Sage Journals: Discover world-class research

Abstract

Dimethyl sulfoxide (DMSO) is a very common organic solvent used for dissolving lipophilic substances, for example for in vitro cell-based assays. At the same time, DMSO is known to be cytotoxic at high concentrations. Therefore, it is important to define threshold concentrations of DMSO for cells but relevant data at the molecular level are very limited. We have focused on conducting microarray analyses of human and rat hepatocytes treated with more than 100 chemicals in attempts to identify candidate biomarker genes. In the present study, the effects of DMSO on gene expression and cytotoxicity were assessed in human cryopreserved hepatocytes and rat primary cultured hepatocytes. A cytotoxicity test with lactate dehydrogenase (LDH) activity demonstrated DMSO to be noncytotoxic up to a concentration of 2% (v/v) in both cases and there were only few effects on the gene expression profiles up to 0.5% (v/v). The observed differences from controls were considered to be of little toxicological importance, but still need to be taken into account in interpretation of findings when DMSO is used at high concentration.

Keywords

gene expression profile DMSO human cryopreserved hepatocytes rat primary cultured hepatocytes cytotoxicity

Introduction

A large-scale gene expression database, termed TG-GATES (Genomics Assisted Toxicity Evaluation System), has been established by the Toxicogenomics Project in Japan.¹ About 150 chemicals, mainly for medicinal use, were selected, and gene expression in rat liver, rat kidney, rat primary cultured hepatocytes, and human cryopreserved hepatocytes is being comprehensively analyzed using Affymetrix GeneChip system (Santa Clara, CA, USA). In the project, rat and human hepatocytes are treated with toxicological prototype drugs, in three dose-ranges, and samples are collected 2, 8, and 24 hr after a single treatment. One of the main aims of our project is to identify candidate biomarker genes to predict and/or diagnose toxicity.

The actual dose-ranges are set according to dose-finding studies. The maximum concentration was set at 10 mM in the case of a chemical with high solubility. The concentration of a chemical with low solubility is determined with reference to solubility in 0.1% (v/v) DMSO, which is the commonly used concentration for in vitro assays. However, in some cases, this means that the concentration is too low for the effects of the chemical to be adequately reflected in the gene expression profile. It is then necessary to increase the chemical concentration, but this also necessitates increasing the DMSO concentration to obtain a solution. Since DSMO is itself toxic at high concentration, this could result in misleading findings. Since there have been no reports of comprehensive gene expression profiles after treatment with DMSO, the present study was conducted using human cryopreserved hepatocytes and rat primary cultured hepatocytes.

There are several reports about influence of DMSO on expression of genes in cells. For example, it has differentiation-inducing effects on embryonic stem (ES) cells, which are most sensitive towards the cytotoxic effects of DMSO including Oct-4.² Klinken et al. also reported induction of differentiation in murine erythroleukemia (MEL) cells with alteration of proto-oncogene levels.³ Similar findings have been described for HL60 cells.⁴ As far as housekeeping genes are concerned, Nishimura et al. examined the effects of DMSO on the expression of beta-actin (Actb), glyceraldehyde-3-phosphate dehydrogenase (Gapdh), beta-glucuronidase (Gusb), phosphoglycerate kinase 1 (Pgk1), peptidylprolyl isomerase A (Ppia), and transferrin receptor (Tfrc) mRNA in cultures of C2C12 myotubes and the mRNA levels of some housekeeping genes were affected by exposure of DMSO concentrations of 0.5% (v/v) or more.⁵ The induction of some drug-metabolizing enzyme genes by DMSO has also been reported.^6–8

Thus, for in vitro assays, it is important to define the threshold concentration of DMSO for cells in accordance with the endpoint. The aim of the present study was to identify the lowest concentration of DMSO at which no significant effects on gene expression profile were observed, with particular attention to candidate biomarker genes. For this purpose, human cryopreserved hepatocytes and rat hepatocytes were examined for effects of DMSO on gene expression with a DNA microarray system and cytotoxicity with a lactate dehydrogenase (LDH) leakage method.

Materials and methods

Cell culture and in vivo studies

Human cryopreserved hepatocytes, purchased from CellzDirect, Inc. (Durham, NC, USA), were recovered in cryopreserved hepatocyte-recovered medium (CellzDirect, Inc.) and seeded in HCM Bullekit^®(Cambrex Corp., East Rutherford, NJ, USA) supplemented with 10% FBS (Invitrogen Corp., Carlsbad, CA, USA) in 6-well collagen type-1 coated plates (Becton, Dickinson and Company, Franklin Lakes, NJ, USA) at a concentration of 1.2 × 10⁶ cells/2 mL/well for 4 hours in a humidified atmosphere (37°C, 5% CO₂). After 4-hour culture, medium was replaced with HCM Bullekit^®without FBS and human hepatocytes were cultured for 20 more hours.

At day 2, medium was replaced with HCM Bullekit^®containing 0% (v/v), 0.1% (v/v), 0.5% (v/v), 0.75% (v/v), 1% (v/v), or 2% (v/v) DMSO (Kanto Chemical Co., Inc., Tokyo, Japan, purity > 99.7%), and the hepatocytes were cultured for 24 more hours in a humidified atmosphere (37°C, 5% CO₂).

At day 3, medium was collected for measurement of LDH activity and total RNA samples were collected for gene expression analysis. Extraction of total RNA was conducted using a RNeasy Mini Kit (Qiagen, Hilden, Germany) before quantification with a spectrophotometer DU-7400 (Beckman Coulter, Fullerton, CA, USA) and assessment of ribosomal RNA integrity using a 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA, USA).

Rat primary-cultured hepatocytes, prepared from a 5-week-old male Spraque-Dawley rat (Charles River Japan Inc., Kanagawa, Japan), were seeded in HCM Bullekit^®supplemented with 10% FBS in 6-well collagen type-1 coated plates at a concentration of 1.0 × 10⁶ cells/2 mL/well for 2-3 hours in a humidified atmosphere (37°C, 5% CO₂). After 2-3 hours culture, medium was replaced with HCM Bullekit^®without FBS and rat hepatocytes were cultured for 16 hours.

At days 2 and day 3, rat primary cultured hepatocytes were treated the same as human cryopreserved hepatocytes.

Gene expression analysis

Microarray analysis was conducted on three samples for each group using HGU133 plus 2.0 probe arrays and RG230 2.0 probe arrays (Affymetrix). The procedures were basically conducted following the manufacturer’s protocol, as previously reported.^9,10 The obtained image files were analyzed with the Affymetrix data suite system, Microarray Suite 5.0 (MAS 5.0) and derived signal values were globally normalized and targeted to all probe sets equal to 500 before comparative analysis to examine gene expression differences between treatment and control samples.

Quantitative RT-PCR assay validation

To confirm the gene expression data using microarray, we conducted quantitative RT-PCR for some rat genes. Five genes (Ccl20, sulfotransferase family, cytosolic, 1A, phenol-preferring, member 1 [Sult1a1], Cyp1a1, carbonic anhydrase 2 [Car2], and Cyp2c12) were measured by real-time PCR using TaqMan® Gene Expression Assays and the 7900HT Fast Real-Time PCR System (Applied Biosystems, Foster City, CA, USA). cDNAs were generated from 12 ng of total RNA using Invtirogen reverse transcription reagents (SuperScript III Reverse Transcriptase (10,000 units), RNaseOUT, Random Primers, 10 mM dNTP, and 0.1 M DTT). Three replicates were run for each gene for each sample in a 384-well format plate.

Measurement of LDH activity

The collected medium was centrifuged at 1500 × g or more for 5 min at 4°C. About 0.8 mL of the supernatant was applied to a biochemical autoanalyzer TBA-200FR (Toshiba, Tokyo, Japan) with the UV-rate method.

Data analysis

Raw probe intensities in Affymetrix CEL files were normalized using the MAS 5.0 algorithm with default parameters. After the MAS 5.0 process, we selected each probe for which a detection call is “present” among all conditions in this study. The fold change values were calculated as the ratio of gene expression values between DMSO treatment data versus control data. To compare gene expression levels among five different DMSO concentrations, we normalized the expression data using Tukey’s biweight method and converted them into signal/control log₂ratios. We used the smoothing spline clustering method¹¹ to sort and identify gene expression patterns that were dependent on the DMSO concentration. To find significantly and differentially expressed gene sets between control and DMSO treatment data, gene set enrichment analysis (GSEA)¹² was performed. The gene set data for the human and the rat were extracted from the Kyoto encyclopedia of genes and genomes (KEGG) pathway database.¹³

Results

Gene expression analysis

In human cryopreserved hepatocytes, there were only 4 probe sets, which altered in expression (p < 0.05 and more than 2-fold) at a concentration of 0.1% (v/v) DMSO compared with control (0% DMSO). There were 16 probe sets altered in expression at 0.5% (v/v) DMSO, 57 probe sets at 0.75% (v/v) DMSO, 133 probe sets at 1% (v/v) DMSO and 752 probe sets at 2% (v/v) DMSO. In rat primary cultured hepatocytes, there were only 1 probe set, which altered in expression (p < 0.05 and more than 2-fold) at a concentration of 0.1% (v/v) DMSO compared with control (0% DMSO). There were 10 probe sets altered in expression at 0.5% (v/v) DMSO, 22 probe sets at 0.75% (v/v) DMSO, 67 probe sets at 1% (v/v) DMSO, and 497 probe sets at 2% (v/v) DMSO (Figure 1a–d).

Figure 1.

The numbers of differentially expressed genes (probe sets) in human cryopreserved hepatocytes (a) and rat primary cultured hepatocytes (c) are shown. Solid and dashed lines indicate up- and down-regulated genes, respectively. Circles and squares indicate 2.0- and 1.5-fold changes, respectively. The number of the probe sets was significantly increased between 1% (v/v) and 2% (v/v) dimethyl sulfoxide (DMSO). The numbers of differentially expressed genes (probe sets) commonly observed from 0.1% (v/v) DMSO to higher concentrations in human cryopreserved hepatocytes (b) and rat primary-cultured hepatocytes (d) are shown.

The GSEA detected differentially expressed gene sets in each DMSO treatment data compared to the control data. We analyzed 165 and 195 gene sets for human and rat hepatocytes, respectively. The differentially expressed gene sets were shown in Table 1. In the case of human hepatocytes, the top 10 differentially expressed gene sets at 0.1% (v/v) DMSO were not within the top 10 at 1% (v/v) DMSO. Distinct gene set changes were observed between 0.75% (v/v) DMSO and 1% (v/v) DMSO. In rat primary cultured hepatocytes, of the top 10 gene sets detected at 0.1% (v/v) DMSO, only 2 gene sets were detected at 2% (v/v) DMSO. The data indicate that 0.1% (v/v) DMSO had distinctly less influence than higher DMSO concentration. From 0.5% (v/v) DMSO to 1% (v/v) DMSO, some overlapping gene sets like RNO04621_ NOD-LIKE_RECEPTOR_SIGNALING_PATHWAY were observed.

Table 1.

The differentially expressed gene sets in human and rat hepatocytes are shown^a

DMSO Con. (%)	Gene Set (KEGG Pathway ID and Name)	NOM p Value	FDR q Value	FWER p Value
0.1	HSA03010_RIBOSOME	0.00	0.00	0.00
	HSA03410_BASE EXCISION REPAIR	0.00	0.06	0.07
	HSA04110_CELL CYCLE	0.00	0.31	0.47
	HSA03050_PROTEASOME	0.02	0.24	0.48
	HSA03030_DNA REPLICATION	0.03	0.25	0.57
	HSA00760_NICOTINATE AND NICOTINAMIDE METABOLISM	0.03	0.21	0.58
	HSA00480_GLUTATHIONE METABOLISM	0.02	0.54	0.92
	HSA00240_PYRIMIDINE METABOLISM	0.02	0.48	0.92
	HSA03430_MISMATCH REPAIR	0.05	0.57	0.97
	HSA05012_PARKINSON’s DISEASE	0.00	0.86	0.99
	HSA04020_CALCIUM SIGNALING PATHWAY	0.03	0.98	1.00
0.5	HSA03010_RIBOSOME	0.00	0.03	0.03
	HSA00590_ARACHIDONIC ACID METABOLISM	0.00	0.27	0.40
	HSA00591_LINOLEIC ACID METABOLISM	0.02	0.27	0.54
	HSA04742_TASTE TRANSDUCTION	0.04	0.20	0.54
	HSA00480_GLUTATHIONE METABOLISM	0.00	0.34	0.85
	HSA05012_PARKINSON’s DISEASE	0.05	0.83	1.00
	HSA04080_NEUROACTIVE LIGAND-RECEPTOR INTERACTION	0.05	0.74	1.00
0.75	HSA00590_ARACHIDONIC ACID METABOLISM	0.00	0.02	0.02
	HSA03010_RIBOSOME	0.00	0.20	0.31
	HSA00591_LINOLEIC ACID METABOLISM	0.00	0.14	0.32
	HSA00190_OXIDATIVE PHOSPHORYLATION	0.00	0.16	0.46
	HSA05012_PARKINSON’s DISEASE	0.00	0.15	0.50
	HSA00760_NICOTINATE AND NICOTINAMIDE METABOLISM	0.02	0.30	0.85
	HSA03050_PROTEASOME	0.04	0.42	0.95
	HSA05322_SYSTEMIC LUPUS ERYTHEMATOSUS	0.02	0.46	1.00
1	HSA04130_SNARE INTERACTIONS IN VESICULAR TRANSPORT	0.00	0.52	0.35
	HSA05322_SYSTEMIC LUPUS ERYTHEMATOSUS	0.00	0.29	0.38
	HSA05120_EPITHELIAL CELL SIGNALING IN HELICOBACTER PYLORI INFECTION	0.00	0.44	0.65
	HSA04912_GNRH SIGNALING PATHWAY	0.00	0.37	0.69
	HSA00591_LINOLEIC ACID METABOLISM	0.02	0.39	0.80
	HSA00601_GLYCOSPHINGOLIPID BIOSYNTHESIS - LACTO AND NEOLACTO SERIES	0.02	0.36	0.84
	HSA04142_LYSOSOME	0.00	0.39	0.90
	HSA04010_MAPK SIGNALING PATHWAY	0.03	0.69	1.00
2	HSA04912_GNRH SIGNALING PATHWAY	0.00	1.00	0.76
	HSA05322_SYSTEMIC LUPUS ERYTHEMATOSUS	0.03	0.85	0.85
	HSA04916_MELANOGENESIS	0.00	0.89	0.91
	HSA04070_PHOSPHATIDYLINOSITOL SIGNALING SYSTEM	0.03	0.98	0.98
0.1	RNO03420_NUCLEOTIDE_EXCISION_REPAIR	0.00	0.06	0.04
0.5	RNO04621_NOD-LIKE_RECEPTOR_SIGNALING_PATHWAY	0.00	0.52	0.50
	RNO03420_NUCLEOTIDE_EXCISION_REPAIR	0.00	0.41	0.74
	RNO04623_CYTOSOLIC_DNA-SENSING_PATHWAY	0.00	1.00	1.00
	RNO04062_CHEMOKINE_SIGNALING_PATHWAY	0.00	0.83	1.00
	RNO05010_ALZHEIMER’s_DISEASE	0.00	0.69	1.00
	RNO05016_HUNTINGTON’s_DISEASE	0.00	0.67	1.00
	RNO05012_PARKINSON’s_DISEASE	0.02	0.98	1.00
	RNO00190_OXIDATIVE_PHOSPHORYLATION	0.02	1.00	1.00
0.75	RNO04621_NOD-LIKE_RECEPTOR_SIGNALING_PATHWAY	0.00	0.81	0.49
	RNO05016_HUNTINGTON’s_DISEASE	0.00	0.58	0.64
	RNO00071_FATTY_ACID_METABOLISM	0.00	0.44	0.71
	RNO00051_FRUCTOSE_AND_MANNOSE_METABOLISM	0.05	0.60	0.90
	RNO05012_PARKINSON’s_DISEASE	0.00	0.60	1.00
	RNO04062_CHEMOKINE_SIGNALING_PATHWAY	0.00	0.55	1.00
	RNO03420_NUCLEOTIDE_EXCISION_REPAIR	0.00	0.50	1.00
	RNO04330_NOTCH_SIGNALING_PATHWAY	0.00	0.46	1.00
	RNO05010_ALZHEIMER’s_DISEASE	0.00	0.49	1.00
	RNO04120_UBIQUITIN_MEDIATED_PROTEOLYSIS	0.00	0.45	1.00
	RNO00330_ARGININE_AND_PROLINE_METABOLISM	0.04	0.54	1.00
1	RNO04621_NOD-LIKE_RECEPTOR_SIGNALING_PATHWAY	0.00	0.05	0.02
	RNO05016_HUNTINGTON’s_DISEASE	0.00	0.06	0.11
	RNO05012_PARKINSON’s_DISEASE	0.00	0.10	0.24
	RNO00051_FRUCTOSE_AND_MANNOSE_METABOLISM	0.00	0.26	0.76
	RNO04330_NOTCH_SIGNALING_PATHWAY	0.00	0.28	0.85
	RNO00534_HEPARAN_SULFATE_BIOSYNTHESIS	0.00	0.33	0.92
	RNO04062_CHEMOKINE_SIGNALING_PATHWAY	0.00	0.35	1.00
	RNO05010_ALZHEIMER’s_DISEASE	0.00	0.32	1.00
	RNO00190_OXIDATIVE_PHOSPHORYLATION	0.00	0.33	1.00
	RNO04620_TOLL-LIKE_RECEPTOR_SIGNALING_PATHWAY	0.00	0.32	1.00
2	RNO00280_VALINE,_LEUCINE_AND_ISOLEUCINE_DEGRADATION	0.00	0.02	0.00
	RNO04621_NOD-LIKE_RECEPTOR_SIGNALING_PATHWAY	0.00	0.03	0.04
	RNO04062_CHEMOKINE_SIGNALING_PATHWAY	0.00	0.02	0.04
	RNO05010_ALZHEIMER’s_DISEASE	0.00	0.05	0.13
	RNO04620_TOLL-LIKE_RECEPTOR_SIGNALING_PATHWAY	0.00	0.06	0.13
	RNO00071_FATTY_ACID_METABOLISM	0.00	0.12	0.35
	RNO04330_NOTCH_SIGNALING_PATHWAY	0.00	0.13	0.44
	RNO00600_SPHINGOLIPID_METABOLISM	0.00	0.12	0.53
	RNO04130_SNARE_INTERACTIONS_IN_VESICULAR_TRANSPORT	0.00	0.12	0.53
	RNO04622_RIG-I-LIKE_RECEPTOR_SIGNALING_PATHWAY	0.00	0.12	0.56
	RNO04623_CYTOSOLIC_DNA-SENSING_PATHWAY	0.00	0.16	0.76
	RNO00640_PROPANOATE_METABOLISM	0.00	0.17	0.76
	RNO00100_STEROID_BIOSYNTHESIS	0.00	0.17	0.76
	RNO04142_LYSOSOME	0.00	0.16	0.76
	RNO04060_CYTOKINE-CYTOKINE_RECEPTOR_INTERACTION	0.00	0.17	0.81
	RNO04140_REGULATION_OF_AUTOPHAGY	0.00	0.19	0.86
	RNO05215_PROSTATE_CANCER	0.00	0.32	1.00
	RNO04660_T_CELL_RECEPTOR_SIGNALING_PATHWAY	0.00	0.34	1.00

^aThe pathway gene sets were extracted from KEGG pathway database and assessed using GSEA (gene set enrichment analysis). The GSEA can determine differentially expressed gene sets between two biological states or conditions. NOM-pval, FDR-qval and FWER-pval indicate nominal p-value, false discovery rate and family-wise error rate p-value, respectively.

The alterations in expression of Phase I, II, III drug-metabolizing enzymes are summarized in Table 2. Of the total of 728 probe sets (Phase I: 295, Phase II: 325, Phase III: 108), 161 genes demonstrated significantly altered expression levels in human hepatocytes. In rat hepatocytes, of the total of 409 probe sets (Phase I: 170, Phase II: 183, Phase III: 56), 57 genes were significantly altered. The alteration in expression of most drug-metabolizing enzymes was not severe up to 0.75% (v/v) DMSO in both human and rat cases.

Table 2.

Up- and down-regulated drug metabolizing enzymes in human and rat hepatocytes are shown^a

Expression	Gene Symbol
Strongly up-regulated	AGXT2L1, BHMT, CYP2C8, CYP39A1, CYP3A4, CYP3A5, CYP4F3, CYP7B1, OAT, TAT
Up-regulated	ABAT, ABCC5, ABCC9, ABCD1, ABCD3, ABCD4, ADH6, ALDH1L1, ALDH2, ALDH3A2, ALDH5A1, ALDH6A1, ALDH7A1, ALDH9A1, BCAT1, BCAT2, BHMT2, COQ5, CYB5A, CYB5R4, CYP1B1, CYP20A1, CYP27A1, CYP2A6, CYP2A7, CYP2B6, CYP2B7P1, CYP2E1, CYP2U1, CYP4A11, CYP4A22, CYP4F11, DPYD, EPHX1, FMO3, FMO5, FTSJD2, GPT2, GPX4, GSTM2, HGSNAT, LRTOMT, MAOB, MSRB2, N6AMT2, NAT12, NAT15, PCMTD1, PCMTD2, PRMT2, PSAT1, SETD7, SLCO4C1, TPMT, TXNDC12, ZADH2
Strongly down-regulated	ABCA1, ABCC3, ADH1B, ADH1C, ADH4, AKR1B10, AKR1D1, ALDH1A1, ALDH1B1, ALPL, CYP2C18, CYP2C9, HNMT, METT10D, METTL12, METTL8, METTL9, NAT11, NAT8, PRMT1, SLCO1B1, SLCO1B3, SULT1A1, SULT1B1, TRMT6, UGT1A6, UGT2A3, UGT2B15, UGT2B28, UGT2B4
Down-regulated	ABCB10, ABCB6, ABCC2, ABCC3, ABCC4, ABCC5, ADH5, AGXT2L2, AKR1A1, AKR1B1, AKR1C1, AKR1C4, AKR7A3, ALDH4A1, AOX1, ARD1A, AS3MT, ASMTL, COMTD1, COQ3, CYB5B, CYB5D1, CYB5R3, CYP51A1, DMAP1, DNMT1, FMO4, GPX1, GSTA1, GSTCD, GSTO1, HEMK1, HNMT, MAOA, METT5D1, METTL1, METTL11A, METTL13, METTL3, METTL5, MGST1, MTR, NAT10, NAT11, NAT13, NAT2, NAT5, NAT9, PEMT, PRMT5, RG9MTD1, RG9MTD2, RNMTL1, RRP8, SETD8, SHMT1, SULT1A1, SULT1A2, SULT2A1, TRDMT1, TRMT1, TRMT11, TRMT5, TRMT6, ZADH2
Up-regulated	Abca4, Abca5, Abcb1a, Aldh1a2, Aldh6a1, As3mt, Bhmt2, Ces3, Coq3, Cyp1a1, Cyp1a2, Cyp4a3, Cyp4f17, Cyp4f5, Gpx3, Gstm7, Gstt1, Gstt2, Nnmt, Pcmtd1, Pcmtd2, Slco3a1, Sult1a1, Tpmt, Trmt12
Down-regulated	Abcb11, Abcc2, Adh1, Akr1b8, Akr1c12, Akr1d1, Akr7a3, Ard1a, Bcat1, Comtd1, Cyp26b1, Cyp2b3, Cyp2c12, Cyp2c22, Cyp3a9, Fmo5, Gpt2, Gstm2, Gstt3, Maoa, Maob, Mett10d, Mettl2, Mgst2, N6amt1, Prmt1, Psat1, Rrp8, Slco1a1, Slco1b2, Ugt2b17, Ugt2b36

^aAll drug metabolizing enzymes are classified using the smoothing spline clustering¹¹. This method is able to classify a series data such as time series data. Several genes are shown multiple times due to redundant probe sets in the microarray.

Cytotoxicity of DMSO

Human cryopreserved hepatocytes and rat primary cultured hepatocytes were treated with DMSO for 24 hours at five different concentrations in order to examine its cytotoxic effects. In both human and rat hepatocytes, no toxic effects were morphologically observed up to the concentration of 2% (v/v) DMSO.

In addition, the cytotoxicity of DMSO was examined by the conventional LDH test. LDH activity response curves for DMSO in human and rat hepatocytes are shown in Figure 2. Again, DMSO did not show any toxic effects up to the concentration of 2% (v/v).

Figure 2.

Lactate dehydrogenase (LDH) activity response curves for dimethyl sulfoxide (DMSO) in human cryopreserved hepatocytes and rat primary cultured hepatocytes. The experiments were conducted with hepatocytes in three independent wells (n = 3). Standard error bars are shown.

Discussion

We have been focusing on identification of candidate biomarker genes to predict and/or diagnose toxicity in our project (TGP2, Toxicogenomics Informatics Project). Gene expression data using rat primary hepatocytes or human cryopreserved hepatocytes as well as gene expression data using rat liver are being analyzed. However, the effects of some chemicals were found to not be adequately reflected in the gene expression profile due to the low concentrations dictated by solubility in the set concentration of DSMO. The necessity to increase the DMSO concentration is the reason for the present study. The fact that there were not many probe sets with altered in expression up to 0.75% (v/v) DMSO is therefore very important.

In particular, there were remarkably few probe sets with altered in expression up to 0.5% (v/v) DMSO. As the concentration of DMSO increased, the number of probe sets with alteration and the magnitude of alteration increased. In human hepatocytes, there was only 1 probe set (histone cluster 1, H2bd; HIST1H2BD), which was upregulated (p < .05 and more than 2-fold) at a concentration of 0.1% (v/v) DMSO, 4 probe sets (HIST1H2BD, BTB [POZ] domain containing 11 [BTBD11], chemokine [C-X-C motif] ligand [CXCL11], and another with no annotation) at 0.5% (v/v) DMSO. And there were 3 probe sets (endoplasmic reticulum aminopeptidase 1 [ERAP1], natural killer-tumor recognition sequence [NKTR], and cleavage and polyadenylation factor subunit, homolog [PCF11]), which were down-regulated (p < 0.05 and less than 1/2) at a concentration of 0.1% (v/v) DMSO, 12 probe sets (insulin-like growth factor binding protein 3 [IGFBP3], dehydrogenase/reductase (SDR family) member 9 [DHRS9], LIM and cysteine-rich domains 1 [LMCD1], solute carrier family 13 (sodium-dependent citrate transporter), member 5 [SLC13A5], IGFBP1, thyroid hormone responsive [THRSP], interleukin 1 receptor antagonist [IL1RN], chromosome 10 open reading frame 108 [C10orf108], palmdelphin [PALMD], and choroideremia-like (Rab escort protein 2) [CHML]) at 0.5% (v/v) DMSO. In rat hepatocytes, there was no probe set, which was up-regulated (p < 0.05 and more than 2-fold) at a concentration of 0.1% (v/v) DMSO, 6 probe sets (Ccl2, radical S-adenosyl methionine domain containing 2 (Rsad2), Ccl20, Cxcl2, DEAD (Asp-Glu-Ala-Asp) box polypeptide 60 [Ddx60], and Cxcl10) at 0.5% (v/v) DMSO and there was only 1 probe set (with no annotation), which was down-regulated (p < 0.05 and less than 1/2) at 0.1% (v/v) DMSO, and 4 probe sets (Car3, D4, zinc and double PHD fingers, family 3 [Dpf3], LOC681825, and another with no annotation) at 0.5% (v/v) DMSO. CXCL11 was up-regulated and IGFBP4 was down-regulated in common, but neither gene is reported to have any relation with DMSO.

To determine whether an a priori defined set of genes shows statistically significant, concordant differences between control and DMSO treatment, GSEA was carried out. In human hepatocytes, a large gap was observed between 0.75% (v/v) and 1% (v/v) DMSO. And only 1 gene set was shared in common. In contrast, there were 3 gene sets in common between 0.1% (v/v) and 0.5% (v/v) DMSO, 4 gene sets in common between 0.1% (v/v) and 0.75% (v/v) DMSO, and 4 gene sets in common between 0.5% (v/v) and 0.75% (v/v) DMSO. In rat hepatocytes, large gaps were observed between 0.1% (v/v) and 0.5% (v/v) DMSO and between 0.1% (v/v) and 0.75% (v/v) DMSO. There was only 1 gene set in common in each gap. In contrast, 6 gene sets were in common from 0.5% (v/v) to 1% (v/v) DMSO (Table 1).

With regard to Phase I, II, III drug-metabolizing enzymes, DMSO concentration-dependent probe sets were clustered with the smoothing spline clustering method, and 161 genes in human cryopreserved hepatocytes and 57 genes in rat primary cultured hepatocytes were altered in expression level, respectively (Table 2). However, the magnitude of alteration for expression of most drug-metabolizing enzymes was within 1 standard deviation in both cells.

In human hepatocytes, there were 66 enzymes, which were up-regulated, and 95 enzymes, which were down-regulated. Of 66 enzymes, 10 enzymes including CYP3A4 and CYP3A5, were strongly up-regulated enzymes and 56 enzymes including CYP1B1, CYP2A6 and CYP2E1 were moderately up-regulated. Of 95 enzymes, 30 enzymes including UGT1A6 were strongly down-regulated enzymes and 65 enzymes were moderately down-regulated. Nishimura et al. earlier investigated the gene expression of CYP1A1, CYP1A2, CYP1B1, CYP2A6, CYP2E1, CYP3A4, CYP3A5, CYP3A7, UGT1A6, UGT1A9 and ABCB1 after 24 hours of exposure to 0.1% (v/v), 0.5% (v/v), and 2.5% (v/v) DMSO in primary culture of human hepatocytes.⁶ Our results were in excellent agreement with their findings. Wilkening and Bader also reported the response after 24 hours exposure to 0%-1% (v/v) DMSO in primary human hepatocytes.⁷ They found CYP3A4 but not CYP3A7 to be induced by DMSO and this also corresponds with our data. Furthermore, Choi et al. reexamined induction of CYP1A1, CYP1A2, CYP2B6, CYP2C8, CYP2C9, CYP2C19, CYP2D8, CYP3A4, CYP3A5, UGT1A1, UGT1A4, UGT1A6, UGT1A9, and UGT2B7 after exposure to 1% (v/v) DMSO over 20 days in Huh7 cells, which was established from a 57-year-old male with a well-differentiated hepatocellular carcinoma in 1952.⁸ In their report, all of the examined enzymes, expect for CYP1A1, showed significant increase in their expression by DMSO treatment. Since their experimental condition was very different from our own, we are not able to make an easy comparison, but CYP2B6, CYP2C8, CYP3A4, and CYP3A5 were induced by DMSO in both data sets. On the other hand, there is a discrepancy between the two concerning CYP2C9 and UGT1A6. In rat hepatocytes, there were 25 enzymes which were up-regulated and 32 enzymes which were down-regulated. Some enzymes including betaine-homocysteine methyltransferase 2 (Bhmt2), protein-L-isoaspartate (D-aspartate) O-methyltransferase domain containing 2 (Pcmtd2), aldo-keto reductase family 1, member D1 (delta 4-3-ketosteroid-5-beta-reductase; Akr1d1), and methyltransferase 10 domain containing (Mett10D) were up-regulated or down-regulated in both human and rat hepatocytes. The magnitude of change of these transcripts in rat hepatocytes was often larger than that in human hepatocytes.

As for housekeeping genes, Nishimura et al. examined the effects of DMSO on the expression of Actb, Gapdh, Gusb, Pgk1, Ppia, and Tfrc mRNA in cultures of C2C12 myotubes.⁵ They reported that Actb, Pgk1 and Tfrc were significantly (p < 0.05 and less than 1/2) decreased at 2.5% (v/v) DMSO after 24 hours. At 2% (v/v) DMSO, we found the above 6 genes to show similar expression change as their results.

To confirm the gene expression data using microarray, we conducted quantitative RT-PCR for some rat genes (Ccl20, Sult1a1, Cyp1a1, Car2, and Cyp2c12). Ccl20, which was up-regulated (p < 0.05 and 2.13-fold compared with control) in microarray analysis at a concentration of 0.5% (v/v) DMSO, showed up-regulation (dCt = −0.67) in expression in using quantitative RT-PCR. It was also confirmed that Sult1a1 and Cyp1a1 were up-regulated in quantitative RT-PCR analysis (dCt = −0,09 and dCt = −0.26, respectively) as well as in microarray analysis (1.59-fold and 1.84-fold) at 2% (v/v) DMSO. On the other hand, Car2 and Cyp2c12 showed down-regulation in both quantitative RT-PCR analysis and microarray analysis (eTable). The Quantitative RT-PCR data were well coincident with the microarray data.

As regards cytotoxicity, no toxic effects were observed up to the concentration of 2% (v/v) in either of the cells with either of the tests employed.

In conclusion, we showed that there are only very few probe sets altered in expression at doses up to 0.75% (v/v) DMSO in both human and rat hepatocytes. In particular, there are remarkably few probe sets altered in expression up to 0.5% (v/v) DMSO and the magnitude of alteration for expression of most drug-metabolizing enzymes was within 1 standard deviation in both cells. Alteration of housekeeping genes was also very small. Furthermore, DMSO did not show any toxic effects up to the concentration of 2% (v/v) and there are few differences in effects of DMSO between human and rat hepatocytes with regard to change of transcripts. These results in human cryopreserved hepatocytes and rat primary cultured hepatocytes suggested that a DMSO concentration up to 0.5% (v/v) can be tolerated, although care must be taken in interpretation at higher concentrations.

Footnotes

This study was supported in part by a grant from the Ministry of Health, Labor and Welfare (H19-Toxico-001).

References

Urushidani

Prediction of hepatotoxicity based on the toxicogenomics database. In: Sahu

(ed.) Hepatotoxicity from Genomics to In Vitro and In Vivo Models. Hoboken: Wiley & Sons, 2007; 507–529.

Adler

Paparella

Pellizzer

Hartung

Bremer

The detection of differentiation-inducing chemicals by using green fluorescent protein expression in genetically engineered teratocarcinoma cells. Alternatives to Laboratory Amimals. Chicago: ATLA, 2005; 33: 1–13.

Klinken

Holmes

Morse

3rd Thorgeirsson

. Transcriptional and post-transcriptional regulation of c-myc, c-myb, and p53 during proliferation and differentiation of murine erythroleukemia cells treated with DFMO and DMSO. Exp Cell Res 1988; 178: 185–198.

Jiang

Tang

Wang

Zhang

Down-regulation of TRRAP-dependent hTERT and TRRAP-independent CAD activation by Myc/Max contributes to the differentiation of HL60 cells after exposure to DMSO. Int Immunopharmacol 2006; 6: 1204–1213.

Nishimura

Nikawa

Kawano

Nakayama

Ikeda

. Effects of dimethyl sulfoxide and dexamethasone on mRNA expression of housekeeping genes in cultures of C2C12 myotubes. Biochem Biophys Res Commun 2008; 367: 603–608.

Nishimura

Ueda

Naito

Effects of dimethyl sulfoxide on the gene induction of cytochrome P450 isoforms, UGT-dependent glucuronosyl transferase isoforms, and ABCB1 in primary culture of human hepatocytes. Biol Pharm Bull 2003; 26: 1052–1056.

Wilkening

Bader

. Differential regulation of CYP3A4 and CYP3A7 by dimethylsulfoxide in primary human hepatocytes. Basic Clin Pharmacol Toxicol 2004; 95: 92–93.

Choi

Sainz

Jr Corcoran

Uprichard

Jeong

. Characterization of increased drug metabolism activity in dimethyl sulfoxide (DMSO)-treated Huh7 hepatoma cells. Xenobiotica 2009; 39: 205–217.

Uehara

Kiyosawa

Hirode

Omura

Shimizu

Ono

Gene expression profiling of methapyrilene-induced hepatotoxicity in rat. J Toxicol Sci 2008; 33: 37–50.

10.

Uehara

Kiyosawa

Shimizu

Omura

Hirode

Imazawa

Species differences in coumarin-induced hepatotoxicity as an example of how toxicogenomics help assessing risks for human. Human Exp Toxicol 2008; 27: 23–35.

11.

Castillo-Davis

Zhong

Liu

. A data-driven clustering method for time course gene expression data. Nucleic Acids Res 2006; 34: 1261–1269.

12.

Subramanian

Tamayo

Mootha

Mukherjee

Ebert

Gillette

Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expressionprofiles. Proc Natl Acad Sci USA 2005; 102 (43): 15545–15550.

13.

Kanehisa

Goto

Kawashima

Okuno

Hattori

. The KEGG resource for deciphering the genome. Nucleic Acids Res 2004; 32: D277–D280.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB

Effects of DMSO on gene expression in human and rat hepatocytes

Abstract

Keywords

Introduction

Materials and methods

Cell culture and in vivo studies

Gene expression analysis

Quantitative RT-PCR assay validation

Measurement of LDH activity

Data analysis

Results

Gene expression analysis

Cytotoxicity of DMSO

Discussion

Footnotes

References

Supplementary Material