Tumor Biol. DOI 10.1007/s13277-015-3723-5

RESEARCH ARTICLE

Association between mismatch repair gene and irinotecan-based chemotherapy in metastatic colon cancer Junli Ma 1 & Yan Zhang 1 & Hong Shen 2 & Linda Kapesa 1 & Wenqiang Liu 1 & Mengsi Zeng 1 & Shan Zeng 1

Received: 6 May 2015 / Accepted: 25 June 2015 # International Society of Oncology and BioMarkers (ISOBM) 2015

Abstract Mismatch repair (MMR) gene is closely related to the pathogenesis of colon cancer. This study aimed to evaluate the association between MMR status and efficacy of irinotecan-based chemotherapy. As a target of 5-FU, thymidylate synthase (TS) expression level might be influenced by irinotecan. Understanding whether this influence of TS is related with MMR status is helpful to the further exploration of the mechanism of irinotecan sensitivity in metastatic colon cancer with different MMR status. One hundred eightyfour patients with metastatic colon cancer receiving irinotecan-based chemotherapy for the first-line treatment were included. Correlations between MMR and clinicopathological characteristics and prognosis were determined. Two pairs of colon cancer cell lines (HCT-116-hMLH1Vector (deficient MMR, dMMR) versus HCT-116-hMLH1+ (proficient MMR, pMMR); SW480-shRNA-hMLH1 (dMMR) versus SW480-shRNA-Control (pMMR)) were established by regulating MMR status. Sensitivity of these cell lines to irinotecan was determined by MTT assay. Regulation of TS by irinotecan was evaluated by western blotting and quantitative real-time PCR assay. dMMR accounted for 18.5 % and was related with proximal colon cancer (p=0.005), poorly differentiated tumors (p=0.018) and favorable efficacy with a higher disease Electronic supplementary material The online version of this article (doi:10.1007/s13277-015-3723-5) contains supplementary material, which is available to authorized users. * Shan Zeng [email protected] 1

Department of Oncology, Xiangya Hospital, Central South University, Changsha, Hunan 410008, China

2

Institute of Medical Sciences, Xiangya Hospital, Central South University, Changsha, Hunan, China

control rate (DCR), a longer progression-free survival (PFS) and a trend of longer overall survival (OS). dMMR colon cancer cells were more sensitive to irinotecan. TS expression level was reduced more in dMMR cells after irinotecan treatment (p<0.05). Our study favors an increased sensitivity of irinotecan in colon cancer with dMMR status. MMR status may be a predictive biomarker of response to irinotecan-based chemotherapy in metastatic colon cancer. Keywords Colon cancer . Mismatch repair gene . Irinotecan . 5-Fluorouracil . Thymidylate synthase

Introduction Colon cancer is the third most common malignant tumor worldwide. The overall survival (OS) is still unsatisfactory and 5-year survival rate is no more than 10 % in the metastatic patients [1]. Although chemotherapy plays an important role in the treatment, the curative effects of cytotoxic drugs are limited and a significant proportion of patients do not response to chemotherapy. Thus, it is of important clinical values to determine the predictive markers of chemotherapy for individualized treatment. DNA mismatch repair (MMR) genes (hMLH1, hMSH2, hMSH6, and hPMS2) recognize and fix DNA single-base mismatches and small insertion/deletion loops during DNA replication or recombination. Loss of expression of hMLH1, hMSH2, hMSH6, or hPMS2 was defined as deficient MMR (dMMR). dMMR can lead to an accumulation of DNA replication errors and produce a genetic phenotype called microsatellite instability (MSI), which is characterized by a defect in any of the MMR genes and numerous insertion/deletion mutations of short tandem repeats (STR) termed microsatellites, which are spread over the whole genome. It has been reported

Tumor Biol.

that dMMR/MSI is closely related to the pathogenesis of colon cancer. Mutations in the MMR genes, usually hMSH2 or hMLH1, have been correlated with susceptibility to Lynch Syndrome. As a result of MMR gene (usually hMLH1) methylation, dMMR was observed in 10–20 % of patients with sporadic colorectal cancer (CRC) [2–4]. The dMMR/MSI colon cancers may display some distinct clinicopathological features, such as poor histological differentiation or mucous type, location in the proximal colon, tumor-infiltrating lymphocytes, and so on [5, 6]. It is uncertain whether dMMR/MSI is a predictive marker for the prognosis and chemotherapy response in colon cancer. A correlation between MMR and fluorouracil-based chemotherapy had been documented in some studies. Colon cancer with dMMR status usually exhibited resistance to 5-FU treatment [7–11]. Currently, irinotecan-based treatment is clinically applied as the first-line chemotherapy in metastatic colon cancer. The predictive value of dMMR in response to irinotecan-based chemotherapy has been less evaluated. Some reports [12–16] tried to ascertain the correlation between MMR and treatment efficacy of irinotecan in colon cancer. It seemed that dMMR tumors were more sensitive to irinotecan than proficient mismatch repair (pMMR) tumors [13–15]. However, some scholars held an opposite opinion [16]. As a result, the predictive value of dMMR in colon cancer needs to be further validated. Irinotecan-based chemotherapy in metastatic colon cancer mainly refers to combined chemotherapy of irinotecan and fluorouracil. These two drugs show a certain degree of synergy. Irinotecan had been documented to reduce thymidylate synthase (TS) expression, a critical target of 5-FU and enhance cytotoxicity of the latter in a sequence-dependent manner [17, 18]. As mentioned above, there are some relationship between 5-FU treatment and MMR status, so we have reasons to connect TS expression with MMR status. It is of important interest to explore whether regulation of TS by irinotecan is related with MMR status. If so, that will be helpful to the further exploration of the mechanism of irinotecan sensitivity in metastatic colon cancer with different MMR status. In a word, the purpose of this study was to evaluate the association between MMR status and treatment efficacy of irinotecan-based chemotherapy, as well as the predictive value and the possible molecular mechanisms of MMR in the patients of metastatic colon cancer.

Methods Patients and tissue specimens of metastatic colon cancer The tumor tissues were randomly collected from 219 patients with metastatic colon cancer between January 2008 and March 2011 in Xiangya Hospital, Central South University, China. They were all histologically confirmed as metastatic

colon cancer by pathological or combined imageological examination and subjected to irinotecan-based chemotherapy as the first-line treatment. Those who received combined targeted agents (n=17) and whose tumor tissues were insufficient for immunohistochemistry (IHC) analysis (n=11) were excluded. Exclusion criteria also included prior cytotoxic or radiation therapy (n=5) and severe cardiopulmonary disease or other systematic diseases precluding standard chemotherapy (n = 2). At last, 184 subjects were finally included (Supplementary Fig. Fig. S1). Among whom, 170 cases received FOLFIRI of a 2-week schedule (irinotecan 180 mg/m2 i.v. on day 1; 5- fluorouracil 400 mg/m2 bolus on day 1, followed by 2,400 mg/m2 i.v. 46 h). One received a 3-week schedule for XELIRI (irinotecan 180 mg/m2 i.v. on day 1; capecitabine 1 g/ m2 p.o. on day 1–14); 13 received irinotecan only (180 mg/m2 i.v. every 2 weeks). Chemotherapy doses were adjusted properly according to toxicities and the patient’s performance status. Collection of follow-up data Clinical follow-up data were obtained from the patients through telephone and outpatient clinic. Follow-up was terminated on June 2014. The followup period was defined as the interval between the date of diagnosis and that of the patient’s death or the last followup. Follow-up time ranged from 7 to 36 months, with a median follow-up time of 17.6 months. The data of deaths from other causes and the surviving patients at the last follow-up were all censored. Written informed consent was obtained from all recruited participants before entry into the study. The subjects’ responses to irinotecan-based chemotherapy were evaluated for patients with measurable disease according to World Health Organization (WHO) criteria. CEA level, computed tomography (CT) scan, and/or ultrasonography were performed every 2 months on average or when evidence of recurrence was confirmed by clinical examination. OS was defined as the time quantum from diagnosis to death or the last inquiry, while progression-free survival (PFS) was defined as the time interval between the first time of chemotherapy and the first evidence of recurrence. Except for the proportion of patients with progressive disease (PD) patients, the rest was recorded as disease control rate (DCR). IHC staining and scoring IHC was performed on the total of 184 formalin-fixed paraffin-embedded specimens. The 4-μmthick paraffin sections were deparaffinized in xylene and dehydrated in gradient alcohol. Antigen retrieval was carried out using 6.5 mM citrate buffer (pH 6.0) pressure-cooking and endogenous peroxidase activity was blocked with 2.5 % hydrogen peroxide in methanol for 30 min at room temperature. Then the sections were incubated with 10 % goat serum for 30 min and followed by the primary monoclonal antibodies against hMLH1 (Cell Signaling Technology, Boston, MA; 1:200 dilution) and hMSH2 (Cell Signaling; 1:250 dilution),

Tumor Biol.

respectively, at 4 °C overnight. The slides were subsequently incubated with the corresponding secondary antibodies (Sigma-Aldrich, St Louis, MO). The negative control slides were probed with PBS instead of the primary antibodies under the same experimental conditions. The IHC staining was scored according to the percentage of positive-staining cells and staining intensity as previously described [19]. The positive cell percentage of cells was scored as follows: 0=less than 10 %, 1=10–25 %, 2=26–50 %, 3=51–75 %, and 4=more than 75 % positive-staining cells. The staining intensity was classified as: 1 = weak, 2 = moderate, and 3 = strong. Immunoreactive score (IS) of each specimen was determined by multiplication of these two scores listed above. According to IS, hMLH1 or hMSH2 was defined to be negative (IS<2) or positive (IS≥2). The tumors negative for at least one of the MMR genes (hMLH1 or hMSH2) were defined as dMMR, and those which were positive for both hMLH1 and hMSH2 were identified as pMMR tumors.

screening medium (500 μg/ml for SW480; 400 μg/ml for HCT-116). Colonies formed were individually picked for further expansion. Expression of hMLH1 in the transfected SW480 and HCT-116 cell lines were confirmed by reverse transcription (RT-PCR) and Western blotting assay.

Cell lines Two colon cancer cell lines of HCT-116 (hMLH1−, dMMR) and SW480 (hMLH1+, pMMR), which were obtained from the Institutes of Biomedical Sciences (IBS), Fudan University, were cultured in RPMI 1640 (GIBCO, Grand Island, NY) supplemented with 10 % fetal bovine serum, 100 units/ml penicillin sodium, and 100 μg/mL streptomycin sulfate (GIBCO) at 37 °C under an atmosphere of 95 % air and 5 % CO2.

RNA isolation and qPCR Total RNA was isolated using TRIzol Reagent (Invitrogen) following the manufacturer’s instructions. First-strand cDNA synthesis and qRT-PCR assay based on SYBR Green fluorescent (TaKaRa Bio Inc., Otsu, Japan) were performed as previously described [20]. The specific primers for hMLH1 were as follows: forward primer: 5′TTCAGGCCTCCCTCTT-TAACAATC-3′, reverse primer: 5′-CGTTCGTGG CAGGGGTTATT-3′). Primers for TS (GenBank NM_001071) were forward primer: 5′- GGGCGT AGCTGGCGATGT- 3′, reverse primer: 5′- GAATCA TCAT GTGCGCTTGGA -3′. β-Actin was used as an internal control, with forward primer: 5′-TGACGTGGACATCCGCAA AG-3′, reverse primer: 5′-CTGGAAGGTGGACAGCGA GG-3′. Amplifications were carried out with an ABI 7500 thermal cycler (Applied Biosystems, Carlsbad, CA) in triplicate according to the following cycling parameters: 50 cycles at 95 °C for 5 s, 60 °C for 20 s. The relative mRNA expression levels were calculated based on the threshold cycle (Ct) values. The relative expression of targeting gene was calculated by 2−ΔCt [ΔCt=Ct (targeting gene)-Ct (β-actin)] method and normalized to the internal control of β-actin mRNA.

Plasmid construction and transfection Three candidate shRNA sequences targeting human MLH1 gene (GenBank Accession No.: NM_000249) in SW480 were designed as follows (GeneChem, Shanghai, China), sequence 1: sense 5′- CTGATCTATACAAAGTCTT -3′, antisense 5′- AAGA CTTTGTATAGATCAG -3′; sequence 2: sense 5′-GTTCTT CTTTCTCTGTATT-3′, antisense 5′-AATACAGAGAAAGA AGAAC-3′; sequence 3: sense 5′-AGACTATTTCTCTTTG GAA-3′, antisense 5′-TTCCAAAGAGAAATAGTCT-3′, respectively. The control of non-silencing shRNA: sense 5′TTCTCCGAACGTGTCACGT-3, antisense 5′-ACGTGA CACGTTCGGAGAA-3′. HCT-116 cells were stably transfected with pEGFP-N1-hMLH1 or pEGFP-N1 vector as the blank control (GeneChem, Shanghai, China). Sequences of pEGFP-N1-hMLH1: sense: 5′-TCCGCTCGAGATGTCG TTCGTGGCAGGGGTTATTC-3′; antisense 5′-ATGGGGTA CCGTACACCTCTCAAAGACTTTGTATAG-3′. SW480 and HCT-116 cells were transfected using Lipofectamine 2000 (Invitrogen, Carlsbad, CA). Briefly, the cells (covered 80 % of the petri dish) were washed twice with PBS and once with serum-free medium, then transfected with the above plasmids (diluted by serum-free medium) respectively. Six hours later, the medium was removed and replaced by complete medium supplemented with 10 % FBS. Another 24 h later, cells were trypsinized and exposed to G418

Chemosensitivity detection The cells were seeded in 96-well plates (Costar, Cambridge, MA) at 3.0×103 per well and incubated at 37 °C. 24 h later, the freshly prepared irinotecan was, respectively, added to the culture medium at different final concentrations of 0, 1, 2, 4, 8, 16, and 32 μM for 48 h. Viable cells were determined by methyl thiazolyl tetrazolium assay (MTT). Brief procedures were as follows: 20 μl MTT was added to each well, followed by incubation at 37 °C for 4 h. Then culture medium was removed and replaced by 150 μl DMSO per well. The absorbances at 490 nm were measured. Dose-dependent curves of irinotecan were plotted and 50 % growth inhibition (IC50) values were determined.

Western blotting The cells were plated at 5×104 per well in six-well plates and exposed to irinotecan on the following day. The concentrations of irinotecan were 0, 1, 4 μM in SW480 cells, and were 0, 1, 8 μM in HCT-116 cells (the working concentrations of irinotecan were below and above respective IC50 value in each group). Total protein was extracted and separated by SDS-PAGE under reducing conditions and then transferred onto PVDF membrane (Millipore, Bedford, MA). The membrane was blocked in PBS containing 0.1 % Tween20 (PBS-T) with 5 % skim milk at room temperature for 2 h, subsequently incubated with the primary antibody against TS (Cell Signaling Technology, Boston, MA; 1:1000 dilution) at

Tumor Biol.

4 °C overnight. HRP-conjugated IgG (KPL, dilution 1:3000, Gaithersburg, MD) was used as the secondary antibody. βTubulin protein was determined by the specific antibody (Sigma-Aldrich; 1:2,000 dilution,) as an internal control. The bands were visualized by employing the enhanced chemiluminescence (ECL) method.

SPSS, version 19.0 (SPSS, Inc, Chicago, IL). The inhibition rates were analyzed by t test. All statistical tests in the study were two-sided and statistical significance was set at p<0.05.

Statistical analysis GraphPad Prism 5.0 Software (GraphPad Software, Inc., La Jolla, CA) was applied for statistical analysis. The correlation between MMR status and clinicopathologic features were determined by χ2 tests. The patients’ survival curves were calculated with Kaplan–Meier method and log-rank test. The Cox proportional hazards regression model was established to analyze the factors independently associated with PFS and OS. The inhibition rates of irinotecan determined by MTT method and IC50 values were assessed using

dMMR was associated with proximal and poorly differentiated metastatic colon cancer Positive hMLH1 and hMSH2 protein expression were located in nuclei (Fig. 1a). dMMR was found in 34 out of 184 (18.5 %) patients and related with proximal colon cancer and poorly differentiated tumors. The rate of dMMR-positive staining was 26.3 % in proximal metastatic colon cancer versus 13.0 % in the distal side (p=0.033) and 26.8 % in the poorly and undifferentiated tissues versus 13.3 % in the moderate-high differentiated tissues (p=0.031).

Fig. 1 hMLH1 and hMSH2 expression and the survival curves. a Immunohistochemical staining of hMLH1, hMSH2 (located in nuclei) and TS (mainly located in cytoplasm). (a1), (a2) Positive staining for hMLH1; (a3), (a4) negative staining for hMLH1; (b1), (b2) positive staining for hMSH2; (b3), (b4) negative staining for hMSH2; (c1), (c2)

positive expression of TS; (c3), (c4) negative staining for TS. Image a1, a3, b1, b3, c1, c3: Original magnification ×200; a2, a4, b2, b4, c2, c4: Original magnification ×400. b Patients with dMMR status have better prognoses than those with pMMR status

Results

Tumor Biol.

No correlation was found between MMR expression and other clinicopathological features, including gender, age, ECOG performance, chemotherapy regimens, and TS expression (p>0.05, Table 1). dMMR was correlated with favorable prognosis DCR in patients with dMMR and pMMR status were 84.8 and 66.2 % respectively (p=0.037). Subgroup analysis indicated that DCR of combined chemotherapy group was 86.7 versus 66.9 % in single-agent irinotecan chemotherapy group, with statistical significance (p=0.044). Mean PFS was 8.6 months (95 % CI: 7.1–10.0 months) in dMMR group and 6.6 months (95 % CI, 6.0–7.3 months) in pMMR group, and the difference was statistically significant (p=0.018). Mean OS was 18.7 months (95 % CI, 17.0–20.4 months) in dMMR tumors, compared with 17.5 months (95 % CI, 16.7–18.3 months) in patients with pMMR status, but the difference was not statistically significant (p=0.246; Fig. 1b). The multivariable Cox proportional hazards model analysis indicated that MMR status was an independent predictive factor for PFS (p=0.010), not for OS (p=0.103). Tumor location was associated with OS (p=0.028; Table 2). Cells with dMMR status were more sensitive to irinotecan MMR status of the two colon cells HCT-116 and SW480 were checked before the whole cytological experiment. hMLH1 was expressed normally in SW480 cells while absent in HCT-116 cells (Supplementary Fig. Fig. S2). Sequence 3 was selected to be the targeted hMLH1 shRNA in SW480 cells, with an inhibition ratio of 51.7 % (Fig. 2a). Through Table 1 Relationship between MMR expression and clinicopathological features of colon cancer

Clinicopathological features

Gender Age (years) Location Differentiation TS expression ECOG performance Chemotherapy regimens

depleting hMLH1 expression by shRNA in SW480 cells and overexpress hMLH1 in HCT-116 cells in vitro, we established two pairs of colon cell lines: SW480-shRNA-hMLH1 versus SW480-shRNA-Control; HCT-116-hMLH1 Vector versus HCT-116-hMLH1+. hMLH1 status in the above transfected SW480 cells and HCT-116 cells were confirmed by RT-PCR and Western blotting analysis. Location of hMLH1 (167 bp) was between 200 and 150 bp (Fig. 2b). Western blotting results showed that hMLH1 protein was expressed in pMMR cell lines (SW480-shRNA-Control and HCT-116-hMLH1+) while absent in dMMR cell lines (SW480-shRNA-hMLH1 and HCT-116-hMLH1Vector; Fig. 2c). The two pairs of colon cell lines with different MMR status were established successfully. We investigated the association between MMR status and sensitivity to irinotecan in the above colon cancer cells. As shown in Fig. 3, the cells with dMMR status were more sensitive to irinotecan compared with pMMR cells (Fig. 3). After being treated of irinotecan for 48 h, IC50 values were 2.46± 0.09 μM against HCT-116-hMLH1Vector (dMMR) cell, and 3.51±0.54 μM against HCT-116-hMLH1+ (pMMR) cell. IC50 values in SW480-shRNA-hMLH1 (dMMR) cell and SW480-shRNA-Control (pMMR) cell were 3.46±0.11 and 6.84±1.31 μM respectively. Irinotecan reduced the TS expression level, especially in dMMR colon cancer cells Every cell line was treated with irinotecan at its own IC10/IC25/IC50 concentration (Supplementary Table S1) for 24 h, followed by respective IC10/IC25/IC50 concentration of 5-FU (Supplementary

Variable

Male Female ≤53 >53 Proximal Distal Well-mod Poor-undifferentiated Positive Negative 0–1 ≥2 Irinotecan Combined schemeb

p value

MMR statusa n

Proficient MMR

Deficient MMR

111 73 115 69 76 108 113 71 96 88 129 55 13

89 (80.2) 61 (83.6) 96 (83.5) 54 (78.3) 56 (73.7) 94 (87.0) 98 (86.7) 52 (73.2) 80 (83.3) 70 (79.5) 106 (82.2) 44 (80.0) 10 (76.9)

22 (19.8) 12 (16.4) 19 (16.5) 15 (21.7) 20 (26.3) 14 (13.0) 15 (13.3) 19 (26.8) 16 (16.7) 18 (20.5) 23 (17.8) 11 (20.0) 3 (23.1)

171

140 (81.9)

31 (18.1)

0. 698 0.434 0.033 0.031 0.571 0.836 0.710

Both of hMLH1 and hMSH2 positive expression were considered to be pMMR, either of them negative was considered to be dMMR

a

b

Combined scheme: Irinotecan combined fluorouracil

Tumor Biol. Table 2 Multivariable Cox proportional hazards models analysis for predictive factors of PFS and OS

Clinicopathological variables n

Gender Male Female Age ≤53 >53 Location Proximal Distal Differentiation Well-mod Poor-undifferentiated TS Positive Negative MMR Positive Negative ECOG 0–1 ≥2 Chemotherapy regimens Irinotecan Combined scheme

Table S2) for another 24 h, then cell viability was determined. The cell survival rates were lower in the dMMR cell lines than the pMMR cell lines (Fig. 4; p<0.05, t test). We further explored the mechanism of combined effect of irinotecan and 5-FU in colon cancer cell lines with different MMR status. After being exposed to different concentrations of irinotecan for 24 h, TS expression level was reduced in a dosage-dependent manner in all cell lines and this downregulation was more significant in cells with dMMR status. TS regulation was determined by qPCR and western blot analysis (p≤0.001; Fig. 5).

Discussion MMR system is one of the important bases mismatch repair forms and closely associated with colorectal cancer. However, sensitivity of irinotecan in the tumors with different MMR status is not so clearly defined. Our data for the first time show that dMMR is a possible predictive marker of irinotecan treatment for colon cancer at both the preclinical and clinical levels.

PFS multivariable analysis HR(95 % CI) P value

OS multivariable analysis HR(95 % CI) P value

111 73

1 0.884 (0.641–1.218) 0.450

1 0.921 (0.653–1.298) 0.637

115 69

1 1.219 (0.866–1.716) 0.256

1 1.214 (0.847–1.741) 0.291

108 76

1 0.989 (0.713–1.371) 0.946

1 0.683 (0.487–0.959) 0.028

113 71

1 1.392 (0.977–1.983) 0.076

1 1.095 (0.757–1.583) 0.629

96 88

1 1.166 (0.852–1.596) 0.337

1 1.026 (0.743–1.417) 0.877

150 34

1 0.570 (0.372–0.875) 0.010

1 0.693 (0.445–1.077) 0.103

129 55

1 1.140 (0.795–1.635) 0.476

1 1.157 (0.792–1.688) 0.451

13 171

1 0.596 (0.324–1.097) 0.096

1 0.562 (0.287–1.101) 0.093

This study suggests involvement of MMR in the curative effect and prognosis of metastatic colon cancer. The incidence of dMMR based on IHC assay was 18.5 %, close to the reported 20 %. If IHC is combined with PCR assay to screen for tumors with MSI/dMMR status, the effectiveness would be higher [21]. dMMR is associated with proximal, poor differentiated tumors and favorable efficacy of irinotecan chemotherapy, with a longer PFS and a trend of longer OS, which is consistent with other researches [15, 22–24]. However, some small-scale investigations denied the predictive value of MMR to irinotecan [25, 26]. Kim JE and co-workers [27] suggested that dMMR colon cancer patients had a trend of advantage over those with pMMR status in the aspects of ORR (overall response rate), PFS, and OS, but none reached statistically significant difference. These inconsistent results of studies may be related to the diversity of the pathogenesis involved in colon cancer, the MMR gene polymorphism, and complicated metabolic pathway of irinotecan [28]. What’s more, most of the studies are small-scale, retrospective, or nonrandomized, with a significant bias. Besides, MMR gene can affect the prognosis of colon cancer through interacting with other potential biomarkers, such as CIMP, BRAF, and KRAS [29–32].

Tumor Biol.

Fig. 2 Two pairs of colon cancer cell lines were established by regulating MMR status. a Downregulation of hMLH1 expression by constructed shRNA-hMLH1 in SW480 cells. qRT-PCR and Western blot demonstrate that shRNA sequence 3 has the highest inhibition efficiency of 51.7 % in the three candidate sequences. The shRNA sequence 3 is named as shRNA-hMLH1 and used for subsequent experiments. b Location of hMLH1 expression is identified by reverse

transcription PCR. The result shows that hMLH1 expression (167 bp) is between 200 and 150 bp. c Western blot analysis is used to examine the expressions of hMLH1 in these transfected colon cancer cell lines. hMLH1 protein was expressed in pMMR cell lines (SW480-shRNAcontrol and HCT-116-hMLH1+), while absent in dMMR cell lines (SW480-shRNA-hMLH1 and HCT-116-hMLH1Vector). β-Tubulin is used as an internal control

The result that colon cancers with dMMR status are chemosensitive to irinotecan was reinforced in vitro.

Contrary to two studies [16, 33], we and other authors [16, 34, 35] demonstrated that dMMR colon cancer cells display

Fig. 3 MTT assay demonstrated significantly decrease cell viability of colon cancer cell lines with different status. a HCT-116-hMLH1Vector compared to HCT-116-hMLH1+. b SW480-shRNA-hMLH1 compared to SW480-shRNA-Control. ★★p<0.01

Tumor Biol.

Fig. 4 Survival rates after irinotecan and 5-FU treatment. After been exposed to irinotecan and 5-FU at their own IC10, IC25, and IC50 concentrations, respectively, dMMR cells showed lowed survival rates.

a HCT-116-hMLH1Vector versus HCT-116-hMLH1+ and b SW480shRNA-hMLH1 versus SW480-shRNA-Control. ★p<0.05; ★★p<0.01

Fig. 5 TS expression level after been exposed to irinotecan was examined by qRT-PCR and Western blot. TS expression reduced in a dosage-dependent manner and the down-regulation was more significant in cells with dMMR status. a–b TS mRNA and protein

expression in HCT-116-hMLH1Vector and HCT-116-hMLH1+ cell lines; c–d TS mRNA and protein expression in SW480-shRNA-hMLH1 and SW480-shRNA-Control cell lines. ★★p<0.01;★★★p<0.001

Tumor Biol.

hypersensitivity to irinotecan. In line with our findings, Vilar [16] showed that three cell lines displaying MSI-H due to an hMLH1 inactivation were four- to nine- fold more sensitive to irinotecan than an MSS/pMMR cell line and that this difference in sensitivity is apparently independent from the original cause of the hMLH1 deficiency (methylation or mutation). Different from others, our study investigated the drugsensitivity in the same cell line by regulating MMR status, so influence from other molecules among different cell lines might can be eliminated, and the result may be more persuasive. Irinotecan is selectively sensitive to colon cancer cells with dMMR status; this could be related to the following reasons: irinotecan and active metabolites SN-38 make transient DNA single-strand breaks convert into permanent DNA doublestrand breaks, which cannot be repaired totally and timely because of deficiency in mismatch repair system. Besides, deficient MMR system can affect homologous recombination, an important repair method [16]. Another study [36] indicated that dMMR cells are able to repair the double-strand breaks by homologous recombination more efficiently but are more error-prone than pMMR cells. dMMR genes frequently tend to accumulate mutations within microsatellite repeats of genes implicated in DSB repair pathway and cell lines harboring mutations in MRE11 and RAD50 were more sensitive to irinotecan [16, 37]. There is another explanation for hypersensitivity to irinotecan in dMMR cell involving in p53 gene. A study [35] indicated that the immediate reaction to irinotecan was a p53- and hMLH1-independent G2/M arrest, which was maintained for a much longer time in the hMLH1+ than in the hMLH1− cell line. Irinotecan-induced G2/M arrest can prevent apoptosis, so lack of hMLH1 protein enhance irinotecan-induced apoptosis. Irinotecan has demonstrated activity against 5-FU refractory colorectal cancer patients, and the inhibition rate of combination therapy of irinotecan and 5-FU is significantly higher than that of monotherapy. The combination has been explored in a number of different schedules [12, 38–42], in vivo and in vitro data showed additive or synergistic effect in sequential delivery with an administration of irinotecan followed by 5FU [12, 39, 41, 42], although some reports hold the reverse sequence [38, 40]. According to the results of most studies and clinical drug use, we did the research using the irinotecan-5FU combination. In our study, after been exposed to irinotecan and 5-FU sequentially at their own IC10, IC25, and IC50 concentrations, respectively, the inhibition rates in dMMR cell lines were higher than those with pMMR status. This more significant combination effect in dMMR colon cancer maybe contribute to the better DCR and PFS in the combined chemotherapy group of patients with dMMR status. Different mechanisms for the combine effect of these two drugs have been studied [12, 43], including cell cycles change, induction of apoptosis, and regulation of the activity of one

agent at specific target sites. It has been reported that irinotecan can improve fluorouracil cytotoxicity by downregulating the expression of TS [12, 13, 43, 44]. In view of the different inhibition rates in cell lines with different MMR status, we regard to demonstrate the relationship between TS regulation and MMR status. We have determined for the first time that TS expression was reduced by irinotecan in all cell lines involved and this down-regulation of TS is related to MMR status, with more reduction of TS protein and mRNA expression level in dMMR cells (p<0.05). So the significant combined effect of irinotecan and 5-FU in dMMR cell lines may be related to more reduction of TS. However, in the clinical level, it is difficult to determine the TS expression level within tumors and it is uncertain whether the expression level in the serum could be an instead. Investigations suggested a certain relationship between TS and MMR, but the results were uncertain and lacked of cytological studies [45–47]. The mechanism of regulation of TS according to MMR status needs a further study, which is our further research of interest.

Conclusions In conclusion, the results of cytological study support the clinical level research. Our study favors an increased sensitivity of irinotecan in colon cancer with dMMR status and dMMR patients may benefit more from irinotecan-based chemotherapy with a longer PFS and a trend of longer OS. Irinotecan combining with 5-FU may be a good schedule for the metastatic colon cancer with dMMR status. Acknowledgments This study was supported by the grants from the National Nature Science Foundation of China (nos. 30770971, 81172470, 81070362, and 81372629). Compliance with ethical standards The manuscript has been performed with the approval of the ethics of Xiangya Hospital, Central South University. Informed consent was obtained from all the patients before enrollment in the study. This study was prospectively performed and approved by the institutional Ethics Committees of our hospital and conducted in accordance with the ethical guidelines of the Declaration of Helsinki. Conflict of interest The authors declare that they have no competing interests.

References 1. 2.

Siegel R, Desantis C, Jemal A. Colorectal cancer statistics, 2014. CA Cancer J Clin. 2014;64:104–17. Gibson J, Lacy J, Matloff E, Robert M. Microsatellite instability testing in colorectal carcinoma: a practical guide. Clin Gastroenterol Hepatol. 2014;12:171–6.

Tumor Biol. 3.

4. 5.

6.

7.

8.

9. 10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

Buecher B, Cacheux W, Rouleau E, Dieumegard B, Mitry E, et al. Role of microsatellite instability in the management of colorectal cancers. Dig Liver Dis. 2013;45:441–9. Brenner H, Kloor M, Pox CP. Colorectal cancer. Lancet. 2014;383: 1490–502. Pino MS, Chung DC. Microsatellite instability in the management of colorectal cancer. Expert Rev Gastroenterol Hepatol. 2011;5: 385–99. Lin CC, Lai YL, Lin TC, Chen WS, Jiang JK, et al. Clinicopathologic features and prognostic analysis of MSI-high colon cancer. Int J Color Dis. 2012;27:277–86. Meyers M, Wagner MW, Hwang HS, Kinsella TJ, Boothman DA. Role of the hMLH1 DNA mismatch repair protein in fluoropyrimidine-mediated cell death and cell cycle responses. Cancer Res. 2001;61:5193–201. Duldulao MP, Lee W, Le M, Chen Z, Li W, et al. Gene expression variations in microsatellite stable and unstable colon cancer cells. J Surg Res. 2012;174:1–6. Devaud N, Gallinger S. Chemotherapy of MMR-deficient colorectal cancer. Familial Cancer. 2013;12:301–6. Sinicrope FA, Foster NR, Thibodeau SN, Marsoni S, Monges G, et al. DNA mismatch repair status and colon cancer recurrence and survival in clinical trials of 5-fluorouracil-based adjuvant therapy. J Natl Cancer Inst. 2011;103:863–75. Juo YY, Johnston FM, Zhang DY, Juo HH, Wang H, et al. Prognostic value of CpG island methylator phenotype among colorectal cancer patients: a systematic review and meta-analysis. Ann Oncol. 2014;25:2314–27. Sato K, Kitajima Y, Kohya N, Koga Y, Ohtaka K, et al. CPT-11 (SN-38) chemotherapy may be selectively applicable to biliary tract cancer with low hMLH1 expression. Anticancer Res. 2007;27:865– 72. Bendardaf R, Lamlum H, Ristamaki R, Korkeila E, Syrjanen K, et al. Mismatch repair status is a predictive factor of tumour response to 5-fluorouracil and irinotecan chemotherapy in patients with advanced colorectal cancer. Tumour Biol. 2007;28:212–20. Vilar E, Scaltriti M, Balmana J, Saura C, Guzman M, et al. Microsatellite instability due to hMLH1 deficiency is associated with increased cytotoxicity to irinotecan in human colorectal cancer cell lines. Br J Cancer. 2008;99:1607–12. Bertagnolli MM, Niedzwiecki D, Compton CC, Hahn HP, Hall M, et al. Microsatellite instability predicts improved response to adjuvant therapy with irinotecan, fluorouracil, and leucovorin in stage III colon cancer: Cancer and Leukemia Group B Protocol 89803. J Clin Oncol. 2009;27:1814–21. Bhonde MR, Hanski ML, Stehr J, Jebautzke B, Peiro-Jordan R, et al. Mismatch repair system decreases cell survival by stabilizing the tetraploid G1 arrest in response to SN-38. Int J Cancer. 2010;126:2813–25. Torigoe S, Ogata Y, Matono K, Shirouzu K. Molecular mechanisms of sequence-dependent antitumor effects of SN-38 and 5fluorouracil combination therapy against colon cancer cells. Anticancer Res. 2009;29:2083–9. Fukushima M, Sakamoto K, Ohshimo H, Nakagawa F, Taguchi T. Irinotecan overcomes the resistance to 5-fluorouracil in human colon cancer xenografts by down-regulation of intratumoral thymidylate synthase. Oncol Rep. 2010;24:835–42. Zeng S, Tao Y, Huang J, Zhang S, Shen L, et al. WD40 repeatcontaining 62 overexpression as a novel indicator of poor prognosis for human gastric cancer. Eur J Cancer. 2013;49:3752–62. Shen LF, Chen J, Zeng S, Zhou RR, Zhu H, et al. The superparamagnetic nanoparticles carrying the E1A gene enhance the radiosensitivity of human cervical carcinoma in nude mice. Mol Cancer Ther. 2010;9:2123–30. Yoon YS, Yu CS, Kim TW, Kim JH, Jang SJ, et al. Mismatch repair status in sporadic colorectal cancer: immunohistochemistry and

22. 23.

24.

25.

26.

27.

28. 29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

microsatellite instability analyses. J Gastroenterol Hepatol. 2011;26:1733–9. Boland CR, Goel A. Microsatellite instability in colorectal cancer. Gastroenterology. 2010;138:2073–87. Fallik D, Borrini F, Boige V, Viguier J, Jacob S, et al. Microsatellite instability is a predictive factor of the tumor response to irinotecan in patients with advanced colorectal cancer. Cancer Res. 2003;63: 5738–44. Kim JC, Roh SA, Cho DH, Kim TW, Yoon SN, et al. Chemoresponsiveness associated with canonical molecular changes in colorectal adenocarcinomas. Anticancer Res. 2009;29:3115– 23. Braun MS, Richman SD, Quirke P, Daly C, Adlard JW, et al. Predictive biomarkers of chemotherapy efficacy in colorectal cancer: results from the UK MRC FOCUS trial. J Clin Oncol. 2008;26: 2690–8. Koopman M, Kortman GA, Mekenkamp L, Ligtenberg MJ, Hoogerbrugge N, et al. Deficient mismatch repair system in patients with sporadic advanced colorectal cancer. Br J Cancer. 2009;100:266–73. Kim JE, Hong YS, Ryu MH, Lee JL, Chang HM, et al. Association between deficient mismatch repair system and efficacy to irinotecan-containing chemotherapy in metastatic colon cancer. Cancer Sci. 2011;102:1706–11. Kim TW, Innocenti F. Insights, challenges, and future directions in irinogenetics. Ther Drug Monit. 2007;29:265–70. Kim JH, Shin SH, Kwon HJ, Cho NY, Kang GH. Prognostic implications of CpG island hypermethylator phenotype in colorectal cancers. Virchows Arch. 2009;455:485–94. Lochhead P, Kuchiba A, Imamura Y, Liao X, Yamauchi M, et al. Microsatellite instability and BRAF mutation testing in colorectal cancer prognostication. J Natl Cancer Inst. 2013;105:1151–6. Samadder NJ, Vierkant RA, Tillmans LS, Wang AH, Weisenberger DJ, et al. Associations between colorectal cancer molecular markers and pathways with clinicopathologic features in older women. Gastroenterology. 2013;145:348–56. Kim JH, Kang GH. Molecular and prognostic heterogeneity of microsatellite-unstable colorectal cancer. World J Gastroenterol. 2014;20:4230–43. Fedier A, Schwarz VA, Walt H, Carpini RD, Haller U, et al. Resistance to topoisomerase poisons due to loss of DNA mismatch repair. Int J Cancer. 2001;93:571–6. Jacob S, Aguado M, Fallik D, Praz F. The role of the DNA mismatch repair system in the cytotoxicity of the topoisomerase inhibitors camptothecin and etoposide to human colorectal cancer cells. Cancer Res. 2001;61:6555–62. Magrini R, Bhonde MR, Hanski ML, Notter M, Scherubl H, et al. Cellular effects of CPT-11 on colon carcinoma cells: dependence on p53 and hMLH1 status. Int J Cancer. 2002;101:23–31. Elliott B, Jasin M. Repair of double-strand breaks by homologous recombination in mismatch repair-defective mammalian cells. Mol Cell Biol. 2001;21:2671–82. Tahara M, Inoue T, Sato F, Miyakura Y, Horie H, et al. The use of Olaparib (AZD2281) potentiates SN-38 cytotoxicity in colon cancer cells by indirect inhibition of Rad51-mediated repair of DNA double-strand breaks. Mol Cancer Ther. 2014;13:1170–80. Inoue Y, Tanaka K, Hiro J, Yoshiyama S, Toiyama Y, et al. in vitro synergistic antitumor activity of a combination of 5-fluorouracil and irinotecan in human colon cancer. Int J Oncol. 2006;28:479–86. Guichard S, Cussac D, Hennebelle I, Bugat R, Canal P. Sequencedependent activity of the irinotecan-5FU combination in human colon-cancer model HT-29 in vitro and in vivo. Int J Cancer. 1997;73:729–34. Inoue Y, Miki C, Watanabe H, Hiro J, Toiyama Y, et al. Scheduledependent cytotoxicity of 5-fluorouracil and irinotecan in a colon cancer cell line. J Gastroenterol. 2006;41:1149–57.

Tumor Biol. 41.

Cao S, Rustum YM. Synergistic antitumor activity of irinotecan in combination with 5-fluorouracil in rats bearing advanced colorectal cancer: role of drug sequence and dose. Cancer Res. 2000;60:3717– 21. 42. Azrak RG, Cao S, Slocum HK, Toth K, Durrani FA, et al. Therapeutic synergy between irinotecan and 5-fluorouracil against human tumor xenografts. Clin Cancer Res. 2004;10: 1121–9. 43. Ide H, Kikuchi E, Hasegawa M, Hattori S, Yasumizu Y, et al. Therapeutic enhancement of S-1 with CPT-11 through downregulation of thymidylate synthase in bladder cancer. Cancer Med. 2013;2:488–95. 44. Takiuchi H, Kawabe S, Gotoh M, Katsu K. Thymidylate synthase gene expression in primary tumors predicts activity of s-1-based

chemotherapy for advanced gastric cancer. Gastrointest Cancer Res. 2007;1:171–6. 45. Jensen SA, Vainer B, Kruhoffer M, Sorensen JB. Microsatellite instability in colorectal cancer and association with thymidylate synthase and dihydropyrimidine dehydrogenase expression. BMC Cancer. 2009;9:25. 46. Kristensen MH, Weidinger M, Bzorek M, Pedersen PL, Mejer J. Correlation between thymidylate synthase gene variants, RNA and protein levels in primary colorectal adenocarcinomas. J Int Med Res. 2010;38:484–97. 47. Ohrling K, Karlberg M, Edler D, Hallstrom M, Ragnhammar P. A combined analysis of mismatch repair status and thymidylate synthase expression in stage II and III colon cancer. Clin Colorectal Cancer. 2013;12:128–35.