SEARCH

SEARCH BY CITATION

Keywords:

  • thymidine kinase;
  • thymidylate synthase;
  • thymidine phosphorylase;
  • dihydropyrimidine dehydrogenase;
  • epithelial ovarian cancer;
  • radioenzymatic assay;
  • capecitabine

Abstract

  1. Top of page
  2. Abstract
  3. MATERIAL AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

TK is a pyrimidine metabolic pathway enzyme involved in salvage DNA synthesis. What roles TK may play in epithelial ovarian cancer and the relationships between TK and the other pyrimidine pathway enzymes remain unclear. We examined TK1 gene expression by RT-PCR and related it to gene expression of TS, TP and DPD in 69 samples from epithelial ovarian cancer, 8 low-malignant-potential tumors, 16 benign ovarian tumors and 34 normal ovaries. Additionally, cytosolic and serum TK activities were determined by radioenzymatic assay. TK1 gene expression, the ratio of TK1 to TS gene expression, that of TK1 to TP and that of TK1 to DPD were significantly higher in epithelial ovarian cancer than in normal ovaries. In epithelial ovarian cancer, TK1 gene expression correlated with cytosolic and serum TK activities, TS and TP gene expression and the ratio of TP to DPD gene expression. Patients with high-TK1 gene expression had a significantly poorer survival than those with low TK1 gene expression. Combined analysis demonstrated that the relative risk of cancer death for tumors with high TK1, high TS and high TP gene expression was greater than that for tumors with high TK1 gene expression alone. TK1 gene expression together with TS, TP and DPD gene expression may play important roles in influencing the malignant behavior of epithelial ovarian cancer. Combination therapy including TK inhibitor is a possible therapeutic intervention in patients with epithelial ovarian cancer. © 2002 Wiley-Liss, Inc.

TK is an important pyrimidine metabolic pathway enzyme, which catalyzes the phosphorylation of thymidine to deoxythymidine monophosphate using either exogenous or endogenous metabolic thymidine as a substrate.1 This salvage pathway is the only one by which thymidine is introduced into DNA synthesis. Two isoforms of the TK gene (TK1 and TK2) have been described.2, 3 In addition, a modification of the sequence for TK1 has been described.4 TK1, which is located in the cytoplasm, is a cell cycle–regulated enzyme.5 Its activity fluctuates with DNA synthesis, being high in dividing and malignant cells and low in quiescent cells.6, 7 TK2 is considered to be located in the mitochondria; levels of TK2 remain relatively constant throughout the cell cycle. Clinical studies have reported elevated serum TK activity in various human malignancies.8 The majority of these studies concerned hematologic malignancies, but serum TK activity was also elevated in patients with epithelial ovarian cancer.9 Additionally, serum and cytosolic TK activities appear to have prognostic value in some solid tumors, such as breast,7, 10, 11 small cell lung,12 prostate13 and cervical14 cancers. However, little is known about the clinical and prognostic value of tissue TK gene expression in human cancers.

TS, TP, DPD and TK are pyrimidine pathway enzymes for de novo DNA synthesis or salvage nucleotide synthesis. TS is an enzyme for the de novo synthesis of deoxythymidine monophosphate (the same product as TK) and is associated with cell proliferative activity.15, 16 TP catalyzes the reversible phosphorolysis of thymidine (the same substrate as TK) to thymine and 2-deoxy-D-ribose-1 phosphate17 and has angiogenic and antiapoptotic effects.18, 19 Subsequently, thymine is degraded to dihydrothymine by DPD, which is inversely related to malignant behavior.20, 21 These 3 enzymes are also responsible for the efficacy and metabolism of fluoropyrimidine cytostatic agents. TS is a critical target enzyme of these drugs.15 The new agent capecitabine (N4-pentyloxycarbonyl-5′-deoxy-5-fluorocytidine) is converted selectively in tumors to active 5-FU by TP and subsequently 5-FU is inactivated to dihydrofluorouracil by DPD.22 The efficacy of capecitabine has been correlated with the ratio of TP to DPD activity in xenograft models of various human cancer cell lines because high levels of active 5-FU are selectively produced in tumors with a high ratio of TP to DPD activity.22 Previously, we demonstrated that high TS gene expression and a high ratio of TP to DPD gene expression (TP/DPD) correlated significantly with poor survival in patients with epithelial ovarian cancer and suggested that capecitabine might be used in these patients.23, 24 Although TK together with TS, TP and DPD cooperatively may affect malignant behavior and may influence the efficacy of this drug, the relationship between TK and these 3 enzymes in patients with malignant ovarian tumors remains unclear.

In our study, we examined TK1 gene expression by RT-PCR and cytosolic and serum TK activities by radioenzymatic assay in normal ovaries and benign and malignant epithelial ovarian tumors as well as the relationship between TS, TP and DPD gene expression and clinicopathologic factors.

MATERIAL AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIAL AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Patients

Our study included 69 patients with epithelial ovarian cancer, 8 patients with low-malignant-potential tumors and 16 patients with benign ovarian tumors. Patients were excluded preoperatively if they had undergone any therapy, had multiple cancers or had severe complications. All research was conducted with informed consent. Patients with epithelial ovarian cancer were staged according to the 1987 FIGO criteria.25 The FIGO staging system assumes that an adequate staging operation has been performed and was described elsewhere.26 Tumors were classified histologically according to the WHO criteria.27 All patients with epithelial ovarian cancer received cisplatin-based chemotherapy as an adjuvant treatment and were followed to evaluate survival. No patients received fluoropyrimidine-containing regimens. Eight low-malignant-potential tumors were stage I, of which 6 were mucinous and 2 serous. Patients with low-malignant-potential tumors were excluded from the survival analysis. Benign ovarian tumors consisted of 8 cases of serous cystadenoma, 4 of mucinous cystadenoma and 4 of endometrial cyst. Thirty-three normal ovarian specimens obtained from women who underwent oophorectomy for nonovarian conditions were examined as controls. Normal controls included 2 cases in the menstrual phase, 4 in the proliferative phase, 10 in the secretory phase and 17 postmenopausal. For analysis of serum TK activity, 32 healthy normal women were included as controls.

Tissue specimens and RNA preparation

A block of fresh tissue, which was prepared to eliminate inappropriate components, was snap-frozen in liquid nitrogen as close as possible to the time of surgical removal and stored at –80°C. A portion of the frozen tissue from each case (tumor and normal) was processed to create formalin-fixed and paraffin-embedded tissue blocks, which were matched and oriented relative to the frozen tissue.

Conventional RT-PCR for TK1 gene expression

Conventional RT-PCR for determination of TK1 gene expression was performed according to a method previously described.26 Briefly, cDNA was prepared by random priming from 500 ng of total RNA using the First-Strand cDNA Synthesis kit (Pharmacia-LKB, Uppsala, Sweden). Pairs of oligonucleotide primers for PCR were designed to insert an intron in the corresponding genomic sequence, to eliminate amplification from genomic DNA. Primers for human TK1 gene (GenBank accession number K02581) amplification were CAGCTTCTGCACACATGACC (upstream) and AGTGCAGCCACAATTACGG (downstream) and the expected PCR product was 183 bp. PCR was carried out in a thermal cycler (Perkin-Elmer Cetus, Norwalk, CT) with a mixture consisting of cDNA derived from 5 ng of RNA, 10 pmol of the upstream and downstream primers for the TK1 gene and 5 pmol of primers for the β2-MG gene (GenBank accession number U00567; upstream primer ACCCCCACTGAAAAAGATGAG, downstream primer ATCTTCAAACCTCCATGATGC, producing a 120 bp fragment), 200 μmol of each deoxynucleotide triphosphate, 37 kBq of [α-32P]dCTP and 0.1 U of Taq DNA polymerase with reaction buffer (Life Technologies, Rockville, MD) in a final volume of 10 μl. PCR conditions were as follows: denaturation at 94°C for 1 min, annealing at 58°C for 1 min and extension at 72°C for 1 min. PCR products were separated on 9% polyacrylamide gels. The amount of radioactivity in the gel was determined using a BAS-2000 Bioimage Analyzer (Fujix, Tokyo, Japan). Relative gene expression of TK1 was calculated by determining the ratio between the amount of the radiolabeled PCR product within the linear range of the TK1 and the β2-MG genes. To determine the number of PCR cycles appropriate for quantification, PCR was performed for 20–40 cycles in increments of 2 cycles. The slope of the linear region of both the β2-MG and TK1 amplifications was obtained from 26 to 32 cycles and the expression ratios of TK1 to β2-MG were reasonably constant (Fig. 1). Therefore, in the subsequent experiments, the values at 30 PCR cycles were defined as the expression levels of the target genes. Each TK1 gene expression value reported was the mean from at least 3 independent RT-PCR experiments.

thumbnail image

Figure 1. The slope of the linear regions of amplification of the TK1 and β2-MG genes. PCR amplification included 26, 28, 30 and 32 cycles. The exponential slopes of both the β2-MG and the TK1 amplifications were obtained from 26–32 cycles. TK1 gene expression levels were calculated as the relative yield of the TK1 gene to that of the β2-MG gene.

Download figure to PowerPoint

The RT-PCR method for TS, TP and DPD gene expression was previously described.23, 24, 26 Levels of TS, TP and DPD gene expression were reported in most patients with malignant ovarian tumors and normal ovaries.

Real-time quantitative RT-PCR for TK1 gene expression

TK1 gene expression was determined by real-time quantitative RT-PCR in 5 epithelial ovarian cancers, 7 benign ovarian tumors and 4 normal ovaries. Real-time quantitative PCR was performed using the hybridization probe format LightCycler system (Roche, Hamburg, Germany). It uses 2 independent, single-labeled oligonucleotides that hybridize adjacently on the amplicon internal to the flanking PCR primers. One oligonucleotide is labeled at its 3′ end with fluorescein and the other oligonucleotide is labeled at its 5′ end with LightCycler Red640 and at its 3′ end with phosphorylation to prevent extension. These 2 probes were utilized to detect the specific PCR products on the principle of fluorescence resonance energy transfer. Roughly speaking, during the PCR annealing step, these 2 probes hybridize, head-to-tail, to adjacent regions of the specific PCR products. The fluorophores, which are directly coupled to the probes, are very close in the hybrid structure (1 bp between probes). The donor fluorophore (fluorescein) is excited by an external light source, then passes on part of its excitation energy to the adjacent acceptor fluorophore (LC Red640). The excited LC Red640 then emits measurable light. PCR for TK was performed using 2 μl mastermix (LightCycler FastStart DNA Master Hybridization Probes) containing buffer, dATP, dCTP, dGTP, dUTP and Taq polymerase; 3 mM MgCl2; 0.2 μM fluorescein probe, 0.4 μM LC Red640 probe and 0.5 μM of each primer (Nihon Gene Research, Tokyo, Japan); and cDNA and water to a final volume of 20 μl. The PCR protocol was predenaturation at 95°C for 10 min, then for PCR cycles denaturation at 95°C for 15 sec, annealing at 60°C for 15 sec (detection), extension at 72°C for 7 sec (45 cycles) and cooling at 40°C for 30 sec. For TK1, the primers used in PCR were (forward) 5′-CAGCTTCTGCACACATGACC-3′ and (reverse) 5′-AGTGCAGCCACAATTACGG-3′. Hybridization probes were a fluorescein probe, 5′-CATCGACGAGGGGCAGTTTTTC-3′ and LC Red640 probe, 5′-CTGACATCATGGAGTTCTGCGAGG-3′. PCR for GAPDH (GenBank accession M33197) used the same conditions except for extension at 72°C for 12 sec. For GAPDH, the primers used in PCR were (forward) 5′-TGAACGGGAAGCTCACTGG-3′ and (reverse) 5′-TCCACCACCCTGTTGCTGTA-3′. Hybridization probes were fluorescein probe, 5′-TCAACAGCGACACCCACTCCT-3′ and LC Red640 probe, 5′-CACCTTTGACGCTGGGGCT-3′.

Radioenzymatic assay for cytosolic and serum TK activities

The area surrounding frozen tissues analyzed for TK1 gene expression was used for determination of cytosolic TK activity in 19 epithelial ovarian cancers, 3 benign ovarian tumors and 2 normal ovaries. Cytosols used for TK activity assays were prepared in TRIS homogenization buffer (10 mM TRIS-HCl, 1.5 mM Na2EDTA, 0.5 mM DTT, 10 mM sodium molybdate and 10% glycerol, pH 7.4), as described previously by Romain et al.28 Cytosolic TK activity was determined by the method of Gronowitz et al.,29 using the commercially available radioenzymatic assay Prolifigen TK kit (Daiichi, Tokyo, Japan). Samples were incubated with ATP and 125I-labeled iododeoxyuridine as substrate at 37°C for 4 hr. The enzymatic reaction was then stopped by addition of aluminum oxide tablets, which bind only the radioactive products. The unreacted substrate was washed off and the bound radioactivity counted in a γ-counter. TK activity was calculated from a calibration curve made from a TK standard. Total protein concentration was measured by the method of Lowry et al.30 TK levels are expressed in units per gram protein.

Blood samples were taken before surgery from 24 patients with epithelial ovarian cancer and 32 normal women as controls. These were immediately centrifuged at 800g for 10 min and sera were stored at –80°C until assay. Serum TK activity was evaluated using the Prolifigen TK kit and expressed in units per liter. The cut-off level was 5 U/l, according to the manufacturer's recommendation (low serum TK activity vs. high serum TK activity).

Statistical analysis

The Mann-Whitney U-test (for comparison of 2 groups) or Kruskal-Wallis 1-way ANOVA and Bonferroni's multiple comparison test (for comparison of more than 2 groups) was used to evaluate nonparametric numeric data. The χ2 test or Fisher's exact test was used for comparison of categorical data. The correlation coefficient (r2) between different parameters was determined by simple regression analysis. Survival curves were plotted using the Kaplan-Meier method31 and the log-rank test was used to determine the difference between life tables. Factors significant by univariate analysis were included in the Cox proportional hazards model in multivariate analysis to identify independent factors influencing survival. A value of p < 0.05 was considered statistically significant.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIAL AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Correlation between TK1 gene expression measured by conventional RT-PCR and real-time quantitative RT-PCR

We examined the correlation between TK1 gene expression measured by conventional RT-PCR and real-time quantitative RT-PCR. There was a strong correlation between these 2 parameters (r2 = 0.84, p < 0.0001; Fig. 2).

thumbnail image

Figure 2. Scatter diagram of TK1 gene expression measured by conventional RT-PCR and real-time quantitative RT-PCR in normal ovaries and ovarian tumors. There was a strong correlation between the 2 parameters (p < 0.0001). Solid circles, epithelial ovarian cancer; open circles, benign ovarian tumor; open squares, normal ovary.

Download figure to PowerPoint

Relationship between TK1 gene expression and TS, TP and DPD gene expression

TK1 gene expression measured by conventional RT-PCR was significantly higher in epithelial ovarian cancer than in normal ovary and benign ovarian tumor (p < 0.0001 and p < 0.001, respectively; Table I). Similarly, expression of both the TS and TP genes was significantly elevated in epithelial ovarian cancer (p < 0.0001 and p < 0.001, respectively), while DPD gene expression was significantly lower in epithelial ovarian cancers (p < 0.0001). For analyses of the relationship between TK1 gene expression and TS, TP and DPD gene expression under various ovarian conditions, we calculated the following ratios: TK1/TS, TK1/TP and TK1/DPD. Epithelial ovarian cancer had significantly higher TK1/TS, TK1/TP and TK1/DPD ratios than did normal ovary (p < 0.0001 and p < 0.01, respectively; Table II).

Table I. Gene Expression of TK1, TS, TP and DPD in Ovarian Tumors and Normal Ovaries
 Number of patientsTK1 expressionTS expressionTP expressionDPD expression
Median (range)Median (range)Median (range)Median (range)
  1. Gene Expression of TK1, TS, TP and DPD was significantly different in the 4 ovarian conditions (all p < 0.0001 by Kruskal-Wallis 1-way ANOVA); Bonferroni's multiple comparison test was used to evaluate the difference in each gene expression in the 4 ovarian conditions. TK1:1–3p < 0.05, 1–4p < 0.0001; 2–4p < 0.001; TS:5–7p < 0.0001, 6,7p < 0.001; TP:8–11p < 0.0001, 9–11p < 0.001, 10,11p < 0.01; DPD:12,13p < 0.01, 12–14p < 0.001, 12–15p < 0.0001.

Normal ovary340.02 (0–0.07)10.34 (0.14–0.71)50.33 (0.09–0.96)80.45 (0.10–1.19)12
Benign ovarian tumor160.03 (0.01–0.10)20.29 (0.06–0.67)60.34 (0.19–0.79)90.29 (0.12–0.62)13
Low malignant potential80.08 (0.02–0.18)30.36 (0.28–0.62)0.42 (0.25–0.94)100.20 (0.04–0.29)14
Epithelial ovarian cancer690.10 (0.02–0.56)40.53 (0.20–1.51)70.95 (0.19–5.38)110.19 (0.03–1.39)15
Table II. Relationship among Gene Expression of TK1, TS, TP and DPD in Malignant Ovarian Tumors and Normal Ovaries
 Number of patientsTK1/TSTK1/TPTK1/DPD
Median (range)Median (range)Median (range)
  1. TK1/TS, TK1/DPD and TK1/TP were significantly different in the 4 ovarian conditions (p < 0.0001 to p < 0.01 by Kruskal-Wallis 1-way ANOVA); Bonferroni's multiple comparison test was used to evaluate the difference in each ratio in the 4 ovarian conditions. TK1/TS:1,2p < 0.01, 1–3p < 0.0001; TK1/TS:4–6p < 0.0001; TK1/DPD:7–9p < 0.0001, 8,9p < 0.01.

Normal ovary340.09 (0.01–0.25)10.08 (0.02–0.21)40.06 (0.02–0.68)7
Benign ovarian tumor160.13 (0.01–0.51)0.12 (0.03–0.30)0.15 (0.02–0.79)8
Low malignant potential80.19 (0.06–0.33)20.16 (0.07–0.40)50.38 (0.23–1.39)
Epithelial ovarian cancer690.19 (0.04–0.64)30.13 (0.01–0.58)60.51 (0.07–5.32)9

In epithelial ovarian cancer, TK1 gene expression correlated strongly with TS gene expression (r2 = 0.50, p < 0.0001; Fig. 3a) and weakly with TP gene expression (r2 = 0.09, p < 0.05; Fig. 3b). There was no correlation between TK1 gene expression and DPD gene expression (r2 = 0.02, p = 0.23; Fig. 3c).

thumbnail image

Figure 3. Scatter diagram of expression of the TK1, TS, TP and DPD genes in epithelial ovarian cancer. There were significant correlations between expression of the TK1 gene and expression of the TS (a) and TP (b) genes (p < 0.0001 and p < 0.05, respectively) but no correlation between TK1 and DPD (c) gene expression.

Download figure to PowerPoint

TP/DPD in 69 epithelial ovarian cancers ranged 0.54–58.73, with a median value of 5.03. The median (range) of TK1 gene expression was 0.11 (0.03–0.56) in epithelial ovarian cancers showing high TP/DPD (>median) and 0.08 (0.02–0.22) in those showing low TP/DPD (≤median). TK1 gene expression was significantly higher in epithelial ovarian cancers showing high TP/DPD than in those showing low TP/DPD (p < 0.05).

Relationship between TK1 gene expression and clinicopathologic factors

The characteristics of the 69 patients with epithelial ovarian cancer are summarized in Table II. TK1 gene expression was significantly lower in clear cell carcinoma than in serous cystadenocarcinoma and undifferentiated adenocarcinoma (both p < 0.05, Table III). Advanced clinical stage tended to correlate with elevated TK1 gene expression, but the difference was not significant (p = 0.06).

Table III. Relationship between TK1 Gene Expression and Clinicopathologic Factors in Patients with Epithelial Ovarian Cancer
FactorsNumber of patientsTK1 gene expression
MedianRange
  • 1

    TK1 gene expression tended to be higher in advanced clinical stage, but the difference was not significant (p = 0.06 by Kruskal-Wallis 1-way ANOVA).

  • 2

    TK1 gene expression was significantly different in the 5 histologic types (p < 0.01 by Kruskal-Wallis 1-way ANOVA); Bonferroni's multiple comparison test was used to evaluate the difference in TK1 gene expression in the 5 histologic types.

  • 3

    p < 0.01.

Total690.100.02–0.56
Patient age (years)
 ≤53 (median)350.080.03–0.45
 >53340.110.02–0.56
Clinical stage1
 I170.060.03–0.18
 II60.090.02–0.14
 III380.110.03–0.45
 IV80.130.03–0.56
Histologic grade
 Well150.070.02–0.19
 Moderate290.080.03–0.45
 Poor250.140.03–0.56
Histologic type2
 Serous3330.110.02–0.56
 Mucinous140.120.04–0.20
 Endometrioid130.070.04–0.45
 Clear cell360.040.03–0.07
 Undifferentiated330.110.11–0.35

Relationship between TK1 gene expression and cytosolic and serum TK activities

There was a positive correlation between cytosolic TK activity by radioenzymatic assay and TK1 gene expression in all specimens (r2 = 0.55, p < 0.0001; Fig. 4) and in epithelial ovarian cancer (r2 = 0.46, p < 0.01). Serum TK activity was significantly higher in patients with epithelial ovarian cancer (median 6.6 U/l, range 3.8–14) than in normal women (median 4.2 U/l, range 3.2–9.0; p < 0.001). The percentage of high serum TK activity (>5 U/l) was 67% (16/24) in patients with epithelial ovarian cancer. TK1 gene expression was significantly higher in epithelial ovarian cancer patients with high serum TK activity (median 0.13, range 0.04–0.45) than in those with low serum TK activity (median 0.05, range 0.03–0.16; p < 0.05).

thumbnail image

Figure 4. Scatter diagram of TK1 gene expression and cytosolic TK activity in normal ovaries and ovarian tumors. A weakly positive correlation was observed between TK1 gene expression and cytosolic TK activity in 24 specimens (p < 0.0001) and in 19 cases of epithelial ovarian cancer (p < 0.01). Solid circles, epithelial ovarian cancer; open circles, benign ovarian tumor; open squares, normal ovary.

Download figure to PowerPoint

Relationship between expression of the TK1, TS and TP genes and survival

For survival analyses according to TK1, TS, TP and DPD gene expression, we used the median value as the cut-off point to divide tumor patients into 2 groups (high expression vs. low expression). Patients with high TK1 gene expression had a significantly poorer survival than those with low TK1 gene expression (p < 0.05, Fig. 5). Moreover, high TS gene expression (p < 0.05), high TP gene expression (p < 0.05), clinical stage III–IV (p < 0.01) and positive residual disease (p < 0.05) significantly correlated with poor survival on univariate analysis, whereas patient age, histologic type, histologic grade and DPD gene expression did not. The 5 factors found to be significant by univariate analysis were subjected to multivariate analysis by Cox's proportional hazards model (Table IV). However, neither TK1 gene expression, TS gene expression nor TP gene expression alone was an independent prognostic factor. We then performed a combined analysis for TK1, TS and TP gene expression with respect to survival. The relative risk of death for 18 patients with high TK1, high TS and high TP gene expression was greater than that for high TK1, high TS or high TP gene expression alone (p < 0.0001, Fig. 6). The distribution of TK1 gene expression in these 18 patients (median 0.18, range 0.11–0.56) was similar to that in 34 patients with high-TK1 gene expression. On multivariate analysis, the gene expression status of TK1, TS and TP and clinical stage were found to be independent prognostic factors, whereas status of residual disease was not (Table IV).

thumbnail image

Figure 5. Kaplan-Meier survival curve in patients with epithelial ovarian cancer according to TK1 gene expression. Patients with high TK1 gene expression (>0.10 median) had a significantly poorer prognosis than those with low TK1 gene expression (≤0.10 median, p < 0.05).

Download figure to PowerPoint

Table IV. Univariate and Multivariate Analyses of Prognostic Factors in Patients with Epithelial Ovarian Cancer
Prognostic factorsNumber of patientsUnivariate1Multivariate 12Multivariate 23
pHazard ratio (95% CI)pHazard ratio (95% CI)p
  • CI, confidence interval; NA, not analyzed; NS, not significant.

  • 1

    Log-rank test.

  • 2

    Factors included in Cox's proportional hazard model were FIGO stage, status of residual disease, TK1 gene expression, TS gene expression and TP gene expression.

  • 3

    Factors included in Cox's proportional hazard model were FIGO stage, status of residual disease and gene expressions of TK1, TS and TP.

FIGO stage
 I–II23<0.016.5 (1.1–51.0)<0.052.4 (1.0–10.4)<0.05
 III–IV46
Status of residual disease
 Negative41<0.051.6 (0.5–4.8)NS1.2 (0.7–2.1)NS
 Positive28
TK1 gene expression
 Low (≤0.10 median)35<0.051.6 (0.5–5.6)NSNA
 High (>0.10)34
TS gene expression
 Low (≤0.53 median)35<0.051.8 (0.5–7.0)NSNA
 High (>0.53)34
TP gene expression
 Low (≤0.95 median)35<0.051.4 (0.4–4.9)NSNA
 High (>0.95)34
Gene expressions of TK1, TS and TP
 High18<0.0001NA1.8 (1.0–3.0)<0.05
 All others51
thumbnail image

Figure 6. Kaplan-Meier survival curve in patients with epithelial ovarian cancer according to combined analysis for expression of the TK1, TS and TP genes. Patients with tumors that had high (>median) expression of the TK1, TS and TP genes had a significantly poorer survival than those that did not (p < 0.0001).

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIAL AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

In the current study, we found that gene expression of TK1, TS and TP was significantly elevated in epithelial ovarian cancer compared to normal ovary and benign ovarian tumor; however, DPD gene expression was significantly lower in epithelial ovarian cancer. Although details of the biologic effects remain unclear, the difference in expression for each gene between the normal ovary and epithelial ovarian cancer was 1.6- to 5-fold. TK and TS are highly expressed in rapidly proliferating tissues, including malignant cells, because of their DNA replication functions.16, 29 TP is elevated in cancer tissues showing such typical microenvironmental characteristics as hypoxia and low pH.32, 33 Downregulation of DPD has been reported in some malignant conditions, such as cancers of the liver and colon.34, 35 The balance between salvage and de novo pathways differed according to the state of the ovary. The TK1/TS, TK1/TP and TK1/DPD ratios progressively increased from normal ovary to malignant ovarian tumor and were significantly higher in epithelial ovarian cancer than in normal ovary. Look et al.34 reported that TK activity measured by biochemical assay was 140-fold higher than TS activity in tissue extracts of 13 human epithelial ovarian cancers. Our and their results suggested that the salvage DNA synthesis pathway is predominant over the de novo DNA synthesis or salvage nucleotide synthesis pathways in the malignant ovary. In epithelial ovarian cancer, a strong positive correlation between TK1 and TS gene expression was observed. The TK1 and TS genes may be controlled by several similar mechanisms in epithelial ovarian cancer.

Although serum and cytosolic TK activities in human malignancies have been widely investigated, little is known about the clinical values of TK gene expression. We found that TK1 gene expression correlated significantly with histologic type (lower in clear cell carcinoma) and survival but not with other generally accepted prognostic factors such as FIGO stage, histologic grade and patient's age. The prognostic effect of TK is thought to depend mainly on rapid tumor cell proliferation since TK is involved in salvage DNA synthesis.29 Cytosolic TK activity has been strongly associated with the S-phase fraction by flow cytometry in breast cancer.11 High growth fraction, which was mainly derived from the S-phase fraction, has been shown to be an adverse prognostic factor for patients with epithelial ovarian cancer.37 Additionally, Scanlon et al.38 reported that the cisplatin-resistant human colon carcinoma cell line HCT8DDT exhibited a 3.2-fold increase in TK gene expression compared to its cisplatin-sensitive parental cell line (HCT8S), suggesting that cisplatin resistance in this cell line is associated with enhanced TK gene expression, which may increase the capacity for repair of cisplatin-induced damage by its DNA replication functions. Increased proliferative activity and resistance to cisplatin may have contributed to the negative prognostic effect of TK in patients with epithelial ovarian cancer.

In addition to TK1 gene expression, high expression of the TS and TP genes was a significant negative prognostic factor for patients with epithelial ovarian cancer.24, 26 The relative risk of cancer death from tumors with high expression of the TK1, TS and TP genes was greater than the risk from tumors with high expression of any of these genes alone. Although expression of each of these genes alone was not an independent prognostic factor by multivariate analysis, the combined gene expression status of these 3 enzymes was an independent predictor of survival. TS and TP play important roles in cellular metabolism, which are associated with tumor aggressiveness and prognosis. Ju et al.39 demonstrated that TS protein induced downregulation of p53 activity by decreasing the translational efficacy of p53 mRNA through a ribonucleoprotein complex. TP itself and its degeneration products (2-deoxy-D-ribose and thymine) have angiogenic and antiapoptotic effects.18, 19 Our previous studies on patients with ovarian cancer demonstrated significant correlations between TP gene expression and disordered angiogenesis measured by pulsed Doppler ultrasound38 and decreased apoptotic index.41 Salvage DNA synthesis and resistance to cisplatin by TK, de novo DNA synthesis and p53 downregulation by TS and the angiogenic and antiapoptotic properties of TP may cooperate to worsen the prognosis of patients with epithelial ovarian cancer.

We examined gene expression of target enzymes using a semiquantitative RT-PCR assay with the housekeeping gene β2-MG as an internal control for each sample. This method does not provide absolute quantification and we must consider the semiquantitative limitations. New technologies have became available that allow real-time and on-line monitoring of PCR products by measuring fluorescence in every cycle with a built-in microvolume fluorimeter.40 These techniques allow the measurement of PCR kinetics and more reliable quantification of gene expression. Although the number of pathologic tumor samples was small, we found that TK1 gene expression measured by our conventional RT-PCR strongly correlated with that measured by real-time quantitative RT-PCR. Additionally, there was a significant correlation between TK1 gene expression by conventional RT-PCR and cytosolic TK activity by radioenzymatic assay. RT-PCR for the determination of TK1 gene expression may be more convenient than radioenzymatic assay because it does not require large amounts of tumor tissue. Serum TK activity was significantly elevated in patients with epithelial ovarian cancer compared to normal women, in accordance with a previous report.9 High serum TK activity (>5 U/l), which was observed in 16/24 (67%) patients with epithelial ovarian cancer, correlated significantly with high TK1 gene expression. Serum TK activity, at least in part, reflects TK1 gene expression in tissues of epithelial ovarian cancer and may be a useful tumor marker in patients with epithelial ovarian cancer.

TK may affect the efficacy of the fluoropyrimidine cytostatic agent capecitabine. This drug can inhibit TS activity and its activity may be increased by a high ratio of TP to DPD activity.22 Our previous study demonstrated that a high TP/DPD ratio was an independent negative prognostic factor in patients with epithelial ovarian cancer, suggesting that patients whose tumors show a high TP/DPD ratio might particularly benefit from treatment with capecitabine.23 In epithelial ovarian cancer, we found a predominance of TK1 gene expression compared to TS gene expression, a positive correlation between TK1 and TS gene expression and an association between high TK1 gene expression and a high TP/DPD ratio. These results suggested that if the target enzyme TS could be inhibited by highly activated capecitabine in tumors showing a high TP/DPD ratio, high activity on the alternative salvage pathway by TK could circumvent the induced growth inhibition.28 TK together with TS, TP and DPD may play important roles in the malignant behavior of epithelial ovarian cancer. The camptothecin analogs, which are a new class of antitumor agents that target topoisomerase I and are widely used for the salvage treatment of ovarian cancer,43 potently inhibited TK in colon adenocarcinoma cells and leukemia cells.44 Taken together, a combination therapy consisting of camptothecin analogs and capecitabine may be a therapeutic intervention in patients with epithelial ovarian cancer showing high TK1 gene expression and a high TP/DPD ratio.

Acknowledgements

  1. Top of page
  2. Abstract
  3. MATERIAL AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

This work was supported by a Grant-in-Aid for Encouragement of Young Scientists (13770917) from the Japan Society for the Promotion of Science. We thank Ms. T. Yamada (Department of Obstetrics and Gynecology, Shimane Medical University) for management of medical records.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIAL AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES
  • 1
    Taylor AT, Stafford MA, Jones OW. Properties of thymidine kinase partially purified from human fetal and adult tissue. J Biol Chem 1972;247: 193035.
  • 2
    Harvey D, Bradshaw HD. Molecular cloning and cell cycle-specific regulation of a functional human thymidine kinase gene. Proc Natl Acad Sci USA 1983;80: 558891.
  • 3
    Johansson M, Karlsson A. Cloning of the cDNA and chromosome localization of the gene for human thymidine kinase 2. J Biol Chem 1997;272: 84548.
  • 4
    Berenstein D, Christensen JF, Kristensen T, Hofbauer R, Munch-Petersen B. 2000. Valine, not methionine, is amino acid 106 in human cytosolic thymidine kinase (TK1). J Biol Chem 2000;275: 3218792.
  • 5
    Adler R, McAuslan BR. Expression of thymidine kinase variants is a function of the replicative state of cells. Cell 1974;2: 1137.
  • 6
    Ellims PH, Van der Weyden MB, Medley G. Thymidine kinase isoenzymes in human malignant lymphoma. Cancer Res 1981;41: 6915.
  • 7
    O'Neill KL, Hoper M, Odling-Smee GW. Can thymidine kinase levels in breast tumors predict disease recurrence? J Natl Cancer Inst 1992;84: 18258.
  • 8
    Hallek M, Wanders L, Strohmeyer S, Emmerich B. Thymidine kinase: a tumor marker with prognostic value for non-Hodgkin's lymphoma and a broad range of potential clinical applications. Ann Hematol 1992;65: 15.
  • 9
    Hallek M, Touitou Y, Levi F, Mechkouri M, Bogdan A, Bailleul F, Senekowitsch R, Emmerich B. Serum thymidine kinase levels are elevated and exhibit diurnal variations in patients with advanced ovarian cancer. Clin Chim Acta 1997;267: 15566.
  • 10
    Robertson JF, O'Neill KL, Thomas MW, McKenna PG, Blamey RW. Thymidine kinase in breast cancer. Br J Cancer 1990;62: 6637.
  • 11
    Romain S, Bendahl PO, Guirou O, Malmstrom P, Martin PM, Fern M. DNA-synthesizing enzymes in breast cancer (thymidine kinase, thymidylate synthase and thymidylate kinase): association with flow cytometric S-phase fraction and relative prognostic importance in node-negative premenopausal patients. Int J Cancer 2001;95: 5661.
  • 12
    Gronowitz JS, Steinholtz L, Kallander CF, Hagberg H, Bergh J. Serum deoxythymidine kinase in small cell carcinoma of the lung. Relation to clinical features, prognosis and other biochemical markers. Cancer 1986;58: 1118.
  • 13
    Lewenhaupt A, Ekman P, Eneroth P, Nilsson B. Tumour markers as prognostic aids in prostatic carcinoma. Br J Urol 1990;66: 1827.
  • 14
    Fujiwaki R, Hata K, Moriyama M, Iwanari O, Katabuchi H, Okamura H, Miyazaki K. Clinical value of thymidine kinase in patients with cervical carcinoma. Oncology 2001;61: 4754.
  • 15
    Danenberg PV. Thymidylate synthetase—a target enzyme in cancer chemotherapy. Biochem Biophys Acta 1977;473: 7392.
  • 16
    Navalgund LG, Rossana C, Muench AJ, Johnson LF. Cell cycle regulation of thymidylate synthetase gene expression in cultured mouse fibroblasts. J Biol Chem 1980;255: 738690.
  • 17
    Iltzsch MH, el Kouni MH, Cha S. Kinetic studies of thymidine phosphorylase from mouse liver. Biochemistry 1985;24: 6799807.
  • 18
    Kitazono M, Takebayashi Y, Ishitsuka K, Takao S, Tani A, Furukawa T, Miyadera K, Yamada Y, Aikou T, Akiyama S. Prevention of hypoxia-induced apoptosis by the angiogenic factor thymidine phosphorylase. Biochem Biophys Res Commun 1998;253: 797803.
  • 19
    Moghaddam A, Zhang HT, Fan TP, Hu DE, Lees VC, Turley H, Fox SB, Gatter KC, Harris AL, Bicknell R. Thymidine phosphorylase is angiogenic and promotes tumor growth. Proc Natl Acad USA 1995;92: 9981002.
  • 20
    Canellakis ES. Pyrimidine metabolism—enzymatic pathways of uracil and thymidine degradation. J Biol Chem 1956;221: 31522.
  • 21
    Tuchman M, O'Dea RF, Ramnaraine ML, Mirkin BL. Pyrimidine base degradation in cultured murine C-1300 neuroblastoma cells and in situ tumors. J Clin Invest 1988;81: 42530.
  • 22
    Ishikawa T, Sekiguchi F, Fukase Y, Sawada N, Ishitsuka H. Positive correlation between the efficacy of capecitabine and doxifluridine and the ratio of thymidine phosphorylase to dihydropyrimidine dehydrogenase activities in tumors in human cancer xenografts. Cancer Res 1998;58: 68590.
  • 23
    Fujiwaki R, Hata K, Nakayama K, Fukumoto M, Miyazaki K. Gene expression for dihydropyrimidine dehydrogenase and thymidine phosphorylase influences outcome in epithelial ovarian cancer. J Clin Oncol 2000;18: 394651.
  • 24
    Fujiwaki R, Hata K, Nakayama K, Fukumoto M, Miyazaki K. Thymidylate synthase expression in epithelial ovarian cancer: relationship with thymidine phosphorylase expression and prognosis. Oncology 2000;68: 24752.
  • 25
    International Federation of Gynecology and Obstetrics. Changes in definitions of clinical staging for carcinoma of the cervix and ovary. Am J Obstet Gynecol 1989;156: 2634.
  • 26
    Hata K, Kamikawa T, Arao S, Tashiro H, Katabuchi H, Okamura H, Fujiwaki R, Miyazaki K, Fukumoto M. Expression of the thymidine phosphorylase gene in epithelial ovarian cancer. Br J Cancer 1999;79: 184854.
  • 27
    Serov SF, Scully RE, Sobin LH. International histological classification of tumors, 9. Histological typing of ovarian tumors. Geneva: WHO, 1973.
  • 28
    Romain S, Martin PM, Klijn JG, van Putten WL, Look MP, Guirou O, Foekens JA. DNA-synthesis enzyme activity: a biological tool useful for predicting anti-metabolic drug sensitivity in breast cancer? Int J Cancer 1997;74: 15661.
  • 29
    Gronowitz JS, Kallander FR, Diderholm H, Hagberg H, Pettersson U. Application of an in vitro assay for serum thymidine kinase: results on viral disease and malignancies in humans. Int J Cancer 1984;33: 512.
  • 30
    Lowry OH, Rosebrough NJ, Farr AL, Rano RI. Protein measurement with the folin phenol reagent. J Biol Chem 1951;193: 26575.
  • 31
    Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958;53: 45781.
  • 32
    Brown NS, Jones A, Fujiyama C, Harris AL, Bicknell R. Thymidine phosphorylase induces carcinoma cell oxidative stress and promotes secretion of angiogenic factors. Cancer Res 2000;60: 6298302.
  • 33
    Griffiths L, Dachs GU, Bicknell R, Harris AL, Stratford IJ. The influence of oxygen tension and pH on the expression of platelet-derived endothelial cell growth factor/thymidine phosphorylase in human breast tumor cells grown in vitro and in vivo. Cancer Res 1997;57: 5702.
  • 34
    Jiang W, Lu Z, He Y, Diasio RB. Dihydropyrimidine dehydrogenase activity in hepatocellular carcinoma: implication in 5-fluorouracil–based chemotherapy. Clin Cancer Res 1997;3: 3959.
  • 35
    McLeod HL, Sludden J, Murray GI, Keenan RA, Davidson AI, Park K, Koruth M, Cassidy J. Characterization of dihydropyrimidine dehydrogenase in human colorectal tumours. Br J Cancer 1998;77: 4615.
  • 36
    Look KY, Moore DH, Sutton GP, Prajda N, Abonyi M, Weber G. Increased thymidine kinase and thymidylate synthase activities in human epithelial ovarian carcinoma. Anticancer Res 1997;17: 23536.
  • 37
    Barnabei VM, Miller DS, Bauer KD, Murad TM, Rademaker AW, Lurain JR. Flow cytometric evaluation of epithelial ovarian cancer. Am J Obstet Gynecol 1989;162: 158490.
  • 38
    Scanlon KJ, Kashani-Sabet M, Sowers LC. Overexpression of DNA replication and repair enzymes in cisplatin-resistant human colon carcinoma HCT8 cells and circumvention by azidothymidine. Cancer Commun 1989;1: 26975.
  • 39
    Ju J, Pedersen-Lane J, Maley F, Chu E. Regulation of p53 expression by thymidylate synthase. Proc Natl Acad Sci USA 1999;96: 376974.
  • 40
    Hata K, Fujiwaki R, Nakayama K, Fukumoto M, Miyazaki K. Expression of TP and TIE2 genes in normal ovary with corpus luteum and in ovarian cancer: correlation with ultrasound-derived peak systolic velocity. Mol Hum Reprod 2000;6: 31923.
  • 41
    Hata K, Fujiwaki R, Maede Y, Nakayama K, Fukumoto M, Miyazaki K. Expression of thymidine phosphorylase in epithelial ovarian cancer: correlation with angiogenesis, apoptosis and ultrasound-derived peak systolic velocity. Gynecol Oncol 2000;77: 2634.
  • 42
    Heid CA, Stevens J, Livak KJ, Williams PM. Real time quantitative PCR. Genome Res 1996;6: 98694.
  • 43
    Coleman RL, Miller DS. Topotecan in the treatment of gynecologic cancer. Semin Oncol 1997;24: 5563.
  • 44
    Voeller DM, Grem JL, Pommier Y, Paull K, Allegra CJ. Identification and proposed mechanism of action of thymidine kinase inhibition associated with cellular exposure to camptothecin analogs. Cancer Chemother Pharmacol 2000;45: 40916.