Abstract
Background/Aim: Hereditary leiomyomatosis and renal cell carcinoma (HLRCC) is a rare autosomal dominant disorder characterized by fumarate hydratase (FH) gene mutation. It is associated with the development of very aggressive kidney tumors, characterized by early onset and high metastatic potential, and has no effective therapy. The aim of the study was to establish a new preclinical platform for investigating morphogenetic and metabolic features, and alternative therapy of metastatic hereditary papillary renal cell carcinoma type 2 (PRCC2). Materials and Methods: Fresh cells were collected from pleural fluid of a patient with metastatic hereditary PRCC2. Morphogenetic and functional characteristics were evaluated via microscopy, FH gene sequencing analysis, real-time polymerase chaine reaction and enzymatic activity measurement. We performed bioenergetic analysis, gene-expression profiling, and cell viability assay with 19 anti-neoplastic drugs. Results: We established a new in vitro model of hereditary PRCC2 – the NCCFH1 cell line. The cell line possesses a c.1162 delA – p.Thr375fs frameshift mutation in the FH gene. Our findings indicate severe attenuation of oxidative phosphorylation and glucose-dependent growth of NCCFH1 cells that is consistent with the Warburg effect. Furthermore, gene-expression profiling identified that the most prominent molecular features reflected a high level of apoptosis, cell adhesion, and cell signaling. Drug screening revealed a marked sensitivity of FH−/− cells to mitoxantrone, epirubicin, topotecan and a high sensitivity to bortezomib. Conclusion: We demonstrated that the NCCFH1 cell line is a very interesting preclinical model for studying the metabolic features and testing new therapies for hereditary PRCC2, while bortezomib may be a potential efficient therapeutic option.
Hereditary leiomyomatosis and renal cell carcinoma (HLRCC, OMIM 150800) is an inherited syndrome previously referred to as multiple cutaneous and uterine leiomyomatosis. HLRCC is associated with the development of cutaneous and uterine leiomyomas, papillary renal cell carcinoma type 2 (PRCC2) and, rarely, uterine leiomyosarcoma (1-7). PRCC2 has a high metastatic potential and is characterized by early onset. HLRCC is due to germline heterozygous mutations of the fumarate hydratase (FH) gene. This gene is known to be a tumor suppressor and encodes the enzyme fumarate hydratase (fumarase; FH). Fumarase is the Kreb cycle enzyme that catalyzes the conversion of fumarate to L-malate in the mitochondrial matrix (1, 4, 5, 8, 9). Thus, FH gene mutations result in fumarate accumulation, with the alteration of multiple metabolic intermediates, and activation of oncogenic pathways. It has been shown that a high level of fumarate induces the activation of hypoxia-inducible factor 1α, whose target genes are involved in angiogenesis, cell proliferation, and survival (10-12). The loss of FH leads to deregulation of energy production, which is followed by an enhanced level of glycolysis in the cell and induction of the production of glucose-mediated reactive oxygen species (13). Moreover, fumarate acts as an electrophile, promoting up-regulation of nuclear factor (erythroid-derived 2)-like 2 (NRF2) through succination of the Kelch-like ECH-associated protein 1 (KEAP1) that mediates the anti-oxidant signaling pathway activation (14, 15). This activation results in overexpression of antioxidant-response element (ARE) genes, such as aldo-keto reductase family 1 member B10 gene (AKR1B10) (14).
Previous studies of Yang et al. established UOK 262 and UOK 268 cell lines as the first human FH-deficient HLRCC-associated kidney cancer cell lines (16, 17). They showed that both cell lines display the Warburg phenomenon (18). These studies also described how UOK 262 and UOK 268 cells exhibit severe depression of oxidative phosphorylation and a high level of aerobic glycolysis. Furthermore, the mitochondrial gene profiling of UOK 268 cells revealed strong deregulation of fatty acid metabolism (17).
The latest studies of PRCC2 have provided important insights into the pathways and the bioenergetics features of the carcinogenesis of this type of kidney tumor. However, there is no efficient therapy for patients with HLRCC. The rarity of this disorder limits the number of preclinical models available for study and the number of clinical trials carried out.
In the present study, we describe the NCCFH1 cell line as a new preclinical in vitro model for studying the metabolic abnormalities of rare hereditary PRCC2 with its specific FH-deficient feature.
Materials and Methods
Patient information. A 17-year-old Singaporean patient was treated at the National Cancer Centre Singapore, at the Divisions of Medical Oncology and Cancer Genetics. Written consent was obtained from both next of kin and the patient. The Singhealth Centralized Institutional Review Board approved this study (IRB number: 2008/443/B). The patient had been diagnosed with HLRCC syndrome and died 19 months after diagnosis from disease progression.
Cells and cell culture. The NCCFH1 cell line was established from the pleural fluid of the patient described above. The cells were seeded in three T75 flasks and fed with three different media: Roswell Park Memorial Institute-1640 medium (RPMI, Life Technologies, Foster City, CA, USA), Dulbecco's modified Eagle's medium (DMEM; Life Technologies, Carlsbad, CA, USA) and human umbilical vein endothelial cell medium (HuVec; Life Technologies). Each medium was supplemented with 10% of fetal bovine serum (FBS; Invitrogen, Carlsbad, CA, USA) and 1% of an antibiotic antimycotic solution (Sigma-Aldrich, St. Louis, MO, USA).
The normal human kidney HK-2 cell line was obtained from the American Type Culture Collection. Dr. Len Neckers and Dr. Youfeng Yang provided UOK 262 cells from the Urologic Oncology Branch Tumor Cell line Repository (National Cancer Institute, Bethesda, MD, USA). The HK-2 cell line was maintained in DMEM and FH-deficient kidney cancer UOK 262 cells were cultured in RPMI. Each medium was supplemented with 10% of fetal bovine serum (FBS) and 1% of an antibiotic and anti-mycotic solution. The cells were maintained in a humidified incubator with 5% CO2 at 37°C.
Short tandem repeat (STR) profiling. Short tandem repeat profiling of the cell lines was conducted using the Geneprint 10 system (Promega, Madison, WI, USA) on an ABI3130xl Genetic Analyzer (Life Technologies) using standard protocols. Analysis was conducted using the Gene Mapper V4.0 software (Life Technologies).
Sequencing analysis. Cell DNA was extracted using QIAGEN's Blood and Cell Culture Mini kit (QIAGEN, Valencia, CA, USA), according to the manufacturer's instructions. FH gene mutations were screened by DNA amplification of each exon. The polymerase chain reaction (PCR) products were purified with the ExoSAP PCR purification kit (USB/Affymetrix, Santa Clara, CA, USA) and were sequenced using ABI PRISM BigDye Terminator Cycle Sequencing Ready Reaction kit (version 3, Applied Biosystems, Foster City, CA, USA). The sequencing was performed on an ABI PRISM 3730 Genetic Analyzer (Applied Biosystems). The chromatograms were analyzed by Sequencher Demo version (Applied Biosystems). The primers used for amplification of FH gene were as follows: Exon 1 F: CTACCCAAGCTCCCTCAGC, R: CAGGCA GGAGGGCTGAAG; exon 2 F: ACAGCTGATAAGATGCGATT ACT, R: GCC TAC TTCA TCC AAA ATA GCC; exon 3 F: TGCATCTGCCAAAATAATAAACT, R: AGTATGGCATGGGTCTG AGG; exon 4 F: TCA AAC TCT GTG GCA TAA TCA G, R CTT GTC AAA AAG TGA ATG CTT GT; exon 5 F: GAAGTTT GTTTTTGTTGCCTCTG, R: ATTGGCCATTTG TACCAAGC; exon 6 F: CATCCTTCCCTATACTTTGCTCA, R: CACAAGAATTCAAGACAGGAACA; exon 7 F: TGGAACTTT CTGTTTCACTTGC, R: TGACCAGAG GACCACAGACA; exon 8 F: GTTGGGCCTT GCTTTATTGT, R: ACCCAACTACCCAATGT GGA; exon 9 F: GTGCCTTCAAATGTTCATGC, R: GCTGTTCT CAAACA CTGATCCA; exon 10 F: TCACTGCTAACCCATA TGTCG, R: CCTAGCA CATCCTAGGGTTTTA.
Quantitative reverse-transcription PCR. RNA samples were prepared from HK-2, NCCFH1 and UOK 262 cells using TRIzol® reagent (Life Technologies), according to the manufacturer's recommended protocol. RNA (1 μg) from samples was reverse-transcribed into cDNA with i-script cDNA synthesis kit (Bio-Rad, Foster City, CA, USA). PCR amplifications were performed in an ABI 9700 thermal cycler instrument, and followed by quantitative PCR in a CFX96™ Real-Time PCR Detection System (Bio-Rad).
Taqman probes Hs 00252524_m1 (AKR1B10) and Hs00264 683_m1 (FH) (Applied Biosystems) were used in this study. Relative mRNA levels were calculated using the ΔΔCt method (22). The mRNA expression level for these genes was normalized to that for a housekeeping gene (18S) and the Ct values were further normalized to those obtained from the normal HK-2 cell line.
Immunoblot. The cells were cultured in 6-well plates and harvested 24 h post-seeding. The proteins were extracted with phosphate-buffered saline (PBS) containing 1% triton-X100 (Sigma Aldrich) in the presence of protease inhibitors. Thirty micrograms of proteins were resolved by 12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis and were electroblotted to a nitrocellulose membrane (Thermo Fisher Scientific, Rockford, IL, USA). The membrane was incubated overnight at 4°C with blocking buffer (PBS containing 5% w/v milk and 0.05% v/v Tween-20). Primary and secondary antibody incubations were performed in blocking buffer. Mouse monoclonal antibody against FH (#ab58232; Abcam, Cambridge, Cambridgeshire, UK) was diluted to 1:500, the mouse monoclonal antibody to β-actin (# A1978; Sigma Aldrich) was diluted to 1:20,000. The membrane was washed with PBS containing 0.05% v/v Tween-20 followed by analysis using the Supersignal Chemiluminescent kit (Pierce Chemical Co., Rockford, IL, USA) according to the manufacturer's recommendations.
Determination of FH enzymatic activity. The FH activity assay was performed using Fumarase Specific Activity Microplate Assay Kit (Abcam), according to the manufacturer's protocol. Briefly, cells were lyzed with sample extraction buffer. Protein concentration was determined by OD280 and samples containing 250 μg/ml of protein were used for the assay. Samples were incubated in the 96-well microplate for 3 h and were washed with 1X wash buffer, and 1X activity solution. The absorbance at 450 nm was then measured to determine FH activity. To determine FH quantity, samples were incubated with 1X detector antibody for 1 h at room temperature. Wells were then washed with 1X wash buffer and incubated with 1X horseradish peroxidase (HRP) label for 1 h at room temperature. Finally, wells were washed with 1X wash buffer and HRP development solution was added to each well. The absorbance at 600 nm was measured after 10 min.
Extracellular acidification and oxygen consumption rate. The XF96 Extracellular Flux Analyzer (Seahorse Bioscience, North Billerica, MA, USA) was used to detect for rapid and real-time changes in cellular respiration and glycolysis rates. NCCFH1, UOK 262 and HK-2 control cells were cultured in custom XF96 culture plates and seeded at a density of 30,000 cells/well. The oxygen consumption rate (OCR) and the extracellular acidification rate (ECAR) were measured 24 h after seeding. The OCR reflects cellular respiration and is measured in pMol/min, the ECAR reflects lactate excretion and the glycolysis rate measured in mpH/min. All measurements were performed following the manufacturer's protocols.
Glucose-dependence proliferation assay. HK-2, NCCFH1 and UOK 262 cells were seeded at 2,000, 3,000 and 4,000 cells, respectively, in 96-well plates in triplicate and were incubated for 72 h. Cell proliferation rates were compared at different glucose concentrations (0.5, 1, 1.5, 2, and 4.5 g/l) in the cell culture media. The proliferation rates were assessed using the colorimetric 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium (MTS) assay, according to the manufacturer's protocol (Promega, Madison, WI, USA). For this assay, 100 μl of fresh media and 20 μl of MTS were added to each well and cells were incubated for a further 2 h. Cells were evaluated in triplicate. A Tecan i-control microplate reader (Tecan, Männedorf, Switzerland), at an optical absorbance at 490 nm and reference wavelength at 690 nm, was used to measure cell proliferation.
Microarray profiling datasets. Expression profiles were generated from NCCFH1, UOK 262, and HK-2 cell lines using Affymetrix HG-U133 Plus 2.0TM chipset (Affymetrix) as previously described (21). The Affymetrix GeneChip scanner generated the CEL files. The raw probe intensities were normalized and summarized to probe sets using the RMA algorithm (http://www.ncbi.nlm.nih.gov/pubmed/12925520).
The affymetrix probes were re-annotated using a custom chip definition file from the Molecular and Behavioral Neuroscience Institute that offers more accurate mapping to genes (http://nar.oxfordjournals.org/content/33/20/e175.full?ijkey=zaJMV7qU1XANIci&keytype=ref). The gene-enrichment analysis, the pathways, and the gene/term similarity were carried-out with the Database for Annotation, Visualization and Integrated Discovery (DAVID) bioinformatics resources (DAVID/Version 6.7, Bethesda, MD, USA).
Drug-response screening. HK-2, NCCFH1 and UOK 262 cells were seeded at 2,000, 3,000 and 4,000 cells, respectively, in 96-well plates and incubated overnight. The cells were then treated with 19 drugs at 10 μM for 72 h: 5-fluorouracil (5-FU) (Teva, Petah Tikva, Israel), bortezomib (Millennium Pharmaceuticals, Cambridge, MA, USA), cisplatin (Mayne Pharma, Lake Forest, IL, USA), doxorubicin (Schering Plough, Oslo, Norway), epirubicin (Pfizer, New York, NY, USA), etoposide (Bristol-Myers Squibb, Princeton, NJ, USA), everolimus (Sigma-Aldrich), gemcitabine (Eli Lilly, Indianapolis, IN, USA), irinotecan (Pfizer, New York, NY, USA), methotrexate (Teva, Petah Tikva, Israel), mitoxantrone (Sigma-Aldrich), oxaliplatin (Sanofi-Aventis, Paris, Ile-de-France, France), paclitaxel (Mayne Pharma), pemetrexed (Eli Lilly), raltitrexed (Astra Zeneca, London, UK), sunitinib (Pfizer), temsirolimus (Sigma-Aldrich), topotecan (GlaxoSmithKline, Brentford, Middlesex, UK) and vinorelbine (Mayne Pharma). In addition, cells were also treated separately with bortezomib at 10, 1, 0.1, 0.01 and 0.001 μM for 72 h. Cell viability was assessed via the MTS assay (Promega) as described above. The results are expressed as the percentage of viable cells relative to those for cells treated with 0.56% dimethyl sulfoxide (control) or to the cells without treatment. Experiments were performed in triplicate and repeated at least three times.
Results
Clinical characteristics of the patient. The 17-year-old patient presented with right-sided shoulder discomfort. A magnetic resonance imaging, which showed a large mass at the posterolateral aspect of the humeral head and a greater tuberosity, helped diagnose a distant humeral metastasis. The dimensions of the metastatic mass were 4.1×3.2×5.7 cm. The mass was located at the metaphysis and the epiphysis of the proximal end of the humerus, and extended to the superior subcortical/subarticular surface (Figure 1A). The biopsy of the humeral metastasis revealed that the tumor tissue presented alveolar and tubulopapillary growth patterns. The tumor cells had a cuboidal form with a round, bland nuclei and small amounts of eosinophilic cytoplasm (from the histopathological report).
Clinical history of the patient. A: Magnetic resonance imaging of the patient's right shoulder, showing a large mass at the posterolateral aspect of the humeral head and greater tuberosity. B: Abdominal computed tomographic scan showing a mass of the right upper pole of the kidney. C: Chest radiography showing the left-sided pleural effusion.
Short repeat tandem analysis (STR) of DNA from patient's blood and NCCFH1 cell line. Confidence of matching between blood DNA and NCCFH1 DNA was 97%.
An abdominal computed tomographic scan showed a primary tumor mass, measuring 5 cm in its longest diameter, located in the upper pole of the right kidney (Figure 1B). The paternal grandmother of the patient had developed uterine leiomyoma with PRCC2 and two of his uncles had developed PRCC2. This strong family history, as well as the above clinical data, led to diagnosis of HLRCC-associated metastatic kidney cancer (23). The patient was first treated with sunitinib and displayed an initial clinical response. Despite this treatment, the disease progressed and the patient developed a malignant pleural effusion (Figure 1C). In previous studies of Gardie et al., a germline mutational analysis revealed an heterozygous frameshift mutation located in exon 8: c.1162 delA, p.Thr375fs of the FH gene in serum from this patient (24).
Morphogenetic characteristics of the NCCFH1 cell line. The NCCFH1 cell line was established at the National Cancer Centre Singapore from the pleural fluid of the lung metastasis. The primary cells were seeded in three different media as described in the Materials and Methods section. Within 3 to 4 weeks, the RMPI-seeded cells started to proliferate.
NCCFH1 cells have irregular morphology, with abnormal branching, large and oval nucleus, vested with multiple nucleoli (Figure 2A). The STR analysis of the DNA samples of the blood and of NCCFH1 cells revealed 97% matching in 10 loci (Table I). DNA sequencing was performed in order to determine if NCCFH1 cells possessed the same FH germline mutation as in the patient's sample. As expected, we found that NCCFH1 cells displayed an identical germline frameshift mutation, with adenine deletion at nucleotide 1162 in exon 8, and we were able to visualize the loss of heterozygosity from the chromatograms (Figure 2B). The detailed analysis of the nucleotide sequence after the frameshift mutation showed a premature stop codon at codon 380 (ATG→TGA) in exon 8 of the FH gene.
Morphogenetic characteristics of the NCCFH1 cell line. A: Morphology of NCCFH1 cells at ×10 magnification. B: DNA chromatograms of fumarate hydratase (FH) gene for the normal human genomic DNA (control) and NCCFH1 cells, showing the sequence adjacent to c.1162 delA of exon 8. C: Real-time quantitative reverse transcription PCR showing the relative FH mRNA expression in NCCFH1 cells. The samples were run in duplicate and the experiment was repeated twice. Error bars represent the SD. **p<0.01. D: Immunoblot showing the FH levels in whole-cell lysates of normal kidney HK-2 and NCCFH1 cells. β-Actin protein was used as a loading control. FH activity (E) and quantity (F) in NCCFH1 cells relative to HK-2 cells. Each sample was run in triplicates. Data are the mean±SD. ***p≤0.001.
Relative FH gene expression analysis showed a strong decrease of the FH mRNA level in the NCCFH1 cell line compared to the normal HK-2 cell line (p<0.01) (Figure 2C). Importantly, the immunoblot analysis demonstrated that FH protein was not expressed by these cells (Figure 2D). We noted the correlation between a low FH mRNA level and undetectable level of FH protein in NCCFH1 cells. We further evaluated the enzymatic activity of FH. The NCCFH1 cell lysate practically lacked any quantity and enzymatic activity of FH compared to the HK-2 cell control (Figure 2E and F). We suggest that the c.1162 delA – p.Thr375fs frameshift mutation in the FH gene results in a premature stop codon, in instability of FH mRNA and loss of normal FH expression and enzymatic activity in NCCFH1 cells.
Warburg phenotype of NCCFH1 cells. We next investigated the levels of mitochondrial respiration and glycolysis in FH-deficient NCCFH1 cells by performing XF extracellular flux assays (25, 26). The metabolic phenotype was studied through the basal OCR and the ECAR. Normal kidney HK-2 and UOK 262 cell lines were used as controls.
On the one hand, the basal OCR rate in NCCFH1 cells was around 10 pMoles/min for 3×104 cells and was lower than that of UOK 262 cells, with a rate less than 20 pMoles/min for 3×104 cells. On the other, the basal OCR rate in the positive control, HK-2, was in the range of 130-140 pMoles/min for 3×104 cells (Figure 3 A). These data indicate that NCCFH1 is a respiratory-deficient cell line, characterized by a phenotype with lower oxygen consumption than the UOK 262 cell line. However, ECAR changes in NCCFH1 cells were slightly lower than those obtained for the HK-2 cell line and significantly higher than those for UOK 262 cells (Figure 3B).
In addition, the OCR/ECAR ratio was markedly lower in FH-deficient cells compared to the HK-2 cell line (Figure 3 C). The average level was 0.14 pMoles/mpH for NCCFH1 cells and 0.29 pMoles/mpH for UOK 262 cells, but 1.73 pMoles/mpH for HK-2 cells. Interestingly, we observed that the OCR/ECAR level was two-fold lower in NCCFH1 cells compared to UOK 262 cells. These data showed dependency on glycolysis of the HLRCC cell line and indicate that NCCFH1 cells are likely to have a higher level of metabolic deregulation than UOK 262 cells.
To confirm this result, a cell-proliferation assay was performed to determine whether NCCFH1 cell line exhibits glucose-dependency for survival and proliferation. The glucose concentration in the medium ranged from 0.5 to 4.5 g/l. Normal kidney HK-2 and cancer UOK 262 cells were used as controls. HK-2 cell line survived when the medium had only 0.5 g/l of glucose, and at 1 g/l of glucose, the conditions were already optimal for proliferation and growth. The proliferative rate of NCCFH1 cells was correlated with that of UOK 262 cells and was significantly different from that of HK-2 cells. We found that the FH-mutant cells required between 1.5 g/l to 2 g/l of glucose for cells to survive and around 2 g/l of glucose for optimal growth and proliferation (Figure 3D). The difference in the proliferative rates between HK-2 and FH-mutant cells were significant at 0.5 g/I, 1 g/I, and 1.5 g/I glucose. Glucose at 4.5 g/l markedly increased the proliferation of FH-mutant cells (p<0.01) compared to that of HK-2 cells (Figure 3 D).
In addition, NCCFH1 cells had a lower proliferative rate than UOK 262 cells, with significant differences at 1 g/l and 1.5 g/l concentrations (p<0.001 and p<0.01, respectively). These data confirm that the HLRCC cell lines are dependent on glucose concentration for their survival and proliferation and that NCCFH1 cells are characterized by a phenotype with a higher glucose-dependency than UOK 262 cell line.
Gene-expression profiling and pathway analysis. We used microarray gene expression profiling to determine which genes are specifically deregulated in FH-deficient cells in comparison to normal healthy cells. The microarray assays were performed in duplicates – from two separate total RNA extracts – for NCCFH1 and UOK 262 cell lines, and with one sample for the HK-2 cell line. The analysis of the raw output data showed that more than 100 genes are differentially up-regulated (with fold-change from 8.67 up to 146), and more than 100 genes are down-regulated (with fold-change from 0.0028 up to 0.075) in both the HLRCC cell lines compared to normal kidney cells. The top 40 genes found to be highly deregulated in FH-deficient cells included transmembrane protein with EGF-like and two follistatin-like domains 2 (TMEFF2), serglycin (SRGN), aldo-keto reductase family 1. member B10 (AKR1B10), fibronectin 1 (FN1), caveolin 1, caveolae protein, 22 kDa (CAV1), claudin 11 (CLDN11), epiregulin (EREG), heat shock 70 kDa protein 1A (HSPA1A), and stearoyl-CoA desaturase (delta-9-desaturase) (SCD) (Tables II and III). These genes are involved in protein kinase B and mitogen-activated protein kinase signaling, apoptosis, metabolism, nuclear factor kappa-light-chain-enhancer of activated B-cells and ubiquitin-proteasome pathways.
Subsequently, we performed a functional annotation clustering analysis (DAVID bioinformatics) of the 200 most significantly deregulated genes in FH-mutant cells (27). We found two clusters of up-regulated genes having the most relevant gene ontology (GO) terms for regulation of apoptosis (p<0.01, enrichment score >2.1). One subset of 10 genes of the first cluster and one subset of three genes of the second cluster overlapped with all presented GO terms (Figure 4A). These genes were epidermal growth factor receptor (EGFR), caspase recruitment domain family, member 16 (CARD16), scinderin (SCIN), pit-oct-unc class 3 homeobox 3 (POU3F3), signaling lymphocyte activating molecule family member 7 (SLAMF7), caspase 1, apoptosis-related cysteine peptidase (CASP1), caspase recruitment domain family, member 6 (CARD6), neurofilament, light polypeptide (NEFL), tissue metallopeptidase inhibitor 3 (TIMP3) and sialophorin (SPN). The analysis of the 100 most down-regulated genes revealed two clusters of genes. We found that the first cluster, with a subset of nine genes, is associated with extracellular matrix activity (p<0.001, enrichment score=3.2). Three extracellular matrix GO terms overlapped for seven out of the nine genes. These genes were cochlin (COCH), collagen triple helix repeat containing 1 (CTHRC1), nephronectin (NPNT), microfibrillar-associated protein 2 (MFAP2), nidogen 1 (NID1), secreted protein, acidic, cysteine-rich (osteonectin) (SPARC), and fibronectin 1 (FN1) (Figure 4B). In addition, the second cluster of the down-regulated genes was represented by GO terms for GTPase binding activity (p<0.01; enrichment score=2.8). The subset of genes RAB31, member RAS oncogene family (RAB31), RAB3B, tubulin, beta 2b class IIb (TUBB2B), eukaryotic translation elongation factor 1 alpha 2 (EEF1A2), tubulin, beta 6 class V (TUBB6), RAS-related associated with diabetes (RRAD), GTP-binding protein overexpressed in skeletal muscle (GEM) and dynamin 1 (DNM1), had six GTPase activity GO terms in common (Figure 4B). Consequently, these data show that loss of FH activity results in alteration of v-Akt murine thymoma viral oncogene (AKT)/extracellular-related kinase (ERK) signaling, apoptosis, metabolism, nuclear factor kappa B (NF-κB) and ubiquitin-proteasome pathways, extracellular matrix and GTPase binding activity.
Warburg phenotype. A: Data for basal oxygen consumption rate (OCR). B: Data for basal extracellular acidification rate (ECAR). C: OCR/ECAR ratio. Results are for HK-2, NCCFH1, UOK 262 cells expressed per 3×104 cells. Data are the mean±standard error of the mean. D: In vitro proliferation assay with different D-glucose concentrations. A high level of D-glucose increased proliferation of NCCFH1 and UOK 262 cells. Cell proliferation was determined by the MTS assay and is shown as absorbance values at different glucose concentrations after 72 h of incubation. Data are the mean± standard error of the mean. **p<0.01, ***p<0.001.
Gene-expression profiling data: list of 20 most differentially up-regulated genes in hereditary leiomyomatosis and renal cell carcinoma cell lines compared to normal HK-2 cell line. Data are means (n=2).
Gene-expression profiling data: list of 20 most differentially down-regulated genes in hereditary leiomyomatosis and renal cell carcinoma cell lines compared to normal HK-2 cell line. Data are means (n=2).
Gene expression and functional annotation clustering analysis. A: Annotation clusters of the most up-regulated genes associated with apoptosis Gene Ontology (GO) terms. B: Annotation clusters of the most down-regulated genes associated with extracellular matrix and GTP binding GO terms. DAVID bioinformatics database 6.7 version. C: Real-time quantitative reverse transcription PCR of the relative aldo-keto reductase family 1. member B10 (AKR1B10) mRNA level in HK-2, NCCFH1 and UOK 262 cells. Data are the mean±SD.
Drug screening. Cell viability (%) was determined by MTS assay. HK-2, NCCFH1 and UOK 262 cells were treated for 72 h with different drugs at 10 μM concentration: Sunitinib, everolimus, temsirolimus (A); 5-Fluorouracil (5-FU), methotrexate, pemetrexed, paclitaxel, doxorubicin, irinotecan, cisplatin, oxaliplatin (B); etoposide, gemcitabine, raltitrexed and vinorelbine (C); mitoxantrone and epirubicin (D); topotecan and bortezomib (E). F: Half maximal inhibitory concentration (IC50) assay. Cells were treated with bortezomib for 72 h with a large range of concentrations. Results were normalized to those of cells treated with 0.56% dimethyl sulfoxide. Data are the mean±standard error of the mean (n=3). *p<0.05, **p<0.01, ****p<0.0001.
We validated the expression levels of AKR1B10 gene, found to be up-regulated in our NCCFH1 cell line, using RT-qPCR. We chose to validate this particular gene due its consistently high expression in HLRCC cell lines and in clinical primary tumor samples from patients (14). A previous study of Ooi et al. showed that AKR1B10 is an ARE-controlled gene and its overexpression is a prominent feature in hereditary and sporadic PRCC2. As expected, we found a significant increase of the AKR1B10 mRNA level in both NCCFH1 and UOK 262 cell lines compared to HK-2 control cells (Figure 4C).
Upstream regulatory elements of the most deregulated genes in hereditary leiomyomatosis and renal cell carcinoma cell lines. According to QIAGEN database.
Drug screening: High sensitivity of NCCFH1 cell line to bortezomib. The current targeted therapy for FH-deficient RCC consists of administering tyrosine kinase inhibitors, such as sunitinib and sorafenib, and mammalian target of rapamycin inhibitors, such as temsirolimus and everolimus. We performed cell-viability assays with 19 anti-neoplastic drugs to evaluate the response of FH-deficient cells to each of them. The NCCFH1, UOK 262 and HK-2 cells (control) were treated with 10 μM concentration of each drug. Cell viability was evaluated after 72-h exposure to each drug and was assessed via the MTS assay. For the three cell lines, a marginal effect was obtained after treatment with sunitinib, temsirolimus and everolimus (Figure 5 A). Nearly all of the cells were alive (≈90%) after 72-h of drug treatment. In addition, FH-mutant cells demonstrated a high resistance to cytotoxic and cytostatic drugs, such as cisplatin, etoposide, doxorubicin, gemcitabine, irinotecan, 5-FU, methotrexate, oxaliplatin, paclitaxel, pemetrexed, raltitrexed and vinorelbine (Figure 5B and C). The normal HK-2 cell line was found to be sensitive to most of these drugs compared to NCCFH1 cells. The exception was observed in HK-2 cells in response to irinotecan and cisplatin, where HK-2 cells present higher resistance similar to that for NCCFH1 cells (Figure 5 B).
Mitoxantrone, epirubicin and topotecan had significant cytotoxic effect against the FH-deficient cell line, with viability of NCCFH1 cells of 19% (p<0.01), 28% (p<0.01), and 36% (p<0.05), respectively, compared to NCCFH1 cells treated with DMSO. Similarly, only 1% of UOK 262 cells survived mitoxantrone treatment (p<0.0001), 22% epirubicin (p<0.01) and 21% topotecan (p<0.01) compared to UOK 262 cells treated with DMSO (Figure 5D and E). Interestingly, NCCFH1 cells were found to have a significantly higher level of resistance to mitoxantrone compared to UOK 262 cells (Figure 5D). However, HK-2 cells, like FH-mutant cells, had a high sensitivity to mitoxantrone, epirubicin and topotecan.
Importantly, the cell viability of NCCFH1 and UOK 262 cell lines after treatment with the proteasome inhibitor, bortezomib was only 2% and 1%, respectively (Figure 5E). Bortezomib had a significant cytotoxic effect against FH-null cells compared to FH-null cells without treatment (p<0.0001). However, the specificity of bortezomib for NCCFH1 cells was low as the viability of normal kidney cells treated with bortezomib was even lower at 0.5%.
To examine the sensitivity of NCCFH1 cells to bortezomib at pharmacologically relevant concentrations, an assay was carried out to determine the 50% inhibitory concentration (IC50), where the bortezomib concentration used ranged from 0.001 to 10 μM. We chose bortezomib because of its high efficacy against NCCFH1 cells compared to the other tested drugs. We found that the IC50 was in the range 0.1-0.01 μM (Figure 5F).
Discussion
HLRCC is an inherited, rare autosomal dominant disorder due to FH gene mutations. It results in chronic accumulation of fumarate and alteration of different intermediates of the tricarboxylic acid cycle. This cycle plays a crucial role in the bioenergetics of the cell. The rarity of this disorder limits the number of preclinical studies as well as the number of clinical trials. Yang et al. established and described the FH−/− cell lines UOK 262 and UOK 268 derived from patients with HLRCC-associated kidney cancer (16, 17). These models harbor missense FH gene mutation and exhibit the Warburg phenotype.
In the present study, we established a new preclinical model of FH-deficient NCCFH1 cells, originating from metastatic HLRCC-associated kidney cancer. In this cell line, we also identified a germline frameshift FH gene mutation, with a loss of heterozygosity, which is located in exon 8, c.1162delA, p.Thr375fs. The detailed analysis of the nucleotide sequences after the frameshift mutation revealed a premature stop codon at codon 380 (ATG → TGA) in exon 8.
The morphology of NCCFH1 cells presents typical cancer cell characteristics, with an irregular cell form, abnormal branching, multiple intracytoplasmic vacuoles, large and oval nucleus with multiple nucleoli. Functionally, the relative mRNA expression of FH gene is dramatically low in these cells; protein expression and enzymatic activity of FH were not detected in NCCFH1 cells due to the premature stop codon after frameshift mutation. These NCCFH1 cells exhibit glucose-dependent growth and impaired oxidative phosphorylation. The OCR/ECAR ratio data show that this HLRCC cell line is dependent on glycolysis. Therefore, we can state that NCCFH1 represents a new model for studies of metabolic abnormalities and the Warburg effect in HLRCC.
Comparative gene-expression analysis between HLRCC cell lines and normal kidney cell line determined more than 100 genes as being differentially expressed in cancer cells. Comprehensive analysis of the microarray data identified a group of up-regulated genes (including TMEFF2, SRGN, RARA, ALDH1A1, CT45A3, RASEF, AKR1B10, and TACSTD2) and a group of down-regulated genes (including FN1, CAV1, CLDN11, EREG, and HSPA1A1). The specific overexpression of AKR1B10 gene reflects the ARE signature in these HLRCC cells and correlates with earlier published data by Ooi et al. (14). Furthermore, according to the QIAGEN database, the promoters for the most of these genes have binding sites for upstream transcription regulators such as NKX-homeodomain factor (NKX) family (for TMEFF2, RARA, TACSTD2, CT45A3, CAV1, and EREG); POU transcription factor family (for SRGN, ALDH1A1 and CT45A3) and signal transducer and activator of transcription (STAT) transcription factor family (for SRGN, RARA, RASEF, FN1, CAV1 and HSPA1A) (Table IV). These transcription factor families are involved in regulation of different processes such as metabolic and oxidative stress sensor regulation, immune function regulation, embryogenesis, proliferation, and cell signaling. It has been reported that the POU transcription family member OCT1 can reduce the level of glycolysis in the cell and promotes oxidative phosphorylation efficiency (29). Thus, our profiling data show that NCCFH1 cells exhibits a Warburg-like phenotype that is characterized by embryogenesis development programs with rapid cell proliferation. Finally, DAVID functional annotation clustering analysis of the microarray data showed deregulation of many important signaling pathways and processes, such as apoptosis regulation, GTPase and retrovirus-associated DNA sequences (RAS) activities, defense response, steroid and lipid metabolism.
The latest discoveries in molecular genetics of RCC have led to the development of the targeted therapies that include using anti-angiogenic agents such as sorafenib, sunitinib and bevacizumab, as well as mTOR inhibitors such as everolimus and temsirolimus (30-33). The current recommendations for targeted therapies are mainly related to clear-cell RCC histology, which is the most frequent subtype of RCC. Therefore, more than half of the clinical trials have been performed for this group of patients. However, there is no established, efficient therapy specifically designed for the rare PRCC2. The targeting drugs sunitinib, sorafenib and temsirolimus are currently considered as the standard option to cure inherited and sporadic forms of PRCC2 (34).
Our study is the first where HLRCC cells were treated with a large number of anti-neoplastic drugs to see whether one or more of them would trigger a response.
Our results show that HLRCC cells resist most tested anti-neoplastic drugs. However, we found that three drugs, namely mitoxantrone, epirubicin and topotecan, markedly affect both NCCFH1 and UOK 262 cell viability. These US Food and Drug Administration-approved topoisomerase inhibitors are used in chemotherapy. Mitoxantrone is an anthracene dione commonly used in the treatment of leukemia, of breast and prostate cancer, and of multiple sclerosis (35-39). Epirubicin is an anthracycline and is used in the therapy of breast cancer (40). Topotecan is a derivative of campothecin used in the therapy of small-cell lung cancer, and of ovarian and cervical cancer (41-44).
We found that the cytotoxic effect on HLRCC cells was much more pronounced with mitoxantrone than with epirubicin and topotecan.
Previous studies have shown that by targeting topoisomerase II, mitoxantrone induces double-strand DNA breaks. In HL60 cells (promyelocytic cell line), this leads to the activation of NF-κB, with subsequent induction of apoptosis (45). Ferrer et al. demonstrated the high sensitivity of mantel cell lymphoma cells to mitoxantrone treatment. This sensitivity is due to activation of the mitochondrial apoptosis-signaling pathway due to the loss of mitochondrial transmembrane potential. In addition, the functional integrity of DNA-damage response genes is required for the cytotoxic effect of mitoxantrone in these cells (46). Karl et al. showed that NF-κB triggers tumor suppression after mitoxantrone treatment. In glioblastoma cells, drug-mediated NFκB activation increases DNA damage and activates a pro-apoptotic pathway as a response to mitoxantrone treatment (47). Further research should be carried out to determine the potential mechanisms of mitoxantrone action on HLRCC cells.
Interestingly, the proteasome inhibitor bortezomib, used in myeloma and lymphoma treatment, had a strong cytotoxic effect on FH-deficient cells in a pharmacologically relevant concentration range, from 0.1 to 0.01 μM. Latest studies have already established that bortezomib acts not only as an inhibitor of several signaling pathways including NF-κB and angiogenesis, but also as an activator of apoptosis and the p38 mitogen-activated protein kinase pathway (48-52).
Previous studies have shown that the ARE axis is strongly deregulated in both inherited and sporadic forms of PRCC2, where activation of ARE-controlled genes, such as AKR1B10, is due to deregulation of the NRF2–KEAP1 axis (14, 15). It has been shown that fumarate acts as an oncometabolite, inducing succination of KEAP1 and subsequent stabilization of NRF2. NRF2 is a transcriptional factor that activates the expression of genes involved in the antioxidant response. In addition, activation of the NF-κB pathway in PRCC2 has been suggested as the result of decreasing KEAP1 function, that is, a negative regulator of the NF-κB pathway. Sourbier et al. described the efficacy of bortezomib in combination with cisplatin against UOK 262 cell line and an in vivo model of HLRCC associated kidney cancer (53). Therefore, inhibition of the NFκB pathway could be a potential explanation for the sensitivity of FH-deficient cells to bortezomib treatment.
In conclusion, this novel FH-deficient cell line NCCFH1 represents a preclinical in vitro model and exhibits the Warburg phenotype. This cell line provides a basis on which to study the oncogenic features of hereditary PRCC2. Our findings show the highly, cytotoxic effect of bortezomib against FH-deficient cell lines. Further studies need to be done to determine whether bortezomib could represent a potential therapy for metastatic HLRCC.
Acknowledgements
We would like to thank the participating patient and his family members. We also acknowledge W. Marston Linehan, M.D., the Urologic Oncology Branch Tumor Cell line Repository, National Cancer Institute at Bethesda, MD, USA, for providing the UOK 262 cell line; Li Shi Lim, Institutes of Bioengineering and Nanotechnology, Singapore, Republic of Singapore; Kesavan Sittampalam, Department of Pathology, Singapore General Hospital, Singapore, Republic of Singapore; Dorine Bonte, Gustave Roussy Cancer Campus, Villejuif, France.
- Received April 28, 2015.
- Revision received September 4, 2015.
- Accepted September 25, 2015.
- Copyright© 2015 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved
