Overexpression of JAG2 is related to poor outcomes in oral squamous cell carcinoma

Abstract Objectives JAG2 is one of Notch ligands, which recently appear to exert various carcinogenesis. In the present study, we aimed to unveil the relation of JAG2 expression and clinicopathological features in oral squamous cell carcinoma (OSCC). Materials and Methods We examined JAG2 expression in OSCC plus adjacent nontumorous epithelia in eight patients. Ninety‐one OSCC tissue specimens were immunohistochemically stained with specific antibodies to JAG2. The immunoreactivities of JAG2 were correlated with clinicopathological factors, including the prognosis of patients. Chi‐square test, Kaplan–Meier survival, and Cox proportional hazard analysis were used to determine the statistical value of JAG2 expression in OSCC. Results JAG2 mRNA expression was much expressed in OSCC tissues compared with adjacent tissue specimens in five of eight patients. JAG2 immunoreactivity was found at invasion front in 31 of 91 OSCC. JAG2 immunoreactivity was significantly associated with age, less than 50 years old of patients (P = .048). Kaplan–Meier analysis demonstrated that the patients with JAG2 immunoreactvty have a short overall survival. With the Cox proportional hazard regression mode, the independent factors predictive of poor overall survival included JAG2 immunoreactivity (P < .05). Conclusions The present findings suggest that JAG2 overexpression, especially at the cancer invasion front, has potential prognostic value.

pathobiological mechanism of OSCC progression to develop new therapies.
Notch signaling pathway appeared to have diverse roles, oncogenic or tumor suppressor, in various carcinogenesis (Aster, Pear, & Blacklow, 2017;Yap et al., 2015). Activation of Notch pathway requires the binding of Notch receptors, designated Notch-1 through Notch-4, to surface membrane ligands, that is, DLL1, DLL3, DLL4, JAG1, and JAG2 in human (Kopan & Ilagan, 2009). Despite of recent progress, it is still not fully understood how these ligands interact with various Notch receptors in carcinogenesis. Furthermore, liganddependent Notch activation required mono-ubiquitination of cytoplasmic domain of ligands, followed by internalization of ligands to cytoplasm (Itoh et al, 2003;Takeuchi et al, 2005). Indeed, cytoplasmic localization of Notch ligands is frequently found in various cancer cells (He et al., 2016;Townsend et al., 2019;Zhang et al., 2014a).
Only a few studies have examined JAG2 expression in OSCC (Osathanon et al., 2016a;Zhang et al., 2011). To the best of our knowledge, the prognostic value or the relationship between JAG2 expression and the clinicopathological factors remains unknown. To know the potential value of JAG2 as a molecular target for patients with OSCC, it is important to reveal the clinicopathological features of JAG2 expression in OSCC.
The aim of this study was to find the association between JAG2 expression and prognosis and clinicopathological factors of OSCC.
Regarding OSCC, there are still a few studies that examined the JAG2 expression (Osathanon et al., 2016a;Zhang et al., 2011). A few studies have suggested that JAG2 is also overexpressed in OSCC (Osathanon et al., 2016a;Zhang et al., 2011). However, to the best of our knowledge, the prognostic value or relation of JAG2 expression to clinicopathological factors remains unknown. To know the potential value of JAG2 as a molecular target for patients with OSCC, it is important to unravel the clinicopathological features of JAG2 expression in OSCC.

| Patient tissue specimens
After informed consent was obtained, surgically resected OSCC specimens and surrounding noncancerous tissues were stored frozen at −80 C for RNA extraction. In a retrospective study, we collected data on all surgically treated patients who were primarily diagnosed with OSCC. Archived pathological tissue specimens from 91 cases of OSCC were used in this study. Information of the patients with OSCC is shown. The present study was conducted in accordance with the ethi- The following primers were used for PCR: JAG2, forward reverse 5 0 -GAAGATGGTGATGGGATTTC-3 0 . The expression of each target gene was analyzed by the 2 -ΔΔCt method. (Livak & Schmittgen, 2001) using the LightCycler system. ΔCT values of genes were normalized to that of GAPDH for each triplicate set. The expression ratio of tumor tissue to nontumorous tissue, T/N, was calculated.

| Immunohistochemical staining
All tissue specimens were obtained surgically, fixed in 10% buffered formalin, and embedded in paraffin. A tissue section, which represent the deepest invasion site of each case was immunohistochemically stained. The rabbit antibody to JAG2 (cat no. NBP1-86337, 1:100) was purchased from Novus Biologicals (Littleton, CO). The tissues were immunostained with antibodies using the ImmPRESS™ polymerized reporter enzyme staining system (Vector Laboratories, Inc., Burlingame, CA, USA) as previously reported (Takeuchi et al., 2000).

| Evaluation of immunohistochemical staining and statistical analysis
We evaluated the immunohistochemical staining results as a percentage of immunoreactivity in OSCC cells. The fraction of positive cells stained at the cancer invasion front was scored after examining three high-power fields (400×) per tissue section for each case. The staining was considered as negative if less than 10% of invasive cancer cells exhibited immunoreactivity and as positive if over 10% did.
Curves for overall survival (OS) were drawn using the Kaplan-Meier method, and differences in the survival rates were compared using the log-rank test. The relationship between the clinicopathological parameter was examined using the chi-square test.
Multivariate analyses of Cox proportional hazard model were also used to determine the significant prognostic factors of OS. A P value of <.05 was considered statistically significant.

| JAG2 mRNA expression in OSCC and neighboring tissues
Results are summarized in Figure 1. Notably, JAG2 mRNA was much expressed in OSCC tissues compared with the neighboring noncancerous tissues of five of eight patients. Notably, OSCC from two of two patients of relatively young age (patient #8 was early thirties, and patient #4 was early forties) exhibited robust JAG2 mRNA compared with noncancerous tissues.
F I G U R E 1 Relative JAG2 mRNA expression ratio between tumor (T) and nontumorous (N) tissue specimens. JAG2 mRNA level in OSCC (T) and paired nontumor tissues (N) measured by quantitative RT-PCR. The ratio of concentration of JAG2 mRNA of tumor/nontumor (T/N) is shown. Expression of JAG2 mRNA in paired tumors and nontumorous tissue specimens was analyzed by triplicate and normalized to GAPDH expression. Note the specimens from patients #2, #3, #4, #7, and #8 in which JAG2 mRNA was much expressed in tumor than nontumorous tissue (T/N ratio was over 1; mean and standard deviation are shown). Patients #4 and #8 were early forties and thirties, respectively

| JAG2 expressions in OSCC tissue specimens in relation to clinicopathological parameter
Clinicopathological characteristics of patients were summarized in Table 1. Representative results from the immunohistochemical staining are shown in Figure 2. Little immunoreactivities were found in nontumorous oral mucosa epithelial cells (Figure 2b). By contrast, JAG2 immunoreactivity was found in 31 of the 91 invasive cancer cell samples at the invasion front (Figure 2c).
JAG2 expression was significantly related to age of patents, younger than 50 years old. By contrast, no significant correlation was observed between JAG2 immunoreactivity and other clinicopathological factors, that is, gender, histology, tumor size, status of nodal metastasis, and stages of patients (Table 2).
Kaplan-Meier method followed by the log-rank test revealed that JAG2 immunoreactivity was correlated to poor OS of patients with OSCC (P = .0221; Figure 2d). Using multivariate analysis, JAG2 immunoreactivity appeared to be an independent marker of poor OS (P = .009, Cox proportional hazard regression model; Table 3).

| DISCUSSION
An earlier study demonstrated that JAG2 expression was highly correlated in tongue carcinoma tissues (Zhang et al., 2011  expression is associated with a poor prognosis in patients with hepatocellular carcinoma (Zhang et al., 2014b) and urinary bladder cancer (Li et al., 2013).
The present result indicated that JAG2 was expressed in cancer invasion front of 31 of 91 OSCC tissue specimens. To our knowledge, this is the first study that unveiled that JAG2 expression was related to poor OS of patients with OSCC. Notably, JAG2 expression was significantly related to age, younger than 50 years old, in our OSCC tissue specimen series. OSCC primarily occurs in older age group.
However, in the recent years, incidence of oral cancer in young people has been on rise worldwide (Abdulla et al., 2018). cytoplasm in a part of OSCC at cancer invasion front, might be linked to the novel molecular mechanism, which was responsible for oral carcinogenesis at relative young age. However, it needs further extensive studies to unveil the molecular and pathobiological mechanism of JAG2-mediating oral carcinogenesis.
In conclusion, the present findings indicate that one of Notch ligands, JAG2, was significantly overexpressed at invasion front of OSCC in one third of specimens with relation to poor prognosis.

ACKNOWLEDGMENT
We would like to thank our colleague, Ms. Reiko Shinoda, for providing tissue specimens.