DIMT1 overexpression correlates with progression and prognosis in gastric carcinoma

    DIMT1 overexpression correlates with progression and prognosis in gastric carcinoma Guangyi Liu MD, Xudong Peng MD, Yongqian Cai MM, Anqi Cheng MM, Lang Zha MD, Ziwei Wang MD PII: DOI: Reference:

S0046-8177(17)30192-2 doi: 10.1016/j.humpath.2017.02.034 YHUPA 4240

To appear in:

Human Pathology

Received date: Revised date: Accepted date:

29 November 2016 18 January 2017 10 February 2017

Please cite this article as: Liu Guangyi, Peng Xudong, Cai Yongqian, Cheng Anqi, Zha Lang, Wang Ziwei, DIMT1 overexpression correlates with progression and prognosis in gastric carcinoma, Human Pathology (2017), doi: 10.1016/j.humpath.2017.02.034

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT DIMT1 overexpression correlates with progression and prognosis in gastric carcinoma Guangyi Liu MDa, Xudong Peng MDa, Yongqian Cai MMb, Anqi Cheng MMa, Lang Zha MDc,

a

RI P

T

and Ziwei Wang, MDc,*

Department of Gastrointestinal Surgery, Laboratory Research Center, The First Affiliated

Department of General Surgery, DaZhou Central Hospital, Dazhou, Sichuan, PR China [AU:

NU

b

SC

Hospital of Chongqing Medical University, Chongqing, PR China [AU: postal code?]

postal code?]

Department of Gastrointestinal Surgery, The First Affiliated Hospital of Chongqing Medical

MA

c

ED

University, Chongqing, PR China 400010

PT

*Corresponding Author:

Ziwei Wang, M.D., Department of Gastrointestinal Surgery, The First Affiliated Hospital of

CE

Chongqing Medical University, No. 1 Youyi Road, Yuan Jiagang, Yuzhong District, Chongqing,

AC

PR China, 400010. E-mail: [email protected]

Running title: Gastric carcinoma and DIMTI Conflicts of Interest: None Source of Funding: This research did not receive any specific grants from funding agencies in the public, commercial, or not-for-profit sectors.

1

ACCEPTED MANUSCRIPT Summary We investigated the expression of dimethyladenosine transferase 1 homolog (DIMT1) in human

T

gastric carcinoma (GC) tissues, pericarcinoma histologically normal tissues (PCHNTs), and

RI P

normal gastric tissues and explored its clinical significance. Immunohistochemistry staining was used to detect the expression of DIMT1, and the findings were compared with clinicopathologic

SC

features of patients with GC. The result also was ascertained by Western blotting. The Kaplan-

NU

Meier method and log-rank test were employed to compare the overall survival rate and time in the DIMT1 low-level and high-level expression groups. Immunohistochemical staining indicated

MA

that the expression of DIMT1 in GC tissues (65 of 75; 86.7%) was significantly more common (P < .001) than that in PCHNTs (14 of 75; 18.7%) and normal gastric tissues (2 of 12; 16.7%).

ED

High expression of DIMT1 correlated closely with differentiation (P = .023), invasion (P

PT

= .042), lymph node metastasis (P = .008), distant metastasis (P = .006), and TNM stage (P = .013). Western blotting showed that DIMT1 expression correlated positively with TNM stage

CE

and implied that more advanced TNM stage was accompanied by higher expression of DIMT1

AC

(P < .001). Kaplan-Meier survival analysis showed that high DIMT1 expression correlated significantly (P < .001) with a poor prognosis. Our data suggest that DIMT1 is a useful molecular biomarker to predict tumor progression and prognosis in patients with GC.

Keywords: DIMT1; Gastric carcinoma; Prognosis; Tumor biomarker

2

ACCEPTED MANUSCRIPT 1. Introduction Gastric carcinoma (GC) is a common disease in some geographic areas of the world and

T

is characterized by latency, high malignancy, and a poor prognosis. It is the second leading cause

RI P

of cancer-related death worldwide, and about one of every four cancer patients who dies of disease will die of GC. The report by Chen et al [1] showed that GC was the second most

SC

common cancer in China, with about 679 100 new cases and 498 000 deaths in 2015. Not only

NU

that, but the incidence has increased each year, and the disease has begun to affect younger persons, including in China. Helicobacter pylori (HP) infection [2], chronic atrophic gastritis

MA

(CAG)[3], and environmental pollution are the major causative factors [4]. The causative mechanism depends on multiple factors and genes and involves many kinds of oncogene

ED

activation or anti-oncogene inactivation. However, the etiologic factors and mechanisms of GC

PT

still are not entirely clear.

Surgery, chemotherapy, radiation, and traditional Chinese medicines are the main means

CE

of treatment of GC at present. The strategy to reduce the mortality rate GC requires diagnosis in

AC

an early stage and adequate treatment [5]. Because of deficiencies in diagnosis and treatment, the mortality rate of GC has remained stubbornly high. Facing this dire situation, the exploration of new diagnostic and prognostic biomarkers for GC is urgently needed. The DIM1 dimethyladenosine transferase 1 homolog gene（DIMT1）is located on chromosome 5q12.1 and encodes a protein with 313 amino acids. The expression of the miRNA210 is decreased significantly, and DIMT1 is increased, in the chronic gastritis caused by H. pylori. Suppressed miRNA210 can upregulate expression of DIMT1 and could promote the proliferation of gastric epithelial cells [6]. Other studies also found that there was a close correlation between expression of DIMT1 and NEDD9, UBC, and RIOK2. Those three proteins

3

ACCEPTED MANUSCRIPT are closely associated with proliferation, apoptosis, invasion, migration, and other malignant biological behaviors of tumor cells [7-10]. Furthermore, they can be used to monitor the

T

recurrence or metastasis of the tumor and also to assess a patient’s prognosis. However, the role

RI P

of DIMT1 in GC has not been reported. Therefore, the aim of our study was to evaluate the

SC

clinical significance of DIMT1 in the progression and prognosis of GC patients.

NU

2. Materials and methods

MA

2.1. Ethics statement

Written informed consent was obtained from all patients or their family members. We

CE

2.2. Tissue specimens

PT

Chongqing Medical University.

ED

obtained approval for the study from the Ethics Committee at the First Affiliated Hospital of

AC

This study included 75 consecutive patients who were admitted to the center of Gastrointestinal Surgery of the First Affiliated Hospital of Chongqing Medical University between January 2011 and June 2011. The patients were 48 men and 27 women with a median age of 62 years (range 32–84 years) at the time of surgery; none of the patients had received preoperative anticancer treatment. All patients underwent standard D2 lymphadenectomy according to the criteria of the American Joint Committee on Cancer (AJCC 7th edition 2010) [11]. Each of the cancers was confirmed as primary gastric adenocarcinoma by postoperative histopathologic examination. Among them, 14 tumors were well differentiated, 28 were moderately differentiated, and 33 were poorly differentiated. There were 15 patients with distant

4

ACCEPTED MANUSCRIPT metastases, including 13 with peritoneal cavity or greater omentum metastasis and 1 each with liver and lung metastases. The pathologic TNM stage criteria categorized 12 cases as stage I, 19

T

as stage II, 29 as stage III, and 15 as stage IV. Of the 75 patients, 61 (81.3%) underwent adjuvant

RI P

chemotherapy after radical operation.

The GC tissues and pericarcinoma histologic normal tissues (PCHNTs) were divided into

SC

two groups. One was stored in liquid nitrogen immediately after resection, and the other was

NU

processed into 4% Formamint-fixed [AU: should this be formalin-fixed?], paraffin-embedded blocks, which were archived in the Laboratory of Oncology after examination. The PCHNTs

MA

were cut more than 3 cm from the tumor margin and demonstrated absence of cancer or dysplastic cells. In addition, 12 normal gastric tissue samples were obtained from nontumor

PT

ED

patients examined at our hospital. The clinical data of these cases are true and complete.

2.3. Immunohistochemistry staining

CE

The immunohistochemistry staining was performed to study protein expression in the

AC

matched GC tissues and PCHNTs and normal gastric tissues. The 3-μm paraffin slices were baked for 48 h at 60C, then placed in the oven and baked for 45 min before the start of the experiment. Paraffin slices were dewaxed in xylene and rehydrated in an alcohol gradient. After three rinses in phosphate-buffered saline (PBS) lasting 5 min each, the slices were immersed in 3% hydrogen peroxide for 10 min at 37C to suppress endogenous peroxidase activity. Antigen was retrieved in citric acid buffer (pH 6.0) heated at high heat in a microwave oven for 3 min and at low heat for 30 sec, this heating being repeated for cycles with natural cooling. The slices were blocked with 10% normal goat serum (NGS) at room temperature for 1 h to reduce nonspecific binding. Slices were incubated overnight at 4C with rabbit anti-DIMT1 polyclonal

5

ACCEPTED MANUSCRIPT antibody (catalog number ab184978; Abcam, Cambridge, UK) at 1:500 diluted in NGS. The negative control was prepared with PBS. Slices were kept at room temperature for 1 h, and after

T

three washes, the rest of the steps were performed according to the instructions in the ready-to-

RI P

use SABC test kit (catalog number SA1053; BSOTER, Wuhan, China). Finally, the sections were dehydrated in graded ethanol, coverslipped in dimethylbenzene, and deposited in neutral

SC

balsam [AU: is balsam correct?].

NU

The immunohistochemical staining results were assigned a mean score considering both the intensity of staining and the proportion of tumor cells with an unequivocally positive

MA

reaction, which was defined as brown signals in the nucleus and cytoplasm. The staining intensity (SI) was scored as follows: 0 (negative); 1 (weak staining); 2 (moderate staining); and 3

ED

(strong staining). The percentage of positive cells (PP) was defined as follows: 0 (not detected);

PT

1 (less than 25% reactive); 2 (26% to 50% reactive); 3 (51% to 75% reactive); and 4 (>75% reactive). Ten fields of each slice were used for the immunohistochemical evaluation. When the

CE

staining was heterogeneous, each component was scored independently and summed for the

AC

results. For example, a specimen containing 75% tumor cells with moderate staining intensity (3 × 2 = 6) and another 25% tumor cells with weak staining intensity (1 × 1 = 1) received a final score of 6 + 1 = 7. The raw data were converted into the immunoreactivity score (IRS) proposed by Remmele and Stegner [12]: SI × PP. Theoretically, the IRS could range from 0 to 12. An IRS of 0 was scored as negative (−), 1–4 was considered weak (+), 5–8 was considered moderate (++), and 9–12 was considered strong (+++). Each section was assessed independently by two pathologists without knowledge of the clinical data.

2.4. Western blotting

6

ACCEPTED MANUSCRIPT Paired GC and PCHNTs specimens from all 75 GC patients were used for Western blotting. Total protein was extracted by a mixed solution of lysis buffer (RIPA; catalog number

T

P0013B; Beyotime, Shanghai, China) and protease inhibitors (catalog numbers PMSF and

RI P

ST506; Beyotime). Adjusted to the same concentration, equal volumes of protein were separated by 10% sulfate polyacrylamide slab gel electrophoresis and transferred to a polyvinylidene

SC

fluoride (PVDF) membrane. Next, the membranes were blocked for 2 h at room temperature

NU

with 5% nonfat milk. After incubation overnight with polyclonal rabbit anti-human DIMT1 at 1:200 (catalog number ab184978; Abcam) or a rabbit anti−β-actin as internal reference at 1:1000

MA

dilution (catalog number ab8227; Abcam) at 4C. The PVDF membranes were washed three times in Tris-buffered saline Tween-20 (TBST) for 15 min on an incubator shaker and then

ED

incubated with horseradish peroxidase−conjugated goat anti-rabbit secondary antibody (catalog

PT

number ab6721; Abcam) at 1:3000 for 2 h at room temperature. The membrane then was washed three times in TBST for 15 min before signals were detected using BeyoECL plus catalog

AC

CE

number (catalog number P0018; Beyotime).

2.5. Postoperative follow-up Postoperatively, each patient was scheduled for follow-up examinations for least 5 years or until lost to follow-up or death [13]. Physical examinations, measurement of weight, routine blood counts, liver and kidney function tests, and serum tumor marker assays were done every 3 months for the first year, every 6 months for the second year, and annually thereafter. In addition, gastroscopy and contrast-enhanced computed tomography scans of the thorax and abdomen were done every 6 months for the first year and annually thereafter for as long as 5 years. All deaths were confirmed by the families and verified by the local public security

7

ACCEPTED MANUSCRIPT authorities. Survival for more than 60 months, patients lost to follow-up, and deaths from other

T

than cancer were managed by censoring the data.

RI P

2.6. Statistical analysis

Data were processed using SPSS 22.0 (SPSS Inc, Cary NC). The correlations between

SC

DIMT1 expression and clinicopathologic features were subjected to the Pearson 2 test and

NU

multivariate logistic regression analysis. Survival curves were created by the Kaplan-Meier method, and the log-rank test was used to calculate differences between the curves. The other

MA

data were analyzed statistically using the paired or independent two-tailed Student’s t test. P

ED

< .05 was considered statistically significant.

PT

3. Results

3.1. Expression of DIMT1 in benign and malignant gastric tissues

CE

Expression of DIMT1 protein was seen mainly in the nucleus but faintly in the cytoplasm

AC

in both benign and malignant tissues (Fig. 1). The protein was expressed in gastric cancer tissues, and the total number of positive cases was 65 (86.7%), which included 22 cases (29.3%) with strong expression (+++), 25 cases (33.3%) with moderate expression (++), and 18 cases (24.0%) with weak expression (+) (Fig. 1A and B). Among the 75 matched PCHNTs, 14 (18.7%) were positive for DIMT1 protein expression (Fig. 1C). Only two positive tissues were identified in the 12 patients without cancer (16.7%; Fig. 1D). The expression of DIMT1 was significantly greater in gastric cancer tissues than that in their matched PCHNTs and normal gastric tissues, and there were significant differences by statistical analysis (2 = 70.767, P = .000; 2 = 28.896, P = .000, respectively), but there was no statistical significance between normal gastric tissues and

8

ACCEPTED MANUSCRIPT matched PCHNTs (2 = 0.528, P = .913) (Table 1). The extent of expression of DIMT1 was related to the invasion by the tumor, and DIMT1 was more abundant in the superficial or

T

mucosal part of the tumor and lower in the deeper invasive front (Supplementary Fig. 1).

RI P

Expression of DIMT1 in primary carcinoma was significantly higher than that in metastatic

SC

lymph node and greater omentum metastasis (Supplementary Fig. 2).

Compared with the expression of DIMT1 protein in PCHNTs, Western blotting revealed

NU

a significant elevation in 66 of 75 GC cases. We selected two cases to represent Western blotting results in each of the different TNM stages (Fig. 2A). The average DIMT1 protein expression in

MA

different TNM stages was significantly higher than in PCHNTs (P < .001) (Fig. 2B–F). The relative expression of DIMT1 in different TNM stages was analyzed by the independent-sample t

ED

test. Expression of DIMT1 protein in stages II, III, and IV was significantly higher than in stage I

PT

(all P < .001), and expression in stages II and III was significantly higher than in stage IV (P = .014 and P = .001, respectively), and there was no significant differences between stages II and

CE

III (P = .085). In order to observe the expression of DIMT1 in different clinical stages, we

AC

randomly selected 8 cases of GC tissues and 2 cases in each stage. The expressions of DIMT1 proved to be highest in stage III and lowest in stage I, with expression in stages II and stage IV somewhere in between (Supplementary Fig. 3).

3.2. Expression of DIMT1 correlates with the clinicopathologic factors Correlations between immunohistochemical DIMT1 expression in tissues and various clinicopathologic characteristics of patients with GC were analyzed by χ2 statistics and are listed in Table 2. More than half (64%) of the GC patients were men, but there was no significant difference in the expression rate or expression intensity of DIMT1 between men and women (2

9

ACCEPTED MANUSCRIPT = 6.637; P = .084). When the GC patients were divided into two groups by age (60 or greater and less than 60), there was no statistically significant difference between the groups (2 = 4.428; P

T

= .219). Also, there were no significant correlations between DIMT1 expression and tumor size

RI P

( 2= 1.417; P = .702) or degree of differentiation (2 = 11.282; P = .080). Overexpression of

SC

DIMT1 was significantly related to tumor invasion (2 = 17.797; P = .038), lymph node metastasis (2 = 14.541; P = .002), and distant metastasis (2 = 19.625; P = .000). Expression of

NU

DIMT1 was an independent prognostic factor (2 = 20.995; P = .013).

MA

Based on the single-factor analysis, we used multivariate logistic regression analysis to analyze the relation between clinical factors and the expression of DIMT1. This analysis

ED

revealed that poor pathologic differentiation (P = .023), invasion (P = .042), lymph node metastasis (P = .008), distant metastasis (P = .006), and advanced clinical TNM stage (P = .013)

CE

PT

were independent predictors of high DIMT1 expression (Table 3).

3.3. Expression of DIMT1 is negatively correlated with GC prognosis

AC

Patients were divided into two groups according to the immunohistochemical staining of DIMT1. The first group showed no or weakly positive (−/+) expression, and the other group had moderate or strong (++/+++) expression. The former group had 28 patients. At the end of the 60month follow-up period, 9 of these patients survived, and 19 patients had died. The median survival was 35 mo, and the 5-year overall survival rate was 32.1%. The second group had 47 patients, of whom only one survived. The median survival was 20 mo, and the 5-year overall survival rate was 2.2%. Survival in the low-expression group was significantly longer than that in the high-expression group (2 = 18.552; P = .000) as show in Fig. 3C.

10

ACCEPTED MANUSCRIPT For a closer look at the effect of DIMT1, we analyzed survival based on the expression intensity of DIMT1 by separating stages I and II from stages III and IV. In stages I and II, the

T

low−DIMT1 expression group had 16 patients, of which 9 died, the cumulative 5-year survival

RI P

rate being 43.8%. The high−DIMT1 expression group had 15 patients of whom 14 died, the 5year survival rate being only 6.7%. The median survival time was 43 mo for the former group

SC

and 29 mo for the latter. There was a statistically significant difference between the groups (2 =

NU

7.530; P = .006; Fig. 3A). Only two patients out of 12 survived more than 60 mo in the lowexpression group in Stages III and IV, the 5-year survival rate being 16.7%. None survived more

MA

than 42 months in the high−DIMT1 expression group. The median survival time was 24 and 17 mo, respectively. As expected, the difference between the two groups was significant (2 =

PT

ED

5.739; P = .017; Fig. 3B).

CE

4. Discussion

Nowadays, GC is a common malignant tumor in many countries [1,14]. The latest

AC

findings showed that the incidence and mortality rate of GC are second only to those of lung cancer. However, the incidence and mortality rate of GC rank first among all malignant tumors in many rural areas in China, and the patients with GC in rural areas are almost three times more likely to die than those in urban areas [1]. The disease has already influenced peoples’ productivity and life seriously. Although there now are advanced diagnostic and treatment technologies to use against the disease, including high-resolution computed tomography (HRCT), positron-emission tomography–computed tomography (PET–CT), combination surgery, radiation therapy, and chemotherapy, there has been little improvement the prognosis or quality of life of affected patients. Conventional treatments have some shortcomings, for example, long

11

ACCEPTED MANUSCRIPT treatment cycles, limited options, significant side effects, high recurrence rate, and many other issues; so there has been a search for new anticancer strategies, including targeted drugs,

T

antibodies, and cancer vaccines [15]. Ultimately, any anticancer strategies, whether inexpensive

RI P

or expensive, should be based on the correct diagnosis of the cancer. With earlier correct diagnosis and finding the cancer before the symptoms appear, we will have a better chance of

SC

beating the disease. Therefore, there is an urgent need for early detection using biological

NU

markers and for novel therapeutic targets.

Kiga et al [6] identified DIMT1 as an miR-210 target and demonstrated that inhibition of

MA

miR-210 expression augmented cell proliferation by activating DIMT1 expression in the chronic gastritis model. It is generally known that chronic gastritis is one of the causative factors of GC,

ED

and various amounts of miR-210 are found in different solid tumors. Moreover, one study

PT

reported that there is a close correlation between DIMT1 and NEDD9, UBC, and others. Both NEDD9 and UBC are positioned to interact with a number of the key proteins coordinating

CE

signaling cascades that control proliferation and apoptosis, migration, and invasion [16]; and

AC

Izumchenko et al [17] found that NEDD9 promoted cancer metastasis by activating focal adhesion kinase (FAK) [18]. The invasion, metastasis, and recurrence of GC are the most significant factors affecting the prognosis and also the main cause of death. Numerous studies have confirmed that many genes, including NEDD9, UBC, and FAK, affect the biological behavior of gastric cancer [19,20]. This finding suggests that these genes play some important role in the disease. In view of DIMT1 and its close genetic links to other cancer genes, we think that DIMT1 protein may be an important part of the occurrence and development of GC. It may stimulate metastatic growth by promoting cancer cell proliferation, invasion, migration, and

12

ACCEPTED MANUSCRIPT angiogenesis and by inhibiting apoptosis. In view of the current understanding of GC, our group set out to examine the role of DIMT1 in the disease.

T

In our experimental study, the SP immunohistochemical method was used to detect

RI P

expression of DIMT1 in 12 normal gastric mucosa tissues, 75 GC tissues, and 75 paired PCHNTs. For the first time, it was revealed that the expression of DIMT1 in GC tissues is

NU

difference between the PCHNTs and normal tissue.

SC

significantly higher than that in paired PCHNTs and normal tissues, and there was no obvious

On this basis, the results, combined with clinicopathologic features, were analyzed

MA

statistically. Single-factor analysis showed that the expression of DIMT1 correlated significantly with the depth of tumor invasion, lymph node metastases, distant metastasis, and TNM stage.

ED

The results of multivariate logistic regression also suggested a noticeable relation to the above-

PT

mentioned features. The degree of pathologic differentiation is also included. Usually, multivariate logistic regression is more rational and conforms to reality better than single-factor

CE

analysis. No significant difference was found among patient sex and age and tumor size in the

AC

expression of DIMT1 by these two methods. In fact, the depth of tumor invasion and tumor size are two different concepts. During the development of alimentary canal neoplasms, the tumor can exhibit either endoluminal or extraluminal growth [21]. Depth of tumor invasion depends greatly on the type of growth, although the diameter of some endoluminal growths can be large without the tumor having infiltrated the muscular layer of the stomach wall [22]. Some of the relatively smaller tumors that have extended into the tunica muscularis or serosa may have a high propensity for local aggression and lymph node and distant metastasis. We found that lymph node and distant metastasis correlated significantly with the depth of tumor invasion. In our

13

ACCEPTED MANUSCRIPT experiment, the expression intensity of DIMT1 was highly correlated with deep infiltration of the primary tumor, all other measures being equal.

T

Because of the results of immunohistochemistry staining, the expression of DIMT1

RI P

protein in GC was detected with Western blotting, with classification according to the TNM stage. Interestingly, the results of Western blotting were not completely consistent with what

SC

we expected. We thought the higher the TNM stage, the higher the expression of DIMT1

NU

protein would be. However, the expression of DIMT1 displayed a tendency to decrease from stage III to stage IV. Thus, if we look only at the experimental data, the results of Western

MA

blotting and immunohistochemistry do not match. Actually, we overlooked a crucial factor the IRS scoring system. In this system, a score between 9 and 12 marks tumors with strong

ED

expression, but the results of Western blotting are measurement data, and those two are

PT

completely different. So we hold that the results of the two experiments are compatible. In order to study the influence of DIMT1 on the prognosis of patients with GC, the

CE

Kaplan-Meier survival analyses were used to compute the survival analyses. Whether from an

AC

overall perspective or different tumor TNM stages, higher expression of DIMT1 was associated with a shorter survival time of patients with GC. Thus, overexpression of DIMT1 is negatively related with the prognosis. These results suggested that DIMT1 has an important function in GC progression and prognosis. However, the precise mechanism of DIMT1 action in GC remains elusive, but according to the results of our study, we think it can serve as a marker to diagnose and evaluate GC. Next, we will explore the precise mechanism of DIMT1 action in GC by experiments in vitro and with animals. . 5. Conclusion

14

ACCEPTED MANUSCRIPT To the best of our knowledge, this paper is the first to report the involvement of DIMT1 in GC. We found that expression of DIMT1 is significantly increased in GC tissues compared

T

with matched PCHNTs and normal gastric tissues. In GC patients, the high expression of DIMT1

RI P

correlated closely with tissue invasion, lymph node metastasis, distant metastasis, and a high TNM grade, whereas there was no relation between DIMT1 expression and age, sex, or tumor

SC

size or differentiation. In addition, the expression of DIMT1 in GC tissues was negatively

NU

correlated with the survival time. It appears that DIMT1 is a novel diagnostic and therapeutic target, while the amount of DIMT1 in GC tissues is a useful predictor of prognosis for patients

MA

with GC.

ED

Acknowledgement

Chen W, Zheng R, Baade PD, et al. Cancer statistics in China, 2015. CA Cancer J

AC

[1]

CE

References

PT

We thank MAC Sandrie for English language assistance.

Clinicians 2016;66:115–32. [2]

Cover TL. Helicobacter pylori diversity and gastric cancer risk. mBio 2016;7:e0186901815.

[3]

Inoue M, Tajima K, Matsuura A, et al. Severity of chronic atrophic gastritis and subsequent gastric cancer occurrence: a 10-year prospective cohort study in Japan. Cancer Lett 2000;161:105–12.

[4]

Karagulle M, Fidan E, Kavgaci H, Ozdemir F. The effects of environmental and dietary factors on the development of gastric cancer. J BUON 2014;19:1076–82.

15

ACCEPTED MANUSCRIPT [5]

Sumiyama K. Past and current trends in endoscopic diagnosis for early stage gastric cancer in Japan. Gastric Cancer 2017;20(Suppl 1):20-7. Kiga K, Mimuro H, Suzuki M, et al. Epigenetic silencing of miR-210 increases the

T

[6]

RI P

proliferation of gastric epithelium during chronic Helicobacter pylori infection. Nature Commun 2014;5:4497.

Feng J, Zhao J, Xie H, et al. Involvement of NEDD9 in the invasion and migration of

[8]

NU

gastric cancer. Tumour Biol 2015;36:3621–8.

SC

[7]

Zhang Q, Wang H, Ma Y, et al. Overexpression of Nedd9 is a prognostic marker of

[9]

MA

human gastric cancer. Med Oncol 2014;31:33.

Liu K, Chen H-L, Wang S, et al. High expression of RIOK2 and NOB1 predict human

Cui L, Sun MM, Zhao ZH, et al. BLCA-4 and UBC combined detection for early

PT

[10]

ED

non-small cell lung cancer outcomes. Sci Rep 2016;6:28666.

diagnosis of bladder cancer. J Biol Regulators Homeostatic Agents 2016;30:485–490. Marano L, Boccardi V, Braccio B, et al. Comparison of the 6th and 7th editions of the

CE

[11]

AC

AJCC/UICC TNM staging system for gastric cancer focusing on the “N” parameterrelated survival: the monoinstitutional NodUs Italian study. World J Surg Oncol 2015;13:215. [12]

Remmele W, Stegner HE. Recommendation for uniform definition of an immunoreactive score (IRS) for immunohistochemical estrogen receptor detection (ER-ICA) in breast cancer tissue. Der Pathol 1987;8:138–40.

[13]

Shoda K, Ichikawa D, Komatsu S, et al. Postoperative follow-up of patients with gastric cancer according to nodal status. Anticancer Res 2013;33:4645–9.

16

ACCEPTED MANUSCRIPT [14]

Siegel RL, Miller KD, Jemal A. Cancer statistics, 2015. CA: Cancer J Clinicians 2015;65:5–29. Chui SY, Clay TM, Lyerly HK, Morse MA. The development of therapeutic and

T

[15]

RI P

preventive vaccines for gastric cancer and Helicobacter pylori. Cancer Epidemiol Biomarkers Prevent 2005;14:1883–9.

Kim M, Gans JD, Nogueira C, et al. Comparative oncogenomics identifies NEDD9 as a

SC

[16]

[17]

NU

melanoma metastasis gene. Cell 2006;125:1269–81.

Izumchenko E, Singh MK, Plotnikova OV, et al. NEDD9 promotes oncogenic signaling

[18]

MA

in mammary tumor development. Cancer Res 2009;69:7198–206. Park JH, Lee BL, Yoon J, et al. Focal adhesion kinase (FAK) gene amplification and its

Hayashi A, Aishima S, Inoue T, et al. Decreased expression of focal adhesion kinase is

PT

[19]

ED

clinical implications in gastric cancer. HUM PATHOL 2010;41:1664–73.

associated with a poor prognosis in extrahepatic bile duct carcinoma. HUM PATHOL

Liu Y, Wang D, Zhao KL, et al. NEDD9 overexpression correlates with poor prognosis

AC

[20]

CE

2010;41:859–66.

in gastric cancer. Tumour Biol 2014;35:6351–6. [21]

Seddik H, Rouibaa F, Massit H, Hamdi FZE, Benkirane A. Carcinome épidermoïde de l’œsophage à développement extraluminal. Presse Méd 2011;40:889–90.

[22]

von Rahden BHA, Brücher BLDM, Sarbia M. Esophageal squamous cell carcinoma with entirely intramural growth pattern. Virchows Arch 2006;448:862–6.

17

ACCEPTED MANUSCRIPT

Figure Legends

T

Fig. 1 Immunohistochemical staining of DIMT1 in gastric carcinoma (GC) tissues,

RI P

pericancerous histologically normal tissues (PCHNTs), and gastric normal tissues (original

NU

expression in PCHNTs and normal gastric tissues.

SC

magnification, ×200). A and B, Strong expression of DIMT1 protein. C and D, Absence of

Fig. 2 Expression of DIMT1 was upregulated in GC tissues. A, Western blot analysis of two

MA

representative paired samples of GC tissues and PCHNTs in different TNM stage (stage I, stage IV). Total proteins were prepared from specimens of GC tissues (T) and paired PCHNTs (P) and

ED

probed sequentially with anti-DIMT1 and anti–β-actin antibodies. The same experiment was

PT

repeated at least three times. B-E, Summary of Western blotting results from TNM stage I–IV GCs and paired PCHNTs presented as relative band densities normalized to the β-actin of the

CE

same samples. F, Summary of Western blotting results from all 75 patients. The average DIMT1

AC

protein concentration in GC tissues was significantly higher than that in PCHNTs (*, **, ***, ****, and ***** P < .001).

Fig. 3 Kaplan-Meier survival curves of GC patients with different extents of DIMT1 expression stratified by TNM stage. A, Patients in stages I and II. B, Patients in stages III and IV. C, All GC patients.

18

ACCEPTED MANUSCRIPT Table 1 DIMT1 expression in gastric normal tissues, PCHNTs, and GC n

DIMT1 expression Positive rate − (%) + ++ +++ a Normal 12 10 1 1 0 16.7 PCHNTb 75 61 5 6 3 18.7 c GC 75 10 18 25 22 86.7 NOTE. Immunohistochemical staining intensity: −, no staining; +, weak staining; ++, moderate staining; +++; strong staining. Statistical analyses were performed by the Pearson 2 test. P < .05 was considered significant. Abbreviations: PCHNT, pericancerous histologically normal tissue; GC, gastric carcinoma. a Normal tissue vs PCHNT: P > .05 (2 = .528; P = .913). b PCHNT vs GC: P <. 05 (2 = 70.767; P = .000). c Normal tissue vs GC: P <. 05 (2 = 28.896; P = .00).

AC

CE

PT

ED

MA

NU

SC

RI P

T

Type of tissue

19

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA

NU

SC

RI P

T

Table 2 Relation between DIMT1 expression in GC tissues and clinicopathologic features Clinical parameters n DIMT1 expression Pa 2 − + ++ +++ Sex 6.637 .084 Male 48 3 12 19 14 Female 27 7 6 6 8 Age (y) 4.428 .219 <60 34 6 5 14 9 ≥60 41 4 13 11 13 Tumor size (cm) 1.417 .702 <3.0 43 5 12 15 11 ≥3.0 32 5 6 10 11 Differentiation Well 14 4 6 2 2 11.282 .080 Moderate 28 3 8 9 8 Poor 33 3 4 14 12 Tissue invasion 17.797 .038 T1 4 2 2 0 0 T2 18 2 3 8 5 T3 34 5 12 9 8 T4 19 1 1 8 9 Lymph node metastasis 14.541 0.002 Negative 24 5 11 6 2 Positive 51 5 7 19 20 Distant metastasis 19.625 .000 M0 60 10 18 21 11 M1 15 0 0 4 11 TNM stage 20.995 .013 I 12 2 4 4 2 II 19 4 6 7 2 III 29 4 8 10 7 IV 15 0 0 4 11 NOTE. Immunohistochemical staining intensity: −, no staining; +, weak staining; ++, moderate staining; +++; strong staining. a Statistical analyses were performed by the Pearson 2 test.

20

ACCEPTED MANUSCRIPT

RI P

95% CI .505, 8.163 .147, 1.872 .144, 2.125 1.151, 6.739 1.074, 37.788 2.533, 532.046 3.358, 1280.623 .005, .541

SC

Exp(B) 2.031 .560 .554 2.785 6.369 36.713 65.573 .053

AC

CE

PT

ED

MA

NU

Covariate Sex Age Tumor Size Differentiation Tissue invasion Lymph node metastasis Distant metastasis TNM stage

T

Table 3 Multivariate logistic regression results of DIMT1 protein expression in GC tissues

21

P .318 .320 .389 .023 .042 .008 .006 .013

ACCEPTED MANUSCRIPT

T

DIMT1 overexpression correlates with progression and prognosis in gastric carcinoma

RI P

Guangyi Liu MDa, Xudong Peng MDa, Yongqian Cai MMb, Anqi Cheng MMa, Lang Zha MDc,

SC

and Ziwei Wang, MDc*

MA

NU

Highlights:

DIMT1 is highly expressed in human gastric carcinoma tissues.



The extent of DIMT1 expression may indicate the prognosis of patients with the disease.



It suggested that DIMT1 could serve as a marker for the diagnosis of gastric carcinoma.

AC

CE

PT

ED



22