Cosmc overexpression enhances malignancies in human colon cancer

Abstract Cosmc is known as a T‐synthase‐specific molecular chaperone that plays a crucial role in the process of O‐glycosylation. Cosmc dysfunction leads to inactive T‐synthase and results in aberrant O‐glycosylation, which is associated with various tumour malignancies. However, it is unclear whether Cosmc has some other functions beyond its involvement in O‐glycosylation. In this study, we aimed to investigate the functional role of Cosmc in human colorectal cancer (CRC). We first assessed the expression levels of Cosmc in human CRC specimens and then forcedly expressed Cosmc in human CRC cell lines (HCT116, SW480) to examine its impact on cellular behaviours. The mechanisms for aberrant expression of Cosmc in CRC tissues and the altered behaviours of tumour cells were explored. It showed that the mRNA and protein levels of Cosmc were markedly elevated in human CRC specimens relative to normal colorectal tissues. The occurrence of endoplasmic reticulum (ER) stress may largely contribute to the increased Cosmc expression in cancer tissue and cells. Cosmc overexpression in CRC cells significantly promoted cell migration and invasion, which could be attributed to the activation of the epithelial‐mesenchymal transition (EMT) pathway rather than aberrant O‐glycosylation. These data indicate that Cosmc expression was elevated in human CRC possibly caused by ER stress, which further enhanced malignancies through the activation of EMT but independently of aberrant O‐glycosylation.

T-synthase. 7,8 Loss of Cosmc results in inactive T-synthase and aberrant O-glycosylation, characterized by the expression of Tn antigen. 7,8 Accumulating evidence indicates that dysfunctions in Cosmc such as acquired gene mutations, deletion, or hypermethylation are the prevailing mechanisms responsible for the phenotypes of aberrant O-glycosylation in many human diseases. 7,[9][10][11][12][13][14] Many in vitro experiments also showed that Cosmc knockdown promoted malignant behaviours in various types of cancer cells such as growth, adhesion, migration and invasion through a mechanism depending on O-glycosylation. Transfection of wild-type Cosmc restored O-glycosylation and attenuated malignant behaviours in cells. 4 However, some recently emerged conflicting findings indicated rather opposing roles of Cosmc. For instance, Huang et al reported that Cosmc overexpression promoted malignant behaviours in cancer cells through induction of aberrant O-glycosylation, whereas Cosmc knockdown reduced these malignancies, which indeed contradicted most reports. 15 Lee et al also showed that overexpression of Cosmc in human umbilical vein endothelial cells enhanced T-synthase, Tn antigen expression and cell growth. 16 Moreover, it is noticeable that in many human cancers expression of Cosmc and Tn antigen is equally elevated, 17 which is a paradox hard to explain by the theory of aberrant Oglycosylation. It is unclear whether elevated Cosmc expression has additional effects or is just an outcome resulting from the increased Tn antigen via feedback regulation. So far, the pathological role of Cosmc in human CRC remains largely elusive.
Here we first defined the expression of Cosmc in clinical CRC samples. We observed that Cosmc expression was up-regulated at both mRNA and protein levels in cancer samples compared with normal colorectal tissues. By investigating the mechanisms of elevated Cosmc expression in human CRC tissues, we found that ER stress was likely an important cause for elevated Cosmc expression. We further assessed the functional impact of Cosmc by using Cosmcoverexpressing CRC cell lines and found that Cosmc overexpression significantly promoted cell migration and invasion. Notably, Cosmc overexpression did not affect the process of O-glycosylation, because there were no appreciable changes in Tn antigen, T-synthase and/or ppGalNac-Ts. Interestingly, we found that the epithelialmesenchymal transition (EMT) process was activated prominently in Cosmc-overexpressing cells, which might be responsible for the altered oncogenic features in cells.

| Western blot analysis
Frozen tissues or cells were lysed using RIPA lysis buffer

| Transwell migration and invasion assays
After being starved for 24 hours, 2 × 10 5 cells with the serumfree medium were seeded into the upper chamber (8-μm pore size, BD Bioscience, USA) pre-coated with or without Matrigel (BD Bioscience, USA). Complete medium containing 10% FBS was added to the bottom of each well as a chemoattractant.
After being cultured for 24 hours, non-migrated or non-invaded cells were removed off the upper chamber using a cotton swab while the migratory or invasive cells were counted after fixation with 4% paraformaldehyde and being stained with 0.1% crystal violet.

| Flow cytometry analysis
To analyse the expression of Tn antigen in Cosmc-overexpressing cells and control cells, flow cytometry analysis was performed.
Briefly, 1 × 10 5 cells were routinely trypsinized, suspended in cold PBS and then incubated with 10 μg/mL mouse anti-Tn IgM mAb or mouse IgM isotype-antibody as a control (Santa Cruz, USA) for 2 hours at 4°C. 4 After being washed three times with PBS, cells were incubated with PE-labelled goat antimouse IgM secondary antibody (BD, USA) for 1 hour at 4°C. Cells were washed three times and suspended in 500 μL cold PBS per tube, followed by further analysis on a Flow Cytometer (Canto II, BD Bioscience, USA).

| Statistical analysis
All data were analysed with GraphPad Prism 6.0 (GraphPad Software, La Jolla, CA, USA). All experiments were repeated at least in triplicate, and the data were presented as mean ± standard deviation (SD). Differences were analysed by Student's t test (unpaired, 2-tailed), and P < .05 was considered statistically significant.

| Cosmc expression was up-regulated in human CRC tissues
Cosmc is an essential molecular chaperone for T-synthase and dysfunction of Cosmc results in abnormal exposure of Tn antigen, which is recognized as a hallmark for aberrant O-glycosylation. 8,9 However, the precise role of Cosmc in clinical CRC remains largely

| ER stress contributed to elevated Cosmc expression in human CRC
Although loss of functional Cosmc caused by mutation, deletion, or hypermethylation has been widely recognized as a prevailing mechanism underlying aberrant O-glycosylation observed in several types of cancers including CRC, 7,[9][10][11][12][13][14] little is known why there is an elevated expression of Cosmc in CRC, which appears to contradict the data obtained from some tumour-derived cell lines and certain types of malignancies such as pancreatic cancer. 18 Here we sought to investigate the molecular mechanisms underlying elevated Cosmc expression in CRC tissues. It has been proposed that expression of Cosmc may be induced by ER stress, which occurs frequently in malignancies. 17 Moreover, Tn antigen has been discovered to be abundantly expressed in human CRC, which may promote Cosmc production via a feedback loop. 17 To test these possibilities, we examined the expression levels of ER stress markers such as GRP78 and CHOP in another set of frozen CRC tissues (16 normal colorectal tissues, 32 CRC tissues).
Western blot results showed that both GRP78 and CHOP were significantly up-regulated in CRC tissues as compared with normal controls, indicative of the occurrence of ER stress, which also correlated with the expression of Cosmc (Figure 2A). Notably, there was a strong correlation between expression of Cosmc and ER stress in human CRC tissues ( Figure 2B).
To definitively address this possibility, we treated two types of  Figure 2C). It is noticeable that Cosmc was also significantly increased at its protein levels following ER stress, thereby supporting the likelihood that ER stress may induce Cosmc expression in cancer cells.
In addition, it has been confirmed that Cosmc dysfunction induces Tn antigen expression, 8

| The effects of Cosmc overexpression on CRC cell behaviours
We next studied whether the elevated Cosmc expression influences

| Cosmc overexpression-mediated oncogenic alterations were not attributed to aberrant Oglycosylation
Most studies have confirmed that aberrant O-glycosylation mediated by the loss of functional Cosmc contributes to tumour progression and metastasis in a broad range of human cancers including CRC. 4,9,18 Here we questioned whether the tumour-promoting ef-

| Cosmc overexpression activated the EMT signalling pathway
We further explored how Cosmc overexpression alters CRC cellular behaviours. The EMT process is an essential candidate signalling pathway associated with cancer progression and metastasis. 32 We thereby analysed whether Cosmc overexpression can influence the EMT process. As shown in Figure

| D ISCUSS I ON
The discovery of Cosmc was an important biological event in the research history of O-glycosylation. [7][8][9] More recent studies disclosed that Cosmc is a specific molecular chaperone for T-synthase and Cosmc loss-of-function leads to impaired expression or function of T-synthase, and subsequently an occurrence of aberrant O-glycosylation, which is characterized by the expression of Tn antigen. 8,9,18 Therefore, many studies investigated the functional role of aber- behaviours. 15 In their study, Cosmc knockdown was demonstrated to be able to decrease cancer cellular malignancies, which actually contradicted most recent findings. Thus, the role of Cosmc in addition to mediating O-glycosylation needs to be fully explored.
In this study, we focused our attention on the impact of Cosmc overexpression in human colon cancer. We first assessed the expression levels of Cosmc in human CRC tissues. We found that Cosmc expression was up-regulated in tumour tissues relative to normal colorectal tissues at its mRNA and protein levels. TCGA data set analysis also supported our observations. Because Cosmc loss-offunction is a prevailing mechanism for aberrant O-glycosylation detected in many human cancers, 9,18,22

ACK N OWLED G EM ENTS
This work was supported by National Natural Science Foundation of China (81672338 to Tao Wen). We thank Tongzhong Ju from Emory University, Atlanta, USA for critical editing of the manuscript.

CO N FLI C T O F I NTE R E S T
The authors confirm that there are no conflicts of interest.

AUTH O R CO NTR I B UTI O N S
Tianbo Gao and Tan Du participated in the study design and per-

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.