ToxSci Advance Access originally published online on March 30, 2007
Toxicological Sciences 2007 97(2):582-594; doi:10.1093/toxsci/kfm067
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
The Contribution of Methotrexate Exposure and Host Factors on Transcriptional Variance in Human Liver
* Center for Molecular Medicine
Department of Rheumatology, University of Connecticut Health Center, Farmington, Connecticut 06030
Boehringer-Ingelheim Pharmaceuticals, Inc., Ridgefield, Connecticut 06877
Department of Dermatology, University of Connecticut Health Center, Farmington, Connecticut
¶ Department of Chemical Engineering, University of Connecticut, Storrs, Connecticut 06269
|| Division of Gastroenterology and Hepatology, University of Connecticut Health Center, Farmington, Connecticut
1 To whom correspondence should be addressed: DWR, The Center for Molecular Medicine, University of Connecticut Health Center, 263 Farmington Avenue, Farmington, CT 06030-3101. Fax: (860) 679-7639. E-mail: rosenberg{at}nso2.uchc.edu or GYW, Gastroenterology, 263 Farmington Avenue, Farmington, CT 06030-1845. E-mail: wu{at}nso.uchc.edu.
Received December 21, 2006; accepted March 14, 2007
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAREFERENCES |
|---|
Long-term administration of methotrexate (MTX) for management of chronic inflammatory diseases is associated with risk of liver damage. In this study, we examined the transcriptional profiles of livers from patients treated with MTX. The possibility that expression signatures correlate with grade of fibrosis or underlying rheumatic disease was evaluated. Twenty-seven patients taking MTX were accrued for this study. Ten non-MTX–exposed normal liver specimens were used as controls. Global mRNA expression was assayed using oligonucleotide arrays. A total of 205 genes were significantly altered in MTX-exposed livers. Six of these genes were validated by qPCR. Two genes, CLN8 and ANKH that map to chromosomal locations previously associated with rheumatoid arthritis, were found to be elevated in MTX-exposed samples. Subsequent pathway analysis indicates that MTX exposure is associated with the following key alterations: (1) upregulation of lipid biosynthetic genes, consistent with MTX-induced steatosis, (2) downregulation of proinflammatory chemokines, consistent with the anti-inflammatory effects of MTX, and (3) elevation of complement pathway gene expression. Complement 5, shown earlier to be correlated with liver fibrosis in mice, was found to be elevated (twofold) in MTX-exposed livers. In conclusion, we have found the expression of a number of genes associated with rheumatic disease and/or MTX exposure to be significantly different. Differences in complement expression provide the rationale for future correlative studies between MTX-induced liver fibrosis and C5 alleles in order to identify patients with increased risk for fibrosis.
Key Words: methotrexate; fibrosis; rheumatic disease; steatosis; complement; liver.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAREFERENCES |
|---|
Methotrexate (MTX), a folic acid antagonist, has been used for several decades in the management of various chronic inflammatory diseases, including rheumatoid arthritis (RA) and psoriasis. However, long-term administration of MTX is associated with an increased risk of liver damage (Richard et al., 2000
). Histological abnormalities include nonspecific pathologies such as fatty liver changes, focal liver cell necrosis, portal tract inflammation, and a more specific fibrosis associated with collagen accumulation (Richard et al., 2000). Approximately 3% of RA patients on low-dose MTX develop severe liver fibrosis (Beyeler et al., 1997; Whiting-O'Keefe et al., 1991). On the other hand, psoriatic patients have a much more substantial risk for MTX-induced liver fibrosis, with reported incidences ranging from 8 to 23%, leading to recommendations for assessment of liver damage after every 1.5 g of total intake (Boffa et al., 1995; Kremer, 2002; Malatjalian et al., 1996; Whiting-O'Keefe et al., 1991). The cause for increased risk of liver fibrosis in psoriatics is unknown.
Several possible phenomena contribute to differential susceptibility for liver fibrosis. It is possible that there may be interindividual difference in MTX metabolism that affect MTX accumulation and toxicity (Ahern et al., 1991). Alternatively, exposure to MTX may produce a comparable acute toxicity in patients, but individual response may not be uniform, resulting, e.g., in variable inflammation. These mechanisms may be further aggravated by additional factors such as alcohol intake. Identification of the cellular changes leading to fibrosis may help predict risk or presence of fibrosis. Underlying mechanisms that contribute to MTX-induced liver damage are incompletely understood. It is possible that MTX-induced hepatocyte cell death leads to macrophage stimulation, hepatic stellate cell activation, and subsequent excretion of extracellular matrix (for review see Friedman 2004). Differentiation of hepatic stellate cells into myofibroblasts strongly correlates with fibrosis observed in human livers and in animal models (Levy et al., 2002; Lewindon et al., 2002; Nakatsukasa et al., 1990; Roderfeld et al., 2006; Takahara et al., 1988). In the rat, chronic exposure to MTX has been shown to result in hepatocyte necrosis, cell death, and Kupffer cell enlargement (Hall et al., 1991). How these effects result in mild fibrosis in some patients and severe fibrosis in others is unknown.
Whatever the cause of fibrosis, diagnosing its presence is essential. Importantly, the risk for development of liver disease is not often predicted with the use of standard liver enzymes (Richard et al., 2000), and thus, patients receiving MTX must be monitored routinely by examination of liver biopsies. Potential alternatives to biopsy have been proposed but have not been validated for MTX-induced fibrosis (Cales et al., 2005; Chalmers et al., 2005; Imbert-Bismut et al., 2004; Maurice et al., 2005; Rossi et al., 2003).
In the following study, we have examined transcriptional profiles in human liver biopsy specimens obtained from patients who have been treated for varying lengths of time with MTX. We hypothesized that MTX-related fibrosis may elicit a specific hepatic gene expression profile. Such changes may provide a better understanding of the genes involved in the development of liver damage as a result of MTX treatment. Furthermore, if gene targets can be identified that encode secreted proteins, the information may serve as the basis for development of a blood test for MTX-related fibrosis. In this report, we have performed global transcriptional analyses on liver biopsies from MTX-treated patients in part to assess the value of this approach. In the data reported here, we have identified and validated several genes the expression of which are associated with either chronic or acute MTX exposures.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAREFERENCES |
|---|
Patient population and MTX treatment.
Prior to biopsy, all patients consented to participate in the study in accordance with University of Connecticut Health Center (UCHC) Institutional Review Board policies and procedures. Age, sex, cumulative MTX dose, number of years on therapy, and comorbidity information were determined. In addition, whenever possible, patients were requested to take their dose of MTX at a specified time (2 h) prior to liver biopsy.
Procurement and handling of liver biopsy specimens.
Liver biopsy specimens were obtained from patients with rheumatic disease or psoriasis at the UCHC John Dempsey Hospital (JDH). Immediately following biopsy, the core was ejected into a culture dish containing 2 ml of 25°C saline. If sufficient tissue was present, a portion of the specimen was removed and snap-frozen in liquid nitrogen, stored at – 80°C, and then processed for RNA extraction as described below. Biopsies were 0.5–1.5 cm long with a diameter of 1.4 mm. Unfrozen tissue was formalin fixed, paraffin embedded, and sectioned for routine pathological examination conducted by board-certified pathologists at JDH. Total RNA from normal human livers was obtained from Ambion (Austin, TX), Clontech (Mountain View, CA), and Biochain (Hayward, CA). The normal group of livers consisted of one female and nine males ranging in age from 24 to 70 years.
Pathologic criteria for assessment of liver damage.
Grade I encompassed normal liver histology, including the presence of mild lobular hepatitis, mild focal fatty metamorphosis, and areas of Kupffer cell hyperplasia without increased portal, pericellular, or central fibrosis (Roenigk et al., 1982). Grade II pathology required the presence of fibrous spurs in portal area, patchy areas of fibrosis within the space of Disse, or occasional thickened liver cords, but without fibrotic septa extending into the lobule. Kupffer cell hyperplasia, mild to moderate steatohepatitis, and spotty hepatocellular necrosis may also be present to a varying extent. Grade IIIA liver pathology was represented by mild portal fibrosis, with fibrotic septa extending into the lobule, but without bridging. Grade IIIB represented spotty necrosis or moderate-to-severe septal fibrosis with portal-to-portal or portal-to-central vein bridging. Grade IV cirrhosis was defined by bridging with regenerating nodules.
RNA isolation, amplification, and hybridization.
Total RNA and protein from liver core specimens were extracted using Trizol reagent (Invitrogen, Carlsbad, CA). All RNA samples were subjected to electrophoresis, and recovery was assessed by evaluation of intact 28s and 18s ribosomal RNA bands. Only samples with prominent 28s and 18s ribosomal RNA were used. T7-based RNA amplification (RiboAmpTM RNA amplification kit, Molecular Devices [formerly Arcturus], Sunnyvale, CA) (Baugh et al., 2001) was used to generate adequate amounts of antisense RNA according to the manufacturer's protocol with the following modifications. During in vitro transcription, 25mM ATP, CTP, and GTP with 12.5mM UTP and 12.5mM aminoallyl-UTP were substituted for the IVT Master Mix. Mono-reactive Cy3 or Cy5 dye (Amersham Pharmacia, Piscataway, NJ) was conjugated to the aminoallyl-labeled RNA (t Hoen et al., 2003). Residual dye was removed using RNeasy RNA purification columns (Qiagen, Valencia, CA). A universal reference RNA (Clontech, Universal Reference RNA) was amplified and conjugated with Cy5 for all samples. The fluorescently labeled RNA was hybridized to OHU21K human 21K oligo-arrays (Yale Keck Facility, New Haven, CT) using their recommended protocol. Information regarding the clone set and array preparation can be obtained from http://keck.med.yale.edu. Each sample was run in duplicate or triplicate. Slides were scanned using a Perkin-Elmer ScanArray Express. Images were quantified using BioDiscovery software (El Segundo, CA). Spot and background intensities were measured and poor or empty spots recorded. Prior to analysis, poor and empty spots were removed and data subjected to standard methods of local background correction and Lowess normalization.
Statistical, functional, and pathway analyses.
Regulated genes were identified using the significance analyzer function in Genesight (Biodiscovery) and significance analysis of microarrays (Tusher et al., 2001). Differentially expressed genes were also identified using a random-variance model–based t-test (Wright and Simon, 2003). The most significantly regulated genes appeared using all three methods (data not shown). The p values generated by the Genesight Student's t-test (ST) were used to rank genes, and those most significantly different among partitions were chosen for further investigation. Genes with p values greater than 0.05 using the Bonferroni correction were not investigated unless part of a functional group of regulated genes. Cluster analysis was performed in Genesight with a K-means algorithm using a euclidian distance metric. Gene Ontology analysis was performed using the BioScript Library tool on GeneSpring (Agilent, Santa Clara, CA) with application of the Bonferroni correction. Genes were classified according to their annotated role in biological processes and molecular function from Gene Ontology (The Gene Ontology Consortium). Correlation of cumulative dose with gene expression was accomplished by calculating the Pearson product-moment correlation coefficient for each gene.
Pathway analysis was performed using Ingenuity Pathway Analysis (IPA) software (Ingenuity Systems, Redwood City, CA). Genes with uncorrected p values (using Genesight Significance Analyzer) of 0.05 or less were used as genes available for inclusion into functional groups for this nonstringent analysis. Global functional analysis and global canonical pathways were performed using the genes on the array as a reference, and p values for functional groups were calculated using a right-tailed Fisher Exact Test (FET) as indicated in the text.
Real-time PCR.
Validation of changes identified by the microarrays was accomplished by real-time PCR (qPCR) on unamplified RNA. A 1:1 ratio of oligo-dT and random primers was used with 0.5 µg of total RNA and reverse transcribed with Superscript II (Invitrogen) according to the manufacturer's protocol. An ABI 7500 real-time machine was used to perform qPCR with iTaq SYBR Green (Biorad, Hercules, CA). All reactions were routinely performed in duplicate. Sufficient sample template was added to obtain a Ct value of less than 32. PCR was performed for 40 cycles followed by a disassociation step to verify the amplicon. A standard curve was generated (total RNA input vs. Ct) from serially diluted samples represented on the same plate. Array data were used to select the housekeeping gene, cyclophilin A (PPIA) (Blanquicett et al., 2002), a selection based on its low variability and high expression in all samples (data not shown). PPIA primers are as follows: F-5'CCACCAGATCATTCCTTCTG3', R-5'AGGAAAACATGGAACCCAAA3'. Ratios to PPIA were taken and, where appropriate, ST applied to compare means. Table 4 lists all primer sets used for qPCRs.
|
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAREFERENCES |
|---|
Patient Information and Liver Pathology
A total of 27 patients were accrued for this study (Table 1), representing 18 psoriatic and 9 nonpsoriatic patients. One patient provided multiple biopsies over a 3-year period. As described in Table 1, liver pathology ranged from Roenigk Grade I (n = 12), Grade II (n = 14), or Grade III (n = 3). Exposure to MTX occurred for a minimum of 2 to a maximum of 17 years. Cumulative dose of MTX ranged from 1.5 to 12.6 g. Patients displayed various rheumatic diseases, including psoriasis, lupus, RA, psoriatic arthritis, UDCTD/CREST (undifferentiated connective tissue disease/calcinosis, Raynaud's phenomenon, esophageal dysmotility, sclerodactyly, and telangiectasia), polymyositis, or Sjogren's syndrome (Table 1).
|
Partition by MTX Exposure/Rheumatic Disease
In the following analysis, samples were partitioned into two groups, MTX exposed and nonexposed. Twenty-nine MTX-exposed samples were compared to 10 control samples.
Functional pathway analysis using IPA identified two obvious trends occurring within the livers of MTX-treated individuals. The first was an upregulation in 35 of 42 fatty acid modification genes (Table 2, p = 0.000079, FET). Similarly, canonical pathway analysis showed elevations in genes associated with fatty acid elongation in mitochondria (Table 2, p = 0.0055, FET, see supplementary Fig. 1 for pathway diagram). Pathology reports indicated that at least 70% of biopsy samples displayed the fatty metamorphosis commonly associated with MTX hepatotoxicity (Fig. 1). The elevation of an acyltransferase in MTX-exposed livers, AGPAT3, which converts lysophatidic acid into diacylglycerol 3-phosphate, was validated by qPCR (Table 4). PPAP2B (NM_003713 [GenBank] ), a phosphatase which catalyzes the next step in the formation of diacylglycerol, was also upregulated 1.3-fold by MTX (p = 0.002, ST). In addition, pathway analysis identified a group of genes associated with triacylglycerol synthesis as upregulated (Table 2, p = 0.012, FET).
|
|
The second trend observed with pathway analysis was downregulation of genes involved in monocyte/T lymphocyte activation and movement, with 19 of 23 genes decreased in expression (Table 3), suggesting suppression of immune response. When expression was compared to fibrosis by Pearson correlation, CCL4 and IL8RB were weakly correlated to fibrosis, mostly due to higher values in samples unexposed to MTX (Fig. 2). However, several important mediators of inflammation, including complement factors C3, C5, and C8a, were elevated (Table 3), consistent with activation of the alternative complement pathway in patients exposed to MTX. In general, there was high variability in mRNA expression of complement, differing by as much as 20-fold between patients (Fig. 2).
|
|
We examined genes not associated with canonical pathways. Using gene-by-gene ST, 205 genes were identified with corrected p values less than 0.05 when livers from MTX-exposed individuals were compared to unexposed livers (supplementary Table 1). A heat map of the top 17 regulated genes (based on p values) is shown in Figure 3, displaying correct segregation into MTX-exposed and unexposed after clustering. Mutations in two of these MTX-upregulated genes, ceroid lipofuscinosis neuronal 8 (NM_018941 [GenBank] ), and ankylosing spondylitis homolog (ANK; NM_054027 [GenBank] ), have been associated with disease (McKee et al., 2004
; Santavuori, 1988). Mutations in ANK are also associated with rheumatic disease in mice (Ho et al., 2000). The expression levels of these two genes were further validated by qPCR (Table 4).
|
Several other notable genes altered in MTX-exposed samples were identified by ontology analysis (supplementary Table 3). Cytosolic sulfotransferase 2A (HST), a member of the organic acid catabolism group, is a phase II drug-metabolizing enzyme that is involved in bile acid sulfation (Comer et al., 1993
; Falany et al., 1995) and was found to be elevated 
2.3-fold in MTX-exposed livers. dehydroepiandrosterone sulfotransferase activity has already been reported upregulated in Hep G2 cells exposed to MTX (Chen et al., 2005). CYP4AH1 was 1.8-fold higher in the livers of MTX-exposed patients. Cyp4a proteins have been shown to be upregulated in mouse steatohepatitis but little is known about CYPAH1 (Deng et al., 2005; Li et al., 2004). We also found lower expression of two transcripts encoding the extracellular proteins, undulin and mucin 5. Circulating undulin has been proposed as a marker of liver fibrosis (Schuppan, 1991). However, our data do not support the use of elevated undulin as a marker of MTX-induced liver damage.
Partition by Acute versus Nonacute MTX Exposures
Patients were separated by those biopsied within 20 h of taking MTX (n = 12) versus those who had not received MTX for at least 4 days (n = 12). When genes were analyzed on a gene-by-gene basis, only S100A11 (calcium-binding protein, calgizzarin; NM_005620
[GenBank]
) was found to be significantly downregulated by 1.8-fold following acute MTX exposure. Regardless of the timing of MTX exposure, unexposed livers had higher S100A11 expression levels (1.7-fold difference) compared to controls. As shown in Table 4, changes in S100A11 mRNA levels were validated by qPCR.
Pathway analysis showed highly significant downregulation of immune response genes in liver samples exposed to MTX within 20 h (Table 4). Complement and lipid synthesis genes, however, were not significantly different in the acute/subacute partition (Table 5).
|
Partition by Psoriatic versus Nonpsoriatic
Pathway analysis was done on all samples, comparing samples from psoriatics (high risk for fibrosis) to samples from patients without psoriasis (low risk). The complement pathway was the most significantly regulated canonical pathway (Fig. 4, p = 0.00026, FET) and found to be elevated in psoriatics. Mean fold changes for many of the genes are small. However, individual variability in complement genes is high (Fig. 2), resulting in some patients with much higher fold changes. In this comparison, both MTX-exposed and nonexposed samples occur in the nonpsoriatic group, suggesting that the general elevation in the complement pathway is associated with the presence of psoriasis, not exposure to MTX.
|
Other Partitions
Sex differences were used as a means to provide internal validation of the data. This approach has been used successfully in previous gene expression arrays studies (Whitney et al., 2003
). Patients were partitioned by sex, and gene-by-gene STs done. Genes were ranked by p value, and the number of sex-linked genes examined. As shown in Figure 5, eight of the top 10 genes on this list were located on the sex chromosomes. A heat map of the 15 most regulated genes allowed samples to be clustered by sex (Fig. 5). The conservative Bonferroni correction for multiple testing indicates a p value of 2.34 x 10–6 as the cutoff for significance at the 0.05 level. Eleven of the genes on this list meet this cutoff and the top six genes had p values that were 10 orders of magnitude less than 2.3 x 10–6 (Fig. 5), showing a robust discovery of genes regulated by sex in human liver. Seven of these transcripts (supplementary Table 4) have previously been reported to differ among men and women (Radich et al., 2004; Rinn et al., 2004; Whitney et al., 2003). One sex-linked gene, SMCY homolog (NM_004653
[GenBank]
), is excluded by the Bonferroni correction and thus may be an example of a type II error. Several nonsex-linked genes were within the significance range, suggesting that type I errors will be commonly seen at p values greater than 3 x 10–7 when ST is used on individual genes.
|
Other partitions of the data failed to identify changes in gene expression. For example, comparison of patients with Grade I to Grade II fibrosis resulted in no genes with corrected p values less than 0.05, and mRNA expression profiles were unable to predict grade of fibrosis (data not shown). Patients were also partitioned by age, resulting in no significantly correlating genes identified, suggesting that age is not a strong confounding factor. When the effect of cumulative dose of MTX was compared to individual mRNA expression by correlation analysis, one gene, ENAH (NM_018212 [GenBank] ), was identified with a Pearson coefficient greater to or equal to |0.8|. ENAH encodes a cytoskeleton regulatory protein (Barzik et al., 2005
; Gertler et al., 1996) that was positively correlated with cumulative dose of MTX (supplementary Fig. 2). Because we are unable to assign a confidence value to this result, an independent validation for this gene is required. |
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAREFERENCES |
|---|
While the majority of patients exposed to MTX develop only mild fibrosis, others progress to more severe disease. Our objective in this study was to compare unexposed and exposed samples from different patients in order to discover common changes that might be responsible for this variation. We recognize that gene expression variability among patients is a limiting factor for this study and that the variability may relate to risk. Also, the cause of transcript-level differences between patients unexposed and exposed to MTX may have been due to MTX, or the presence of rheumatic disease.
As shown in Table 2, 35 genes involved in fatty acid synthesis were upregulated in the MTX-exposed samples, consistent with frequent steatosis associated with MTX exposure. Fatty metamorphosis is a hallmark of MTX-induced hepatotoxicity, but unlike advanced fibrosis, the finding of steatosis in biopsies does not require cessation of MTX therapy. We conclude that in early MTX-induced liver disease in humans, gene expression related to fatty change was present but gene expression related to fibrotic change was difficult to detect. Although many of the fold changes are small in the genes used for IPA, they may still be significant. Cell type heterogeneity in liver tissue may cause dilution of fold change when total liver RNA is extracted as done for this study.
Pathway analysis also showed a group of 12 chemokines downregulated by MTX exposure (Table 3), consistent with the general anti-inflammatory effects of MTX. Downregulation of chemokine CCL4 (macrophage inflammatory protein-1ß) in livers from patients chronically exposed to MTX was surprising as patients with liver disease have been reported to have high serum levels of CCL4 (Leifeld et al., 2003). However, in those cases increased serum CCL4 was associated with fulminant hepatic failure and uremia, severe conditions not present in our study (Leifeld et al., 2003; Pawlak et al., 2004). Because CCL4 can be detected in serum, further research is needed to determine if CCL4 can be used to distinguish between mild and severe fibrosis.
Our data show that MTX-induced liver inflammation may be mediated by the complement pathway because C3, C5, and C8a were found elevated in MTX-exposed samples (Fig. 2). If complement proteins are cleaved in the liver, localized inflammation may result. Complement is produced by hepatocytes in high amounts and secreted into blood. Activation of the complement pathway results in a cascade of enzymatic events producing amplification of response. Thus, the pathway must be tightly controlled (see Qin and Gao for review; Qin and Gao, 2006). Complement pathway activation results in neutrophil chemotaxis, leukocyte stimulation, cell lysis, and phagocytosis.
We compared patients who took MTX shortly (within 20 h) before biopsy with those who had not taken MTX for 4 days. Functional pathway analysis indicated a more significant downregulation of immune response genes in acute MTX exposure than the downregulation seen in MTX-exposed versus never-exposed controls (Table 5). This suggests that MTX was responsible for the downregulation rather than the underlying disease/genetic factors. Complement genes, however, were found to be most significantly regulated when patients were grouped by risk of fibrosis (Fig. 4), suggesting that expression differences may be attributed to disease/genetic factors.
At least two genes identified in the current study are already known to be associated with disease, CLN8 and ANKH. It remains to be determined if the RNA levels of these genes are modulated by MTX, rheumatic disease, or actually contribute to rheumatic disease phenotype. CLN8, found elevated in MTX-exposed samples, is expressed in a variety of tissues, as well as liver (Lonka et al., 2005). Mutation of CLN8 can cause neuronal ceroid lipofuscinosis, an autosomal recessive neurodegenerative disease (Weimer et al., 2002). A truncating mutation in mouse Cln8 resulted in a model for these diseases, characterized at the cellular level by accumulation of autofluorescent storage bodies in a variety of cells including liver (Faust et al., 1994). Ceroid lipofuscinosis has also been found associated with alcoholic liver fibrosis (Kishi et al., 1996).
Ankylosing spondylitis is a rheumatic disease associated with the ANKH gene in human and the mouse homolog Ank (Ho et al., 2000; Pendleton et al., 2002). ANK protein is involved in cellular export of inorganic pyrophosphate (PPi), a potent inhibitor of calcification (Fleisch et al., 1965). Disruption of ANKH potentiated calcification of cartilage and thus increased the risk for spondylarthropy. The liver releases a substantial portion of circulating PPi (Rachow and Ryan, 1988). Our data indicated that ANKH is elevated in MTX-exposed liver (Table 4). Although further validation at the protein level is needed, the identification of this gene associated with MTX exposure may lead to a better understanding of the normal function of ANKH in liver. Further investigation is needed to determine if ANKH exports PPi generated by triacylglycerol synthesis and whether increased circulating PPi through upregulation of ANKH contributes to the antiarthritic effect of MTX.
The downregulation of calgizzarin, a calcium-binding protein, in patients with recent MTX exposure, indicates that calgizzarin regulation may be affected by MTX exposure independent of the underlying inflammatory disease. Calgizzarin is one of several proteins that has been found to be upregulated in rat liver by chronic MTX exposure, suggesting a prominent role in fibrosis (Kristensen et al., 2000). None of the genes usually associated with fibrosis, such as ACTA2 or TGFß, were found upregulated when partitioning between acute and nonacute exposure, suggesting that 2–20 h of in vivo exposure to MTX is insufficient time for upregulation of these genes.
Several investigators have conducted single-nucleotide polymorphism (SNP) profiling on genes in the MTX pathway, such as methylenetetrahydrofolate reductase, reduced folate carrier, or thymidylate synthase, and have found association of specific alleles with efficacy of MTX in RA (Dervieux et al., 2004; Hughes et al., 2006; Urano et al., 2002; Wessels et al., 2006). Unfortunately, these studies did not compare grade of fibrosis with allele frequencies, so the possibility of genetic predisposition to fibrosis remains unclear. Although little effort has been made in discovering genes responsible for differential susceptibility to MTX-induced liver fibrosis, loci associated with the underlying inflammatory disease necessitating the use of MTX have been well characterized. Linkage studies have demonstrated a strong association of HLA-DRB1 with RA (Deighton et al., 1989; Wordsworth and Bell, 1991). The HLA locus (6p21) is estimated to account for approximately 40% of the genetic component of RA heritability (MacGregor et al., 2000). Other non-HLA loci may also contribute to RA susceptibility (Jawaheer et al., 2003; John et al., 2006). The individual genes at non-HLA loci responsible for risk have not been identified. In fact, Risch and Merkangas (1996) have calculated that depending on allele frequency, hundreds to thousands of affected sibling pairs are required to detect loci associated with modest genetic risk. As a possible companion strategy to large linkage studies, global transcript expression analysis may help identify the responsible genes in loci associated with rheumatic disease.
We investigated whether any of the genes identified by expression analysis appear near loci linked to RA. CLN8 is located at 8p23, in a region with possible linkage to RA (John et al., 2006). The microsatellite marker D8S277 used in the screen for this region is 4.8 Mb away from CLN8. This proximity warrants consideration of CLN8 as a candidate gene for 8p23. ANKH, located at 5p15.1, may also be close enough (5.5 Mb) to a possible linked locus at 5p15.31 for investigation (Jawaheer et al., 2003). We believe it is unlikely these two genes are associated with regions identified in linkage analysis purely by random chance, and the data suggest that modified expression of these two genes may be due to the presence of rheumatic disease, not MTX exposure. Alleles of ANKH associated with ankylosing spondylitis have already been proposed (McKee et al., 2004; Tsui et al., 2003), and our data provide support for its involvement in generalized rheumatic disease. We believe SNP analysis for CLN8 and ANKH done on families with RA would be interesting. Given the 10-cM resolution used in the large linkage studies done for RA and the modest genetic relative risks for non-HLA–associated loci, mRNA expression analysis may be a useful method for identifying genes conferring risk to rheumatic disease.
Our results indicate that with a relatively small number of heterogeneous samples, genes regulated by the presence of rheumatic disease or MTX exposure can be identified. No significant association was found between fibrotic grade and gene expression. This study was limited by the number of Grade III liver samples available. Therefore, inclusion of greater numbers of Grade III patients is needed to identify gene expression modulated between mild and severe fibrosis. Genes associated with lipid metabolism were found upregulated in samples with MTX exposure, consistent with fatty metamorphosis commonly seen in these patients. We found higher expression of complement genes in MTX-exposed samples consistent with a previous report by Hillebrandt et al. (2005) on hepatitis C virus-induced fibrosis, which raises the possibility that C5 alleles may be predictive for MTX-induced fibrosis. We also found differences in various individual genes with unclear toxicological relevance. Further investigation is needed at the protein level to assess complement binding in liver, validate CLN8 and ANKH as regulated by rheumatic disease or MTX exposure, and, if so, whether alleles influence expression and confer differential susceptibility to rheumatic disease. Studies using larger array platforms and a more homogenous patient population are warranted to identify additional regulated genes.
|
SUPPLEMENTARY DATA |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAREFERENCES |
|---|
Supplementary data are available online at http://toxsci.oxfordjournals.org/.
|
NOTES |
|---|
The authors certify that all research involving human subjects was done under full compliance with all government policies and the Helsinki Declaration.
|
ACKNOWLEDGMENTS |
|---|
This work was fully supported by a generous grant from Boehringer-Ingelheim Pharmaceuticals, Inc. GYW received support from the Herman Lopata Chair in Hepatitis Research. Conflicts of interest: None declared.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAREFERENCES |
|---|
Ahern MJ, Kevat S, Hill W, Hayball PJ, Harley H, Hall PD. Hepatic methotrexate content and progression of hepatic fibrosis: Preliminary findings. Ann. Rheum. Dis. (1991) 50:477–480.
Barzik M, Kotova TI, Higgs HN, Hazelwood L, Hanein D, Gertler FB, Schafer DA. Ena/VASP proteins enhance actin polymerization in the presence of barbed end capping proteins. J. Biol. Chem. (2005) 280:28653–28662.
Baugh LR, Hill AA, Brown EL, Hunter CP. Quantitative analysis of mRNA amplification by in vitro transcription. Nucleic Acids Res. (2001) 29:E29.[CrossRef][Medline]
Beyeler C, Reichen J, Thomann SR, Lauterburg BH, Gerber NJ. Quantitative liver function in patients with rheumatoid arthritis treated with low-dose methotrexate: A longitudinal study. Br. J. Rheumatol. (1997) 36:338–344.
Blanquicett C, Johnson MR, Heslin M, Diasio RB. Housekeeping gene variability in normal and carcinomatous colorectal and liver tissues: Applications in pharmacogenomic gene expression studies. Anal. Biochem. (2002) 303:209–214.[CrossRef][Web of Science][Medline]
Boffa MJ, Chalmers RJ, Haboubi NY, Shomaf M, Mitchell DM. Sequential liver biopsies during long-term methotrexate treatment for psoriasis: A reappraisal. Br. J. Dermatol. (1995) 133:774–778.[Web of Science][Medline]
Cales P, Oberti F, Michalak S, Hubert-Fouchard I, Rousselet MC, Konate A, Gallois Y, Ternisien C, Chevailler A, Lunel F. A novel panel of blood markers to assess the degree of liver fibrosis. Hepatology (2005) 42:1373–1381.[CrossRef][Web of Science][Medline]
Chalmers RJ, Kirby B, Smith A, Burrows P, Little R, Horan M, Hextall JM, Smith CH, Klaber M, Rogers S. Replacement of routine liver biopsy by procollagen III aminopeptide for monitoring patients with psoriasis receiving long-term methotrexate: A multicentre audit and health economic analysis. Br. J. Dermatol. (2005) 152:444–450.[CrossRef][Web of Science][Medline]
Chen X, Baker SM, Chen G. Methotrexate induction of human sulfotransferases in Hep G2 and Caco-2 cells. J. Appl. Toxicol. (2005) 25:354–360.[CrossRef][Web of Science][Medline]
Comer KA, Falany JL, Falany CN. Cloning and expression of human liver dehydroepiandrosterone sulphotransferase. Biochem. J. (1993) 289(Pt 1):233–240.[Web of Science][Medline]
Deighton CM, Walker DJ, Griffiths ID, Roberts DF. The contribution of HLA to rheumatoid arthritis. Clin. Genet. (1989) 36:178–182.[Web of Science][Medline]
Deng QG, She H, Cheng JH, French SW, Koop DR, Xiong S, Tsukamoto H. Steatohepatitis induced by intragastric overfeeding in mice. Hepatology (2005) 42:905–914.[CrossRef][Web of Science][Medline]
Dervieux T, Furst D, Lein DO, Capps R, Smith K, Walsh M, Kremer J. Polyglutamation of methotrexate with common polymorphisms in reduced folate carrier, aminoimidazole carboxamide ribonucleotide transformylase, and thymidylate synthase are associated with methotrexate effects in rheumatoid arthritis. Arthritis Rheum. (2004) 50:2766–2774.[CrossRef][Web of Science][Medline]
Falany CN, Comer KA, Dooley TP, Glatt H. Human dehydroepiandrosterone sulfotransferase. Purification, molecular cloning, and characterization. Ann. NY Acad. Sci. (1995) 774:59–72.[Web of Science][Medline]
Faust JR, Rodman JS, Daniel PF, Dice JF, Bronson RT. Two related proteolipids and dolichol-linked oligosaccharides accumulate in motor neuron degeneration mice (mnd/mnd), a model for neuronal ceroid lipofuscinosis. J. Biol. Chem. (1994) 269:10150–10155.
Fleisch H, Schibler D, Maerki J, Frossard I. Inhibition of aortic calcification by means of pyrophosphate and polyphosphates. Nature (1965) 207:1300–1301.[Medline]
Friedman SL. Mechanisms of disease: Mechanisms of hepatic fibrosis and therapeutic implications. Nat. Clin. Pract. Gastroenterol. Hepatol. (2004) 1:98–105.[CrossRef][Medline]
Gertler FB, Niebuhr K, Reinhard M, Wehland J, Soriano P. Mena, a relative of VASP and Drosophila Enabled, is implicated in the control of microfilament dynamics. Cell (1996) 87:227–239.[CrossRef][Web of Science][Medline]
Hall PD, Jenner MA, Ahern MJ. Hepatotoxicity in a rat model caused by orally administered methotrexate. Hepatology (1991) 14:906–910.[CrossRef][Web of Science][Medline]
Hillebrandt S, Wasmuth HE, Weiskirchen R, Hellerbrand C, Keppeler H, Werth A, Schirin-Sokhan R, Wilkens G, Geier A, Lorenzen J, et al. Complement factor 5 is a quantitative trait gene that modifies liver fibrogenesis in mice and humans. Nat. Genet. (2005) 37:835–843.[CrossRef][Web of Science][Medline]
Ho AM, Johnson MD, Kingsley DM. Role of the mouse ank gene in control of tissue calcification and arthritis. Science (2000) 289:265–270.
Hughes LB, Beasley TM, Patel H, Tiwari HK, Morgan SL, Baggott JE, Saag KG, McNicholl J, Moreland LW, Alarcon GS, et al. Racial/ethnic differences in allele frequencies of single nucleotide polymorphisms in the methylenetetrahydrofolate reductase gene and their influence on response to methotrexate in rheumatoid arthritis. Ann. Rheum. Dis (2006) 65:1121–1123.
Imbert-Bismut F, Messous D, Thibault V, Myers RB, Piton A, Thabut D, Devers L, Hainque B, Mercadier A, Poynard T. Intra-laboratory analytical variability of biochemical markers of fibrosis (Fibrotest) and activity (Actitest) and reference ranges in healthy blood donors. Clin. Chem. Lab. Med. (2004) 42:323–333.[CrossRef][Web of Science][Medline]
Jawaheer D, Seldin MF, Amos CI, Chen WV, Shigeta R, Etzel C, Damle A, Xiao X, Chen D, Lum RF, et al. Screening the genome for rheumatoid arthritis susceptibility genes: A replication study and combined analysis of 512 multicase families. Arthritis Rheum. (2003) 48:906–916.[CrossRef][Web of Science][Medline]
John S, Amos C, Shephard N, Chen W, Butterworth A, Etzel C, Jawaheer D, Seldin M, Silman A, Gregersen P, et al. Linkage analysis of rheumatoid arthritis in US and UK families reveals interactions between HLA-DRB1 and loci on chromosomes 6q and 16p. Arthritis Rheum. (2006) 54:1482–1490.[CrossRef][Web of Science][Medline]
Kishi M, Maeyama S, Koike J, Aida Y, Yoshida H, Uchikoshi T. Correlation between intrasinusoidal neutrophilic infiltration and ceroid-lipofuscinosis in alcoholic liver fibrosis with or without fatty change: Clinicopathological comparison with nutritional fatty liver. Alcohol Clin. Exp. Res. (1996) 20:366A–370A.[Medline]
Kremer JM. Not yet time to change the guidelines for monitoring methotrexate liver toxicity: They have served us well. J. Rheumatol. (2002) 29:1590–1592.
Kristensen DB, Kawada N, Imamura K, Miyamoto Y, Tateno C, Seki S, Kuroki T, Yoshizato K. Proteome analysis of rat hepatic stellate cells. Hepatology (2000) 32:268–277.[CrossRef][Web of Science][Medline]
Leifeld L, Dumoulin FL, Purr I, Janberg K, Trautwein C, Wolff M, Manns MP, Sauerbruch T, Spengler U. Early up-regulation of chemokine expression in fulminant hepatic failure. J. Pathol. (2003) 199:335–344.[CrossRef][Web of Science][Medline]
Levy MT, McCaughan GW, Marinos G, Gorrell MD. Intrahepatic expression of the hepatic stellate cell marker fibroblast activation protein correlates with the degree of fibrosis in hepatitis C virus infection. Liver (2002) 22:93–101.[CrossRef][Web of Science][Medline]
Lewindon PJ, Pereira TN, Hoskins AC, Bridle KR, Williamson RM, Shepherd RW, Ramm GA. The role of hepatic stellate cells and transforming growth factor-beta(1) in cystic fibrosis liver disease. Am. J. Pathol. (2002) 160:1705–1715.
Li A, Jiao X, Munier FL, Schorderet DF, Yao W, Iwata F, Hayakawa M, Kanai A, Shy Chen M, Alan Lewis R, et al. Bietti crystalline corneoretinal dystrophy is caused by mutations in the novel gene CYP4V2. Am. J. Hum. Genet. (2004) 74:817–826.[CrossRef][Web of Science][Medline]
Lonka L, Aalto A, Kopra O, Kuronen M, Kokaia Z, Saarma M, Lehesjoki AE. The neuronal ceroid lipofuscinosis Cln8 gene expression is developmentally regulated in mouse brain and up-regulated in the hippocampal kindling model of epilepsy. BMC Neurosci. (2005) 6:27.[CrossRef][Medline]
MacGregor AJ, Snieder H, Rigby AS, Koskenvuo M, Kaprio J, Aho K, Silman AJ. Characterizing the quantitative genetic contribution to rheumatoid arthritis using data from twins. Arthritis Rheum. (2000) 43:30–37.[CrossRef][Web of Science][Medline]
Malatjalian DA, Ross JB, Williams CN, Colwell SJ, Eastwood BJ. Methotrexate hepatotoxicity in psoriatics: Report of 104 patients from Nova Scotia, with analysis of risks from obesity, diabetes and alcohol consumption during long term follow-up. Can. J. Gastroenterol. (1996) 10:369–375.[Web of Science][Medline]
Maurice PD, Maddox AJ, Green CA, Tatnall F, Schofield JK, Stott DJ. Monitoring patients on methotrexate: Hepatic fibrosis not seen in patients with normal serum assays of aminoterminal peptide of type III procollagen. Br. J. Dermatol. (2005) 152:451–458.[CrossRef][Web of Science][Medline]
McKee S, Pendleton A, Dixey J, Doherty M, Hughes A. Autosomal dominant early childhood seizures associated with chondrocalcinosis and a mutation in the ANKH Gene. Epilepsia (2004) 45:1258–1260.[CrossRef][Web of Science][Medline]
Nakatsukasa H, Nagy P, Evarts RP, Hsia CC, Marsden E, Thorgeirsson SS. Cellular distribution of transforming growth factor-beta 1 and procollagen types I, III, and IV transcripts in carbon tetrachloride-induced rat liver fibrosis. J. Clin. Investig. (1990) 85:1833–1843.[Web of Science][Medline]
Pawlak K, Pawlak D, Mysliwiec M. Hepatitis intensified oxidative stress, MIP-1beta and RANTES plasma levels in uraemic patients. Cytokine (2004) 28:197–204.[CrossRef][Web of Science][Medline]
Pendleton A, Johnson MD, Hughes A, Gurley KA, Ho AM, Doherty M, Dixey J, Gillet P, Loeuille D, McGrath R, et al. Mutations in ANKH cause chondrocalcinosis. Am. J. Hum. Genet. (2002) 71:933–940.[CrossRef][Web of Science][Medline]
Qin X, Gao B. The complement system in liver diseases. Cell. Mol. Immunol. (2006) 3:333–340.[Medline]
Rachow JW, Ryan LM. Inorganic pyrophosphate metabolism in arthritis. Rheum. Dis. Clin. North Am. (1988) 14:289–302.[Web of Science][Medline]
Radich JP, Mao M, Stepaniants S, Biery M, Castle J, Ward T, Schimmack G, Kobayashi S, Carleton M, Lampe J, et al. Individual-specific variation of gene expression in peripheral blood leukocytes. Genomics (2004) 83:980–988.[CrossRef][Web of Science][Medline]
Richard S, Guerret S, Gerard F, Tebib JG, Vignon E. Hepatic fibrosis in rheumatoid arthritis patients treated with methotrexate: Application of a new semi-quantitative scoring system. Rheumatology (Oxford) (2000) 39:50–54.[CrossRef][Medline]
Rinn JL, Rozowsky JS, Laurenzi IJ, Petersen PH, Zou K, Zhong W, Gerstein M, Snyder M. Major molecular differences between mammalian sexes are involved in drug metabolism and renal function. Dev. Cell (2004) 6:791–800.[CrossRef][Web of Science][Medline]
Risch N, Merikangas K. The future of genetic studies of complex human diseases. Science (1996) 273:1516–1517.
Roderfeld M, Weiskirchen R, Wagner S, Berres ML, Henkel C, Grotzinger J, Gressner AM, Matern S, Roeb E. Inhibition of hepatic fibrogenesis by matrix metalloproteinase-9 mutants in mice. Faseb J. (2006) 20:444–454.
Roenigk HH Jr,, Auerbach R, Maibach HI, Weinstein GD. Methotrexate guidelines—Revised. J. Am. Acad. Dermatol. (1982) 6:145–155.[Web of Science][Medline]
Rossi E, Adams L, Prins A, Bulsara M, de Boer B, Garas G, MacQuillan G, Speers D, Jeffrey G. Validation of the FibroTest biochemical markers score in assessing liver fibrosis in hepatitis C patients. Clin. Chem. (2003) 49:450–454.
Santavuori P. Neuronal ceroid-lipofuscinoses in childhood. Brain Dev. (1988) 10:80–83.[Web of Science][Medline]
Schuppan D. Connective tissue polypeptides in serum as parameters to monitor antifibrotic treatment in hepatic fibrogenesis. J. Hepatol. (1991) 13(Suppl. 3):S17–S25.
t Hoen PA, de Kort F, van Ommen GJ, den Dunnen JT. Fluorescent labelling of cRNA for microarray applications. Nucleic Acids Res. (2003) 31:e20.
Takahara T, Kojima T, Miyabayashi C, Inoue K, Sasaki H, Muragaki Y, Ooshima A. Collagen production in fat-storing cells after carbon tetrachloride intoxication in the rat. Immunoelectron microscopic observation of type I, type III collagens, and prolyl hydroxylase. Lab. Investig. (1988) 59:509–521.[Web of Science][Medline]
Tsui FW, Tsui HW, Cheng EY, Stone M, Payne U, Reveille JD, Shulman MJ, Paterson AD, Inman RD. Novel genetic markers in the 5'-flanking region of ANKH are associated with ankylosing spondylitis. Arthritis Rheum. (2003) 48:791–797.[CrossRef][Web of Science][Medline]
Tusher VG, Tibshirani R, Chu G. Significance analysis of microarrays applied to the ionizing radiation response. Proc. Natl. Acad. Sci. U.S.A. (2001) 98:5116–5121.
Urano W, Taniguchi A, Yamanaka H, Tanaka E, Nakajima H, Matsuda Y, Akama H, Kitamura Y, Kamatani N. Polymorphisms in the methylenetetrahydrofolate reductase gene were associated with both the efficacy and the toxicity of methotrexate used for the treatment of rheumatoid arthritis, as evidenced by single locus and haplotype analyses. Pharmacogenetics (2002) 12:183–190.[CrossRef][Web of Science][Medline]
Weimer JM, Kriscenski-Perry E, Elshatory Y, Pearce DA. The neuronal ceroid lipofuscinoses: Mutations in different proteins result in similar disease. Neuromolecular Med. (2002) 1:111–124.[CrossRef][Web of Science][Medline]
Wessels JA, de Vries-Bouwstra JK, Heijmans BT, Slagboom PE, Goekoop-Ruiterman YP, Allaart CF, Kerstens PJ, van Zeben D, Breedveld FC, Dijkmans BA, et al. Efficacy and toxicity of methotrexate in early rheumatoid arthritis are associated with single-nucleotide polymorphisms in genes coding for folate pathway enzymes. Arthritis Rheum. (2006) 54:1087–1095.[CrossRef][Web of Science][Medline]
Whiting-O'Keefe QE, Fye KH, Sack KD. Methotrexate and histologic hepatic abnormalities: A meta-analysis. Am. J. Med. (1991) 90:711–716.[Web of Science][Medline]
Whitney AR, Diehn M, Popper SJ, Alizadeh AA, Boldrick JC, Relman DA, Brown PO. Individuality and variation in gene expression patterns in human blood. Proc. Natl. Acad. Sci. U.S.A. (2003) 100:1896–1901.
Wordsworth P, Bell J. Polygenic susceptibility in rheumatoid arthritis. Ann. Rheum. Dis. (1991) 50:343–346.
Wright GW, Simon RM. A random variance model for detection of differential gene expression in small microarray experiments. Bioinformatics (2003) 19:2448–2455.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  H. Jarvelainen, A. Sainio, M. Koulu, T. N. Wight, and R. Penttinen Extracellular Matrix Molecules: Potential Targets in Pharmacotherapy Pharmacol. Rev., June 1, 2009; 61(2): 198 - 223. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||