryonic development, epithelial differentiation, cell growth, the maintenance of normal functioning of nervous and immune systems, and other processes. Natural and synthetic forms of ATRA were shown to inhibit the proliferation of tumor cells, including gastric cancer cells [29]. A significant amount of ATRA synthesized from alltransretinol in the presence of NAD+ was detected in human gastric mucosa [24]. ATRA biosyn thesis includes multiple enzymes that exhibit various catalytic activities, tissue specific expressions, and substrate and cofactor specificities [3, 4, 10, 14]. How ever, most works focus on the expression of one or sev eral genes and enzymes involved in ATRA biosynthesis and retinoid metabolism [17, 21, 30–32]. Only a few works describe the complex expression profile of retin oid metabolism enzymes in pathological conditions, such as uterine fibroid [33], nonalcoholic fatty liver disease [34], and testicular feminization in mice [35]. Key enzymes responsible for ATRA biosynthesis remain unknown in many tissues, including normal and tumor gastric tissues. Based on an analysis of current biomedical data, we made a list of candidate genes involved in ATRA bio synthesis and analyzed transcriptomic databases in order to identify candidate genes involved in this pro cess in gastric tissues (Table 1). The level of expression of most of these genes is altered in gastric cancer com pared to normal tissues. The microarray data that we used for analysis have been successfully used for the preliminary estimation of gene expression for a long time [27]. Genes subjected to RTPCR and realtime PCR analyses were selected regarding catalytic activity and substrate specificity of the encoded enzymes (Table 1). The oxidation of alltransretinol to retinal in nor mal gastric mucosa can involve ADH (ADH1B, ADH1C, ADH4, and ADH3), as well as NAD+ dependent RDH (RDH10, RDHL, and RDH5). The most prominent alteration is observed in the case of ADH4, ADH1B, and ADH1C genes (Fig. 2, Table 3), i.e., in the case of the most active genes. The level of RDHL mRNA is altered to a lesser extent, while the levels of RDH10, ADH3, and RDH5 in gastric cancer remains similar to those in normal tissues. Total effect of the observed alterations should be estimated regard ing not only the level of gene expression, but also the activity of the encoded enzymes. Although kinetic constants should be compared carefully due to differ

ent methods for their determination, it is evident that ADH4, RDH10, ADH1C, and ADH1B are the most active alltransretinoloxidizing enzymes [10, 13] involved in the oxidation of alltransretinol to retinal in vivo. Thus, mutant Adh1–/– mice synthesize ten times less ATRA than wildtype mice [4, 7]. The addi tion of ADH inhibitors, including 4methylpyrazole, ethanol, ranitidine hydrochloride, or cimetidine, hin ders ATRA synthesis in mice [24, 36]. Oxidoreductases ADH3, RDHL, and RDH5 exhibit low catalytic activity in alltransretinol oxida tion [8, 37, 38], which is why they have less impact to retinol oxidation than ADH4, ADH1, and RDH10 with regard to their low content in gastric tissues. Among all NAD+dependent retinoidactive oxi doreductases, RDH10 has the lowest KМ for alltrans retinol (approximately 0.035 μM) [13]. The published values of catalytic activity of ADH isoenzymes vary in a wide range, although the average values of the kinetic constants of ADH1 and RDH10 for alltransretinol oxidation are similar (KМ = 0.035–0.182 μM for the enzymes of ADH1 family) [39]. It should be noted that the content of ADH1B, ADH1C, and ADH4 mRNA is significantly reduced, even at early stages of gastric cancer. Over 90% of tumor samples exhibit a signifi cant reduction in the level of expression of at least one of these genes (Fig. 2). That is why RDH10, which is less active in gastric tissues, cannot compensate for the reduced expression of these genes. Therefore, in most tumor samples, we have detected a dramatic decrease in the mRNA level of genes that encode the enzymes that oxidize alltransretinol to retinal and are actively expressed in normal gastric tissues, which should reduce the retinal and ATRA synthesis and result in the accumulation of retinol (Fig. 3). The most active retinoloxidating ADH4 is located in the gastrointestinal tract, mostly in gastric tissues. The reduction in ADH4 activity, accompanied by a decrease in ATRA content is associated with enhanced inflammation, atrophy and intestinal metalasia [21]. Kinetical and genetic studies show that ADH1 and ADH4 can fulfill two major physiological functions, i.e., ethanol oxidation in stomach and ATRA synthe sis. The ethanolinduced disturbance of ATRA syn thesis can underlie the pathogenesis of alcoholic syn drome and the malignant transformation of the upper gastrointestinal tract tissues associated with alcohol uptake [40]. Retinal reduction to retinol in gastric tissues involves NADР+dependent RDH (RDH11 and RDH12), AKR (AKR1B10 and AKR1B1), and ADH (mostly ADH4 and less ADH1). Their mRNA content is decreased in the following series: AKR1В10 > ADH1B = ADH1C > ADH4 = RDH12 = RDH11 > AKR1В1. Maxi mal retinalreducing activity is exhibited by RDH12, ADH4, RDH11, and AKR1B10 [10, 14]; the most prominent alteration of gene expression is observed in the case of RDH12, ADH4, AKR1В10, ADH1B and ADH1C, which are the most active genes in gastric tis MOLECULAR BIOLOGY

Vol. 47

No. 2

2013

EXPRESSION OF GENES INVOLVED IN RETINOIC ACID BIOSYNTHESIS ADH1B ADH1C ADH4 ADH3 RDH10

alltransretinol

alltransretinoic acid (ATRA)

alltransretinal

RALDH1

RDHL

CHOH

289

RALDH3 CHO RALDH2

RDH5

RDH12 RDH11

AKR1B10

COOH

↑CYP26A1

AKR1B1 ADH4

degradation products

regulation of target genes transcription

Fig. 3. Disturbance of ATRA biosynthesis in gastric cancer. In normal gastric mucosa, alltransretinol is reversibly oxidized to alltransretinal by ADH (ADH1B, ADH1C, ADH3, ADH4) and NAD+dependent RDH (RDH10, RDHL, RDH5). The transformation of retinal to retinol is catalyzed by NAD+dependent RDH12 and RDH11, AKR1B10 and AKR1B1, as well as by ADH4. Alltransretinal is irreversibly oxidized to ATRA by RALDH1, 2, and 3. ATRA can be degraded by CYP26A1. Black arrows indicate alteration of gene expression in gastric tumor compared to normal tissues. Type weight and size of gene symbols reflect the level of expression in normal gastric tissues; the thickness of arrows reflect the altered level of gene expression in tumor tissues compared to the adjacent normal ones.

sues. Therefore, retinal to retinol conversion is dis turbed in gastric cancer. In human gastric tissues, retinal oxidation to ATRA is catalyzed by RALDH1, 2, and 3. We de monstrated that the RALDH1 gene was expressed at a higher level than RALDH3 and especially RALDH2. The content of retinoic acid in blood of Raldh–/– mice is less than that in wildtype animals [41]. Decreased ATRA synthesis was observed after the addition of ace taldehyde inhibitor of RALDH [24]. The decreased expression of the gene that encodes RALDH1, the major alltransretinaloxidizing enzyme in gastric tis sues, can also result in the reduction of ATRA synthesis. Cytochrome CYP26A1 catalyses ATRA degrada tion to the inactive polar metabolites in gastric tissues [16]. We detected moderate increase of CYP26A1 expression, which indicated that ATRA degradation was increased, although not to a great extent, in gastric tumors, which may also result in the reduction of the ATRA content. Various epithelial tissues can vary not only in the expression pattern of ATRAsynthesizing enzymes, but also in the altered pattern of their expression in carcinogenesis. Thus, expression of AKR1B10 is signif icantly increased in nonsmallcell lung cancer and liver cancer [30, 42, 43] and is significantly decreased in gastrointestinal cancers (colon cancer [44] and gas tric cancer (this work)). In the gastrointestinal tract, ADH, AKR, and RALDH, which are the genes analyzed in our work, can be involved in ATRA biosynthesis, as well as in the detoxication of the organism [45, 46]. The reduced content of these enzymes in the stomach can have var ious effects associated with not only the ATRA defi ciency, but also with the toxicity of endogenous and exogenous spirits and aldehydes. Adh1–/– mice exhibit significantly reduced ATRA biosynthesis and display increased retinol toxicity [7]. CRBP1 regulates retinol MOLECULAR BIOLOGY

Vol. 47

No. 2

2013

transport and can protect retinol from oxidation to retinal and conversion to ATRA. We demonstrated that CRBP1 expression was unaltered in most gastric tumors, which is why the increased retinol level in tumors was associated with an increased amount of free retinol, which might be hydroxilated by cyto chromes P450, which yield toxic byproducts [47]. Several mechanisms ensure the functioning of ATRA, which regulates the expression of numerous genes via binding to and activating nuclear receptors [1]. ATRA interacts with RAR receptors (α, β, γ), which form a heterodimeric complex with Xretinoid receptor (RXRα, β, γ) and regulate the transcription of ATRAdependent genes. Target genes often include retinoic acid response elements (RARE) that interact with RAR/RXR heterodimer. ATRARAR/RXR complex can regulate gene expression independently of RARE. Retinoic acid also affects the activity of nuclear receptors that differ from RAR and RXR. The transcription of target genes is controlled both by direct interaction with nuclear receptor heterodimer and by intermediate transcription factors. ATRAreg ulated genes include wellknown genes that encode cyclins and cyclindependent kinases (CDK) impli cated in cellcycle regulation, CDK inhibitors, homeobox Hox genes that control growth and differ entiation, oncogene RAMP that encodes matrixasso ciated protein and is specifically expressed in gastric cancer, and genes associated with retinoid metabolism and signal transduction (CRBP1, ADH3, CYP26A1, RARα2, RARβ2 etc.). Retinoic acid regulates the tran scription of numerous noncoding RNA via RAR/RXR dimers [1, 48, 49]. The interaction of ATRA with retinoid receptors is required for the func tioning of the retinoid signal system. The alteration of RAR and RXR expression, which affects signal func tion, was detected in various types of cancer, including gastric cancer [48, 50].

290

KROPOTOVA et al.

This work describes the complex expression pattern of numerous genes implicated in ATRA biosynthesis in gastric cancer for the first time. The disturbances were detected at every stage of the process. ATRA rep resents a pleotropic regulator of transcriptional activ ity of multiple genes; it belongs to the group of factors required for normal cell differentiation and prolifera tion. The inhibition of ATRA synthesis might cause multiple pathological changes in normal gastric mucosa [21]. The disturbed balance of retinoic acid is considered to be associated with pathogenesis of pre malignant states and malignant transformations in gastric mucosa [51]. Most key enzymes involved in ATRA biosynthesis are polyfunctional; they are involved not only in retin oid metabolism, but also in neuromediator catabo lism, conversion of steroid hormones and omegafatty acids, synthesis of cholesterol and bile acids, and in protection from exogenous ethanol and endogenous toxic compounds. Therefore, the altered expression of genes, analyzed in this work, can affect the regulation of not only metabolic and signal retinoid pathways, but also of other systems. ACKNOWLEDGMENTS The work was financially supported by the Russian Foundation for Basic Research (100401760a and 120400388a), by the International Scientific and Technical Center (no. 3909), and by the Ministry of Education and Science (state contract no. 16.552.11.7034).

8.

9.

10.

11. 12.

13.

14.

REFERENCES 1. Blomhoff R., Blomhoff H.K. 2006. Overview of retinoid metabolism and function. J. Neurobiol. 66, 606–630. 2. Harrison E.H. 2012. Mechanisms involved in the intes tinal absorption of dietary vitamin A and provitamin A carotenoids. Biochim. Biophys. Acta. 1821, 70–77. 3. Napoli J.L. 2012. Physiological insights into alltrans retinoic acid biosynthesis. Biochim. Biophys. Acta. 1821, 152–167. 4. Kumar S., Sandell L.L., Trainor P.A., Koentgen F., Duester G. 2012. Alcohol and aldehyde dehydrogena ses: Retinoid metabolic effects in mouse knockout models. Biochim. Biophys. Acta. 1821, 198–205. 5. Duester G., Farrés J., Felder M.R., Holmes R.S., Höög J.O., Parés X., Plapp B.V., Yin S.J., Jörnvall H. 1999. Recommended nomenclature for the vertebrate alcohol dehydrogenase gene family. Biochem. Pharma col. 58, 389–395. 6. Estonius M., Svensson S., Höög J.O. 1996. Alcohol dehydrogenase in human tissues: Localization of tran scripts coding for five classes of the enzyme. FEBS Lett. 397, 338–342. 7. Molotkov A., Deltour L., Foglio M.H., Cuenca A.E., Duester G. 2002. Distinct retinoid metabolic functions for alcohol dehydrogenase genes Adh1 and Adh4 in pro

15.

16. 17.

18. 19.

20.

tection against vitamin A toxicity or deficiency revealed in double null mutant mice. J. Biol. Chem. 277, 13804– 13811. Molotkov A., Fan X., Deltour L., Foglio M.H., Martras S., Farrés J., Parés X., Duester G. 2002. Stimulation of ret inoic acid production and growth by ubiquitously expressed alcohol dehydrogenase Adh3. Proc. Natl. Acad. Sci. U. S. A. 99, 5337–5342. Gallego O., Belyaeva O.V., Porté S., Ruiz F.X., Stet senko A.V., Shabrova E.V., Kostereva N.V., Farrés J., Parés X., Kedishvili N.Y. 2006. Comparative functional analysis of human mediumchain dehydrogenases, shortchain dehydrogenases/reductases, and aldoketo reductases with retinoids. Biochem. J. 399, 101–109. Parés X., Farrés J., Kedishvili N., Duester G. 2008. Medium and shortchain dehydrogenase/reductase gene and protein families: Mediumchain and short chain dehydrogenases/reductases in retinoid metabo lism. Cell. Mol. Life Sci. 65, 3936–3949. Lidén M., Eriksson U. 2006. Understanding retinol metabolism: Structure and function of retinol dehydro genases. J. Biol. Chem. 281, 13001–13004. Kallberg Y., Oppermann U., Persson B. 2010. Classifi cation of the shortchain dehydrogenase/reductase superfamily using hidden Markov models. FEBS J. 277, 2375–2386. Belyaeva O.V., Johnson M.P., Kedishvili N.Y. 2008. Kinetic analysis of human enzyme RDH10 defines the characteristics of a physiologically relevant retinol dehydrogenase. J. Biol. Chem. 28, 20299–20308. Belyaeva O.V., Korkina O.V., Stetsenko A.V., Kim T., Nelson P.S., Kedishvili N.Y. 2005. Biochemical prop erties of purified human retinol dehydrogenase 12 (RDH12): Catalytic efficiency toward retinoids and C9 aldehydes and effects of cellular retinolbinding protein type I (CRBPI) and cellular retinaldehydebinding protein (CRALBP) on the oxidation and reduction of retinoids. Biochemistry. 44, 7035–7047. Crosas B., Hyndman D.J., Gallego O., Martras S., Parés X., Flynn T.G., Farrés J. 2003. Human aldose reductase and human small intestine aldose reductase are efficient retinal reductases: Consequences for retinoid metabolism. Biochem. J. 373, 973–979. Ross A.C., Zolfaghari R. 2011. Cytochrome P450s in the regulation of cellular retinoic acid metabolism. Annu. Rev. Nutr. 31, 65–87. Jette C., Peterson P.W., Sandoval I.T., Manos E.J., Hadley E., Ireland C.M., Jones D.A. 2004. The tumor suppressor adenomatous polyposis coli and caudal related homeodomain protein regulate expression of retinol dehydrogenase L. J. Biol. Chem. 279, 34397– 34405. Pasquali D., Thaller C., Eichele G. 1996. Abnormal level of retinoic acid in prostate cancer tissues. J. Clin. Endocrinol. Metab. 81, 2186–2191. Mira Y.L.R., Zheng W.L., Kuppumbatti Y.S., Rexer B., Jing Y., Ong D.E. 2000. Retinol conversion to retinoic acid is impaired in breast cancer cell lines relative to normal cells. J. Cell Physiol. 185, 302–309. Williams S.J., Cvetkovic D., Hamilton T.C. 2009. Vita min A metabolism is impaired in human ovarian can cer. Gynecol. Oncol. 112, 637–645. MOLECULAR BIOLOGY

Vol. 47

No. 2

2013

EXPRESSION OF GENES INVOLVED IN RETINOIC ACID BIOSYNTHESIS 21. Matsumoto M., Yokoyama H., Suzuki H., Shiraishi Yokoyama H., Hibi T. 2005. Retinoic acid formation from retinol in the human gastric mucosa: Role of class IV alco hol dehydrogenase and its relevance to morphological changes. Am. J. Physiol. Gastrointest. Liver Physiol. 289, G429–433. 22. Jemal A., Bray F., Center M.M., Ferlay J., Ward E., Forman D. 2011. Global cancer statistics. CA Cancer J. Clin. 61, 69–90. 23. Davydov M.I., Aksel’ E.M. 2011. Morbidity and mor tality due to malignant neoplasms in Russia and CIS countries in 2009. Vestn. Ross. Onkol. Hauch. Tsentra im. N.N. Blokhina Ross. Akad. Med. Sci., 22, Suppl. 1, 54–123. 24. Yokoyama H., Matsumoto M., Shiraishi H., Miyagi M., Kato And S., Ishii H. 2001. Nicotinamide adenine dinucleotidedependent retinoic acid forma tion from retinol in the human gastric mucosa: Inhibi tion by ethanol, acetaldehyde, and H2 blockers. Alco hol. Clin. Exp. Res. 25, 24S–28S. 25. Persson C., Sasazuki S., Inoue M., Kurahashi N., Iwasaki M., Miura T., Ye W., Tsugane S.; JPHC Study Group. 2008. Plasma levels of carotenoids, retinol and tocopherol and the risk of gastric cancer in Japan: A nested casecontrol study. Carcinogenesis. 29, 1042– 1048. 26. Larsson S.C., Bergkvist L., Näslund I., Rutegård J., Wolk A. 2007. Vitamin A, retinol, and carotenoids and the risk of gastric cancer: A prospective cohort study. Am. J. Clin. Nutr. 85, 497–503. 27. Rhodes D.R., Yu J., Shanker K., Deshpande N., Varambally R., Ghosh D., Barrette T., Pandey A., Chinnaiyan A.M. 2004. ONCOMINE: A cancer microarray database and integrated datamining plat form. Neoplasia. 6, 1–6. 28. Manzeniuk O.Yu., Malakho S.G., Pekhov V.M., Kosorukova I.S., Poltaraus A.B. 2006. Characterization of the universal Russian reagent sets for realtime PCR and its application to molecular oncodiagnosis. Mol. Biol. (Moscow). 40, 305–311. 29. Zhang J.P., Chen X.Y., Li J.S. 2007. Effects of alltrans retinoic on human gastric cancer cells BGC823. J. Dig. Dis. 8, 29–34. 30. Fukumoto S., Yamauchi N., Moriguchi H., Hippo Y., Watanabe A., Shibahara J., Taniguchi H., Ishikawa S., Ito H., Yamamoto S., Iwanari H., Hironaka M., Ish ikawa Y., Niki T., Sohara Y., Kodama T., Nishimura M., Fukayama M., DosakaAkita H., Aburatani H. 2005. Overexpression of the aldoketo reductase family pro tein AKR1B10 is highly correlated with smokers’ non small cell lung carcinomas. Clin. Cancer Res. 11, 1776– 1785. 31. Leclerc J., CourcotNgoubo Ngangue E., Cauffiez C., Allorge D., Pottier N., Lafitte J.J., Debaert M., Jaillard S., Broly F., LoGuidice J.M. 2011. Xenobiotic metabolism and disposition in human lung: Transcript profiling in nontumoral and tumoral tissues. Bio chimie. 93, 1012–1027. 32. Chiang C.P., Jao S.W., Lee S.P., Chen P.C., Chung C.C., Lee S.L., Nieh S., Yin S.J. 2012. Expres sion pattern, ethanolmetabolizing activities, and cel lular localization of alcohol and aldehyde dehydroge nases in human large bowel: Association of the func MOLECULAR BIOLOGY

Vol. 47

No. 2

2013

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

291

tional polymorphisms of ADH and ALDH genes with hemorrhoids and colorectal cancer. Alcohol. 46, 37–49. Zaitseva M., Vollenhoven B.J., Rogers P.A. 2007. Retinoic acid pathway genes show significantly altered expression in uterine fibroids when compared with nor mal myometrium. Mol. Hum. Reprod. 13, 577–585. Ashla A.A., Hoshikawa Y., Tsuchiya H., Hashiguchi K., Enjoji M., Nakamuta M., Taketomi A., Maehara Y., Shomori K., Kurimasa A., Hisatome I., Ito H., Shiota G. 2010. Genetic analysis of expression profile involved in retinoid metabolism in nonalcoholic fatty liver disease. Hepatol. Res. 40, 594–604. O’Shaughnessy P.J., Abel M., Charlton H.M., Hu B., Johnston H., Baker P.J. 2007. Altered expression of genes involved in regulation of vitamin A metabolism, solute transportation, and cytoskeletal function in the androgeninsensitive tfm mouse testis. Endocrinology. 148, 2914–2924. Collins M.D., Eckhoff C., Chahoud I., Bochert G., Nau H. 1992. 4Methylpyrazole partially ameliorated the teratogenicity of retinol and reduced the metabolic formation of alltransretinoic acid in the mouse. Arch. Toxicol. 66, 652–659. Chetyrkin S.V., Belyaeva O.V., Gough W.H., Kedishvili N.Y. 2001. Characterization of a novel type of human microsomal 3αhydroxysteroid dehydroge nase: Unique tissue distribution and catalytic proper ties. J. Biol. Chem. 276, 22278–22286. Wang J., Chai X., Eriksson U., Napoli J.L. 1999. Activ ity of human 11cisretinol dehydrogenase (Rdh5) with steroids and retinoids and expression of its mRNA in extraocular human tissue. Biochem. J. 338, 23–27. Chase J.R., Poolman M.G., Fell D.A. 2009. Contribu tion of NADH increases to ethanol’s inhibition of retinol oxidation by human ADH isoforms. Alcohol. Clin. Exp. Res. 33, 571–580. Yin S.J., Chou C.F., Lai C.L., Lee S.L., Han C.L. 2003. Human class IV alcohol dehydrogenase: Kinetic mechanism, functional roles and medical relevance. Chem. Biol. Interact. 143/144, 219–227. Molotkov A., Duester G. 2003. Genetic evidence that retinaldehyde dehydrogenase Raldh1 (Aldh1a1) func tions downstream of alcohol dehydrogenase Adh1 in metabolism of retinol to retinoic acid. J. Biol. Chem. 278, 36085–36090. Mashkova T.D., Oparina N.Yu., Zinov’eva O.L., Kro potova E.S., Dubovaya V.I., Poltaraus A.B., Fridman M.V., Kopantsev E.P., Vinogradova T.V., Zinov’eva M.V., Laktionov K.K., Kasymova O.T., Zborovskaya I.B., Sverdlov E.D., Kisselev L.L. 2006. Transcription of TIMP3, DAPK1, and AKR1B10 in squamouscell lung cancer. Mol. Biol. (Moscow). 40, 945–951. Scuric Z., Stain S.C., Anderson W.F., Hwang J.J. 1998. New member of aldose reductase family proteins over expressed in human hepatocellular carcinoma. Hepa tology. 27, 943–950. Kropotova E.S., Tychko R.A., Zinov’eva O.L., Zyry anova A.F., Khankin S.L., Cherkes V.L., Aliev V.A., Beresten S.F., Oparina N.Yu., Mashkova T.D. 2010. Downregulation of AKR1B10 expression in colorectal cancer. Mol. Biol. (Moscow). 44, 216–222.

292

KROPOTOVA et al.

45. Martin H.J., Maser E. 2009. Role of human aldo ketoreductase AKR1B10 in the protection against toxic aldehydes. Chem. Biol. Interact. 178, 145–150. 46. Westerlund M., Belin A.C., Felder M.R., Olson L., Galter D. 2007. High and complementary expression patterns of alcohol and aldehyde dehydrogenases in the gastrointestinal tract: implications for Parkinson’s dis ease. FEBS J. 274, 1212–1223. 47. Roberts E.S., Vaz A.D.N., Coon M.J. 1992. Role of isozymes of rabbit microsomal cytochrome P450 in the metabolism of retinoic acid, retinol, and retinal. Mol. Pharmacol. 41, 427–433. 48. Tang X.H., Gudas L.J. 2011. Retinoids, retinoic acid receptors, and cancer. Annu. Rev. Pathol. 6, 345–364.

49. Li J., Ng E.K., Ng Y.P., Wong C.Y., Yu J., Jin H., Cheng V.Y., Go M.Y., Cheung P.K., Ebert M.P., Tong J., To K.F., Chan F.K., Sung J.J., Ip N.Y., Leung W.K. 2009. Identification of retinoic acidregu lated nuclear matrixassociated protein as a novel regu lator of gastric cancer. Br. J. Cancer. 101, 691–698. 50. Hu K.W., Chen F.H., Ge J.F., Cao L.Y., Li H. 2012. Retinoid receptors in gastric cancer: Expression and influence on prognosis. Asian Pac. J. Cancer Prev. 13, 1809–1817. 51. Jiang S.Y., Shen S.R., Shyu R.Y., Yu J.C., Harn H.J., Yeh M.Y., Lee M.M., Chang Y.C. 1999. Expression of nuclear retinoid receptors in normal, premalignant and malignant gastric tissues determined by in situ hybridi zation. Br. J. Cancer. 80, 206–214.

MOLECULAR BIOLOGY

Vol. 47

No. 2

2013