Peripheral Opioid Antagonist Enhances the Effect of Anti-Tumor Drug by Blocking a Cell Growth-Suppressive Pathway In Vivo

The dormancy of tumor cells is a major problem in chemotherapy, since it limits the therapeutic efficacy of anti-tumor drugs that only target dividing cells. One potential way to overcome chemo-resistance is to “wake up” these dormant cells. Here we show that the opioid antagonist methylnaltrexone (MNTX) enhances the effect of docetaxel (Doc) by blocking a cell growth-suppressive pathway. We found that PENK, which encodes opioid growth factor (OGF) and suppresses cell growth, is predominantly expressed in diffuse-type gastric cancers (GCs). The blockade of OGF signaling by MNTX releases cells from their arrest and boosts the effect of Doc. In comparison with the use of Doc alone, the combined use of Doc and MNTX significantly prolongs survival, alleviates abdominal pain, and diminishes Doc-resistant spheroids on the peritoneal membrane in model mice. These results suggest that blockade of the pathways that suppress cell growth may enhance the effects of anti-tumor drugs.


Introduction
Chemoresistance is often observed in tumor therapy, and can lead to a poor prognosis. One potential mechanism of such resistance is the arrest of tumor cell division (i.e., a dormant state), Ethics in Animal Experimentation of the National Cancer Center. Efforts were made to minimize the numbers and any suffering of animals used in the subsequent experiments.

Immunochemistry
Specimens fixed in formalin and embedded in paraffin were cut into 5 μm sections, subsequently dewaxed, and dehydrated. Sections were blocked for DAKO protein block (DAKO, Carpinteria, CA), and stained with a primary antibody against Ki-67 antigen (1:75; DAKO) at room temperature for 1 h. Subsequently, the sections were subjected to DAB (substrate buffer + DAB chromogen [× 50]) for 5 min. The slides were counterstained with hematoxylin and then mounted.

Microarray analysis
Total RNA was isolated by suspending the cells in an ISOGEN lysis buffer (Nippon Gene, Toyama, Japan) followed by precipitation with isopropanol. We used Human Expression Array U95A version 2 (Affymetrix, Santa Clara, CA) for analysis of mRNA expression levels corresponding to 12,600 transcripts. The procedures were conducted according to the suppliers' protocols. The expression value (average difference: AD) of each gene was calculated using Gen-eChip Analysis Suite version 4.0 software (Affymetrix). The mean of AD values in each experiment was 1000 to reliably compare variable multiple arrays. Hierarchical clustering is widely used as one of the unsupervised learning methods. Hierarchical clustering of microarray data was performed by the use of GeneSpring (Agilent Technologies Ltd., Palo Alto, CA), Microsoft EXCEL, and Cluster & TreeView software [29]. All the microarray data have been deposited in a MIAME compliant database, GEO; the accession number GSE47007. By Wilcoxon u-test (p<0.05) and a 2-fold change, 188 genes were selected as specific genes for 18 intestinal-type GCs, and 704 genes were selected as specific genes for 12 diffuse-type GCs. The results of a twodimensional hierarchical clustering analysis of the 892 selected genes are shown.

Reverse transcription polymerase chain reaction (RT-PCR)
Total RNA was isolated from tissues or suspending cells using ISOGEN lysis buffer (Nippon Gene), followed by precipitation with isopropanol. As described in our previous report [29], semi-quantitative RT-PCR was carried out using primer sets (Table S1). For semi-quantitative RT-PCR, we showed data within linear range by performing 25-30 cycles of PCR.

Drug preparation
[Met 5 ]-enkephalin (OGF) was purchased from Wako (Tokyo, Japan). Methylnaltrexone (MNTX) was provided from Drs. H. Nagase & T. Suzuki. OGF and MNTX were dissolved in sterile water or saline. Docetaxel (Doc) was purchased from Aventis Pharma Co., Ltd. (Tokyo, Japan). Doc was prepared according to the manufacturer's instructions and diluted with sterile water or saline.

OGFR shRNA-bearing lentiviral particles transduction
Cells were cultured at a density of 1 x 10 2 cell per well in 24-well plates. After 24 h, cells were transferred to a medium containing Polybrene (Santa Cruz Biochemistry, Santa Cruz, CA) at a final concentration of 5 μg/ml, and we then added the OGFR shRNA lentivial particles (Santa Cruz Biochemistry) at MOI: 8 and incubated overnight. The medium with lentivial particles was removed, and the cells were cultured in a normal medium without polybrene. After 48 h, the cells were replaced with selection medium containing 2 μg/ml Puromycin, and then several colonies were picked up.

Cell growth assays
OGF-induced cell growth inhibition was determined by a cell proliferation assay using MTT assay (1 x 10 3 cells/well, 96-well plates). OGF (10 -4 M), combinations of OGF and MNTX (10 -6 M) or sterile water were added beginning 24 h after seeding. Both media and compounds were replaced daily. Seventy-two hours after treatment, 20 μl of 5 mg/ml MTT solution was added to each well of the culture medium. After incubation for an additional 4 h, the medium was removed and 100 μl of DMSO was added to resolve the formazan crystals. Optical density was measured using a microplate reader with an absorption wavelength of 563 nm. In each experiment, three replicates were prepared for each sample. The proportion of cell growth was determined based on the difference in absorbance between the samples and controls. MNTXinduced cell growth was conducted in 24-well plates (60As6: 2 x 10 4 cells/well and HSC-42: 1 x 10 4 cells/well) under normal nutrient (10% FBS) and low nutrient (2% FBS) conditions. Twenty-four hours after seeding, cells were treated with MNTX (10 -6 and 10 -5 M) or sterile vehicle for 72 h. Cells were harvested with a solution of 0.05% trypsin/0.53 mM EDTA, centrifuged, and counted with TC20 Automated Cell Counter (Bio-Rad, Hercules, CA). Doc-induced cell growth inhibition was conducted in 6-well plates (60As6: 2 x 10 5 cells/well and HSC-42: 5 x 10 4 cells/well). Twenty-four hours after seeding, cells were treated with Doc (10 -9 M) or sterile vehicle for 48 h, and subsequently treated with Doc, combinations of Doc and MNTX (10 -6 M), or a sterile vehicle for 48 h. Cells were harvested with a solution of 0.25% trypsin/0.53 mM EDTA, centrifuged, and counted with a hematcytometer. Cell viability was determined by trypan blue staining.
Co-culture and cell growth assays 1Cs-mM and NIH3T3 cells were plated in 24-well plates (2 x 10 4 cells/well). Twenty-four hours after plating, 60As6-GFP cells (1 x 10 4 cells/well) were seeded on 1Cs-mM and NIH3T3 cells respectively, and treated with MNTX (10 -5 M) or sterile water for 72 h. Cells were harvested with a solution of 0.25% trypsin/0.53 mM EDTA, centrifuged, and counted with a hematcytometer under a fluorescence microscope TS100 (Nikon, Tokyo, Japan).

Tumor cell inoculation
The density of 60As6-Luc or 60As6-GFP cells was adjusted to 1 × 10 6 cells per 1 ml phosphatebuffered saline (PBS). In the experimental group, the cell suspension was injected into the abdominal cavity via a 26 1/2-gauge needle inserted into the central abdomen. In the control group, PBS was injected into the abdominal cavity instead of tumor cells.

Tumor growth and survival studies
The day of tumor cell inoculation was considered day 0. Mice were weighed and measured weekly for tumor growth. For that measurement, whole-body luciferase imaging with an IVIS imaging system was used to visualize the 60As6-Luc cells, as described previously [22]. In vivo photon counting analysis was determined using Living Image software (version 2.50) (Xenogen). On day 7 after tumor inoculation, mice were divided evenly into 4 groups based on the number of photon counts. Mice were then treated intraperitoneally with saline, MNTX (0.3 mg/kg), Doc (0.5 mg/kg) and Doc/MNTX twice weekly until endpoint criteria were met. Endpoint criteria included severe cachexia, significant weight loss exceeding 20% of initial weight, or extreme weakness or inactivity.

Assessment of metastatic tumor cells
An in vivo imaging system OV110 (Olympus, Tokyo, Japan) was used to observe 60As6-GFP cells attached to the mesentery. Forty-nine days after the inoculation, mice were deeply anesthetized with isoflurane and the intestine and mesentery were quickly removed after decapitation. To analyze the number of single cells or spheroids on the mesentery and peritoneal surface, we captured 10 images per each intestine and mesentery.

Behavioral test
Hunching behavior was examined as described previously with some modifications [30]. Briefly, mice were placed individually in the center of an open field arena and observed for 180 sec. The hunching score was the total time (sec) that the mouse exhibited hunching behavior multiplied by the scoring factor, which was defined according to our recent paper [31]: 0 = normal coat luster, displays exploratory behavior. 1 = mild rounded-back posture, displays slightly reduced exploratory behavior, normal coat luster. 2 = severe rounded-back posture, displays considerably reduced exploratory behavior, piloerection, intermittent abdominal contractions. Behavioral testing was performed on day 35 after tumor inoculation.

Western blotting
Cells were lysed in RIPA buffer containing 1% protease inhibitor cocktail. The protein concentration of each sample was measured using a BCA protein assay kit (Thermo Fisher Scientific, Rockford, IL). Whole cell lysates were subjected to 15% SDS-PAGE using standard protocols. The following antibodies were used: p21 (BD Pharmingen, San Diego, CA), OGFR (Proteintech, Chicago, IL), β-actin (Cell Signaling Technology, Danvers, MA), and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) (Santa Cruz Biochemistry). Membranes were probed with secondary anti-rabbit or anti-mouse horseradish peroxidase-conjugated antibodies, and developed using a chemiluminescence Western blotting detection system SuperSignal (Thermo Fisher Scientific).

ELISA
OGF level of ascites and peritoneal washings obtained from patients or mice with peritoneal dissemination were assessed by an EIA kit (Peninsula Labs, San Carlos, CA) according to the manufacturer's instructions. All samples were studied in duplicate.

Flow cytometry
Cells were transferred to serum-free medium for 96 h, and subsequently harvested with 0.05% trypsin-EDTA, and then suspended in cold 80% ethanol. After fixation at -20°C overnight, the samples were washed in PBS (-) and incubated in 500 μl of PI/RNase Staining Buffer (BD Pharmingen) per 1 x 10 6 cells for 30 minutes at room temperature before analysis. Flow cyctometry was performed using FACSCalibur (BD, Franklin Lakes, NJ), and analyzed by a software, FlowJo (Tree Star, Inc., Ashland, OR).

siRNA transfection
OGFR siRNA was introduced to 60As6 cells using Dharma FECT1 (Dharmacon, La Fayette, CO), following the procedure recommended by the manufacturer. RT-PCR and Western blot analyses were carried out after the experiment.
Automated image capture and cell cycle analysis 60As6 cells were seeded at 1 x 10 3 cells per well on 96-well plates the day before treating them with OGF (10 -6 -10 -4 M). Forty-eight hours after treatment, 5-ethynyl uridine (EdU) was added for 6 h and stained with EdU antibody. Cell nuclei were labeled with Hoechst 33342. The DNA content and EdU intensity were quantitatively analyzed by the Cellomics ArrayScan high content microscope with the Cell Cycle BioApplication software (Thermo Fisher Scientific).

Statistics
Statistical analyses were carried out using GraphPad Prism version 5.0a (GraphPad Software, La Jolla, CA). The statistical significance of cell growth data was assessed with one-way ANOVA followed by the Bonferroni multiple comparisons test (cell growth) or two-tailed unpaired Student t test (for co-culture, micrometastasis and behavioral test). Mortality data was compared using Kaplan-Meier plots and Log-rank test. A p value of<0.05 was considered statistically significant.

Expression of PENK encoding OGF in diffuse-type GCs
Solid tumors with diffuse growth are composed of many myofibroblasts and few vessels (e.g., pancreatic cancers and scirrhous type of breast cancer) (S1 Fig). Depending on the conditions related to the microenvironment, such as nutrient deficiency, these tumors show a high prevalence of rarely-proliferative tumor cells. In support of these reports, diffuse-type GC shows a high proportion of Ki-67-negative non-proliferating tumor cells compared to intestinal-type GC (Fig 1A and 1B). We previously established a highly peritoneal metastatic cell line 60As6 by the 6-times transplantation of a diffuse-type GC-derived parental cell line (HSC-60) into a mouse peritoneal cavity [10]. We found that 60As6 cells exhibit apparent resistance to Doc by inducing growth arrest in G1 phase under serum starvation (S2A- S2C Fig), suggesting that the acquisition of this characteristic might be attributed to a particular tumor microenvironment, such as hyponutrition. To address the molecular mechanisms of dormancy related to chemoresistance of diffuse-type GC, we first searched for cell growth-suppressive signal pathways by a comparative gene expression analysis between 12 primary diffuse-type and 18 intestinal-type GCs (Fig 1C). In most of the intestinal-type GCs, CDC6, which is a typical marker of S-phase progression in the cell cycle, was highly expressed (S3A Fig). On the other hand, PENK encoding OGF was identified to be over-expressed in diffuse-type GCs (Fig 1C). PENK mRNA was also confirmed to be preferentially expressed in diffuse-type GCs compared with intestinaltype GCs by RT-PCR (Fig 1D). An increase in OGF secretion in cancer-associated ascites, but not peritoneal washings, was confirmed by ELISA (S3B Fig). Consistent with the expression profile of primary GCs, PENK and OGFR were expressed in the diffuse-type GC cell line HSC-60, 60As6 cells, and their mouse xenografts (60As6 xeno), whereas the intestinal-type GC cell line HSC-42 expressed only OGFR mRNA (Fig 2A). OGF has been reported to bind to two membrane receptors, μand δ-opioid receptors (OPRM1 and OPRD1, respectively), as well as to the nuclear receptor OGFR. Quite low expression of OPRM1 and OPRD1 was detected in both HSC-60 and 60As6 cells (S4A Fig). In various diffuse-type GC cell lines, little or no expression of these two membrane receptors was detected by RT-PCR (data not shown), whereas high expression levels of both PENK and OGFR were found in most of these cell lines (Fig S4B).

Blockade of OGF signaling by MNTX increased the cell growth of a diffuse-type GC cell line under low nutrient conditions
The growth of 60As6 cells treated with OGF, the opioid antagonist MNTX or both was measured by the MTT assay. Treatment with OGF for 72 h significantly inhibited cell growth (69.4 ± 9.8%), which was clearly restored by MNTX (93.4 ± 6.0%, Fig 2B). To investigate the mechanism of OGF-induced cell growth inhibition, a cell population of 60As6 cells after treatment with OGF was acquired from the ArrayScan HCS reader. Cells in different cell-cycle phases G1, S and G2/M were separated by a cell population analysis according to ErdU incorporation and DNA content. Treatment with OGF for 48 h resulted in a dose-dependent increase in the G1 population of 60As6 cells (S5A Fig). Accordingly, the number of cells at the G2/M phase was decreased. Furthermore, this effect was significantly diminished by MNTX ( S5B Fig). It has been reported that the expression of p16 and/or p21 is related to OGF-induced G1-arrest in cancers of the head and neck, ovaries, and pancreas [11][12][13]. In 60As6 cells, p21 protein was evidently increased at 3 h after treatment with OGF (S5C Fig), whereas p16 protein was not detected due to homozygous deletion of the gene (data not shown). With the use of two different RNA-interference experiments, we confirmed that OGFR is required for the suppression of cell growth by OGF. The growth of OGFR siRNA-transfected cells was not suppressed by OGF (S5D and S5E Fig), and a stable transfectant of OGFR shRNA, in which OGFR protein was reduced (Fig 2C), was also not affected by OGF (Fig 2D). We then examined the effect of the inhibition of OGF signaling by MNTX on cell growth under both normal and low nutrient conditions (10% or 2% FBS, respectively). In the low nutrient condition, treatment with MNTX for 72 h clearly increased the growth of 60As6 cells (10 -6 M: 126.4 ± 46.7%, 10 -5 M: 275.6 ± 10.1%, Fig 2E), but not HSC-42 cells (10 -6 M: 90.0 ± 33.6%, 10 -5 M: 92.2 ± 16.5%, Fig 2F). Cultures that were treated with MNTX under the normal nutrient condition showed no differences in the numbers of both cell lines compared to the control (Fig 2E and 2F).

Blockade of OGF released from mesothelial cells by MNTX increased the growth of diffuse-type GC cells in a co-culture system with mesothelial cells
The microenvironment for the peritoneal metastasis of diffuse-type GC is made up of mesothelial cells and myofibroblasts [14][15][16]. The first surface that free tumor cells encounter is the innermost layer of the peritoneum, the mesothelium. The mesothelium forms a cellular monolayer supported by a basement membrane. The adherence of tumor cells to the mesothelium is the second step in the metastatic cascade, which temporarily arrests the tumor cells to their eventual site of metastasis [17]. A recent study suggested that these cells are functionally organized to promote the survival of tumor cells in the host [18]. Interestingly, our originally established mouse mesothelial cell line 1Cs-mM (Fig 3A) also expresses Penk mRNA. Thus, we next investigated whether OGF released by mesothelial cells suppressed tumor cell growth using a co-culture cell system of GFP-expressing 60As6 (60As6-GFP) cells and a mouse mesothelial cell line, 1Cs-mM cells (Fig 3B). Twenty-four hours after 1Cs-mM cells were plated, 60As6-GFP cells were seeded at a 1: 4 (60As6-GFP: 1Cs-mM) ratio and treated with MNTX or vehicle for 72 h. As expected, treatment with MNTX significantly increased 60As6-GFP cell growth in this system (144.0 ± 23.3%, Fig 3C and 3D), indicating that peritoneal mesotheliumderived OGF can also arrest tumor cell growth. Myofibroblasts or carcinoma-associated fibroblasts are the most abundant cell type in the primary tumor and have been shown to promote resistance to anti-tumor drugs [19]. The mouse fibroblast cell line NIH3T3 also expresses Penk mRNA (S6A Fig). As with mesothelial cells, treatment with MNTX was associated with a significant increase in the growth of 60As6-GFP cells in the co-culture system with NIH3T3 cells (122.5 ± 12.6%, S6B and S6C Fig).

Blockade of OGF signaling by MNTX boosts the anti-tumor effect of Doc in vitro
We next evaluated the effect of MNTX on the Doc-induced inhibition of cell growth. In both 60As6 and HSC-42 cells, treatment with Doc alone and Doc/MNTX obviously suppressed cell growth relative to that in vehicle-treated cells (Fig 3E). Notably, the combination of MNTX with Doc significantly decreased cell growth compared with Doc alone in 60As6 cells (Doc: 65.3 ± 6.6%, Doc/MNTX: 40.5 ± 7.1%, Fig 3E). In contrast, this booster effect of MNTX was not observed in HSC-42 cells (Doc: 56.0 ± 15.0%, Doc/MNTX: 55.0 ± 16.7%), in which PENK expression was not detected (Fig 3E). MNTX had the same effect on the pancreatic cancer cell line PANC-1 (Doc: 71.3 ± 8.1%, Doc/MNTX: 51.3 ± 4.2%, Fig 3E), and pancreatic cancers often express OGF and are well known to show diffuse growth (S1 Fig). We further investigated the effect of the combination of MNTX and Doc on primary cultured GC cells (NSC-16C) derived from the ascites of a patient with diffuse-type GC. The results showed that MNTX clearly boosted the anti-tumor effect of Doc in this clinical subject as well as in a GC cell line with a high level of PENK expression (Doc: 39.2 ± 5.8%, Doc/MNTX: 19.2 ± 2.1%, Fig 3E).
Combined use of Doc and MNTX significantly prolongs survival in peritoneal dissemination model mice Conventional therapeutic modalities have failed to improve survival or outcomes among patients with peritoneal dissemination. Recently, intraperitoneal chemotherapy, which is the direct application of chemotherapeutic agents to macro/microscopic tumor seeding, has been considered to be a promising method for reducing the incidence of peritoneal dissemination [20,21]. Therefore, we established mouse models for the intraperitoneal administration of Doc, which corresponded to 3 different phases (early, middle, and late) in the progression of peritoneal dissemination (Fig 4A-4C). The xenografts of 60As6 as well as clinical samples ( Fig  1A) showed a high proportion of Ki-67-negative non-proliferating tumor cells (Fig 4D). A quite low dose of Doc (0.5 mg/kg) was used in these models, and therefore the toxicity of Doc was considered to be extremely low. Among the 3 different models, we selected the middlephase model, which has characteristics similar to those of patients with multiple small tumor nodules that cannot be detected by a computed tomography scan. By in vivo imaging with luciferase-expressing 60As6 cells (60As6Luc) [22], representative chronological tumor growth was observed in the mouse peritoneal cavity treated with saline, Doc or Doc/MNTX (0.3 mg/ kg) (Fig 4E). Beginning on day 7 after inoculation, the mice were divided into 4 groups based on photon counts. The mice were then treated intraperitoneally with the above reagents twice weekly until the endpoint criteria were met. In the saline-treated mice, tumor progression in the peritoneal cavity was observed on day 14 after inoculation. A marked increase in the volume of ascites was noted and moribund mice were observed around day 28. A high concentration of OGF was observed in mouse ascites ( On the other hand, mice that had been treated with Doc or Doc/MNTX tended to be stable with slower tumor growth up to day 28. No ascites formation was noted in any of the mice. Consistent with these results, the intraperitoneal administration of Doc or Doc/MNTX clearly prolonged survival compared with that in saline-treated mice (Fig 5A). Notably, the survival time of Doc/MNTX-treated mice was significantly greater than that of Doc-treated mice ( Fig  5A). Conversely, the combined use of Doc and OGF was likely to shorten survival relative to that with Doc alone (S7B Fig). There was no difference in survival time between MNTX-treated mice and saline-treated mice (S7C Fig). In accordance with these effects, visceral pain-related hunching behavior on day 35 after inoculation was apparently decreased in Doc/MNTX-treated mice compared with Doc-treated mice (Fig 5B). Importantly, a booster effect of MNTX with respect to Doc was not observed in an OGFR shRNA-transfectant, 60As6-OGFR-KD, in vivo (Fig 5C). The eradication of free tumor cells and suppression of the formation of multiple small tumor nodules on the membranes of the abdomen are major challenges in the treatment of peritoneal metastasis. To clarify the effects of Doc/MNTX in this regard, we monitored 60As6-GFP cells on the membranes of the abdomen. In mice treated with saline, several 1-3 mm tumors studding the mesentery adjacent to the bowel were observed on day 31 after inoculation (Fig 6A). In mice that had been treated with Doc or Doc/MNTX, fewer tumor nodules were noted on day 49 (Fig 6A, white box in the upper region of the cecum). However, isolated tumor cells and small tumor cell spheroids, which were located in the vascular bed of the mesentery, were found only in Doc-treated mice (Fig 6A). Based on the results of microscopic observation, there were significantly fewer single cells and spheroids in Doc/MNTX-treated mice than in Doc-treated mice (S6B and S6C Fig).

Discussion
Peritoneal dissemination is the most frequent form of metastasis of diffuse-type GC and a leading cause of death [4,23,24]. Up to half of all advanced GC patients will develop peritoneal dissemination, which is already present in 5-20% of patients who are examined for potentially curative resection [25]. The five-year survival rate in patients with peritoneal dissemination is less than 3%, with overall mean and median survival periods of 6.5 and 3.1 months, respectively [26]. Although systemic chemotherapy can improve median survival up to 12 months in all advanced GC, a similar survival benefit has not been seen in patients with peritoneal dissemination due to their lower sensitivity to chemotherapy [26]. Recent reports have suggested that the molecular mechanism that is responsible for chemoresistance is based on two modalities: a tumor cell-autonomous factor and a tumor microenvironment-related factor [27]. Generally, disseminated tumor cells can regulate stromal cells, such as cancer-associated fibroblasts and mesothelial cells, in promoting the formation of ascitic fluid [28]. The ascites contains cytokines, bioactive lipids and growth factors, which can influence the proliferation of tumor cells [28]. In the present study, we found that OGF is predominantly expressed in diffuse-type GCs compared with intestinal-type GCs. The expression of OGF and its receptor OGFR is also detected in mesothelial cells and fibroblast cells. OGF is highly concentrated in the cancer-associated ascites of not only peritoneal-dissemination model mice but also GC patients. Previous studies and our current data suggest that OGF causes cell growth suppression due to p21-mediated G1-phase arrest of the cell cycle [13]. The present study demonstrated that blockade of OGF signaling by MNTX significantly increases the cell growth of diffuse-type GC cells under a low nutrient condition and/or in the co-culture system with stromal cells that spontaneously produce OGF. As shown in Fig 2C and 2D, S5E Fig, and Fig 5C, down-regulation of OGFR negated these effects of OGF and MNTX in vitro and in vivo. These results suggest that OGF/OGFR signaling is dominant in the microenvironment of diffuse-type GC, which may contribute to the dormant state of tumor cells and the lower sensitivity to anti-tumor drugs.
The present study showed that intraperitoneal treatment with Doc significantly prevents ascites formation in peritoneal dissemination model mice. In addition, isolated tumor cells and spheroids, which may be resistant to Doc, on the peritoneal surface in Doc-treated mice were clearly diminished by the co-administration of MNTX with Doc. Finally, the combined use of MNTX and Doc significantly alleviated abdominal pain and prolonged survival compared with Doc alone. Therefore, intraperitoneal chemotherapy has great potential for stopping the progression of peritoneal metastasis in patients with free tumor cells (peritoneal-lavage cytologypositive patients) or with multiple but very small tumor nodules that cannot be detected by a computed tomography scan.
In conclusion, our findings indicate that MNTX boosts the anti-tumor effect of Doc through the recycling of OGF-induced cell growth arrest, and this booster effect leads to an improved QOL and prolonged survival in peritoneal dissemination model mice. Our data propose that a strategy of awakening and killing tumor cells may have great potential for resolving the problem of chemoresistance caused by tumor cell dormancy.
Supporting Information S1 Table. Primers for RT-PCR of human and mouse genes. ArrayScan HCS Reader and separated by cell population analysis based on EdU incorporation and DNA content (mean ± SD, n = 3 each, Ã p<0.05). B, a cell population analysis of G1 phase after co-treatment with OGF (10 -4 M) and MNTX (10 -5 M) for 48 h. A population of G1 phase was analyzed based on EdU incorporation and DNA content (mean ± SD, n = 3 each, Ã p<0.05). C, OGF induced p21 expression. 60As6 cells were transferred to serum-free medium for 96 h to synchronize cells, and subsequently treated with OGF (10 -4 M) for 3, 6, 9 and 12 h. Total protein was resolved by SDS-PAGE, and blotted with p21-specific antibody. D, RT-PCR and Western blotting analyses of OGFR in 60As6 cells treated with OGFR siRNAs or nontargeting control siRNAs. E, growth of 60As6 cells in the presence or absence of OGF (10 -4 M) 72 h after the transfection of OGFR siRNA. As a control, non-targeting control siRNA was used (mean ± SD, n = 3 each, Ã p<0.05).  Fig. Other in vivo experiments. A, a high concentration of OGF was observed in mouse ascites. The amount of OGF released into the ascites and peritoneal washings (PBS) obtained from mice 28 days after the inoculation of 60As6-Luc cells. The concentration of OGF was measured by ELISA. OGF was only detected in the ascites (mean ± SD, n = 5 each). B, in vivo effects of OGF or MNTX alone. Survival curves of middle-phase peritoneal metastasis model mice treated with saline, Doc, or Doc/OGF. Drug administration was started 7 days after the inoculation of 60As6-Luc cells. Mice were treated with Doc or a combination of Doc and OGF (10 mg/kg) 2 times a week until the endpoint criteria were met (n = 5, Ã p<0.05, vs. saline). C, survival curves of middle-phase peritoneal metastasis model mice treated with saline or MNTX. Mice were treated with saline or MNTX (0.3 mg/kg) 2 times a week until the endpoint criteria were met (n = 5). (TIF)