Clinic correlation and prognostic value of P4HB and GRP78 expression in gastric cancer

Abstract


Introduction
Gastric cancer (GC) is the fth common diagnosed cancer and the third most frequently cause of cancer death global wide, relating to over one million new cases and an estimated nearly zero point eight million deaths in 2018, along with the incidence rate is still markedly elevating in Eastern Asia [1].Since the best treatment period of most GC patients had already missed when they clinically diagnosed, the prognosis of GC remains poor [2].The grim fact that survival trends were little increased in most countries between 1995-1999 and 2010-2014 disappoints us, despite efforts to improve treatment strategies [3].Owing to huge biological heterogeneity, even with similar clinical features, the prognoses of GC would perform great differences [4,5].Therefore, precisely evaluate the prognosis of GC with the aid of better predictors is urgently needed.
The endoplasmic reticulum (ER) is responsible for calcium dynamic balance, lipid biogenesis and participate in folding of either secretory or membrane bound proteins.Once various stresses such as hypoxia break its homeostasis, the resulting abnormal accumulation of unfolded and misfolded proteins together constitute ER stress [6].To reconstruction ER homeostasis, a series of signaling pathways named unfolded protein response (UPR) subsequently activates, which makes up of three main branches: 1) PERK (pancreatic ER kinase-like ER kinase, also known as eIF2AK3) -eIF2α (eukaryotic translation initiation factor 2α ); 2) ATF6 (activating transcription factor-6); 3) IRE1 (inositol-requiring protein 1, also known as ERN1)-XBP1 (X-box binding protein 1) [6][7][8].Tumors trend to accompany with hypoxia, acidosis and nutrient deprivation, frequently induce ER stress, together with follow-up activation of UPR play core roles in tumorigenesis, proliferation and survival of cancer cells [6].Prolyl 4-hydroxylase, beta polypeptide (P4HB, also known as PDIA1) belongs to protein disul de isomerase (PDI) family, close connects with ER stress and UPR, as an enzymatic molecular chaperone to reconstruct misfolded proteins [9].P4HB is proved to be upregulated in a variety of cancers, such as lung cancer [10], diffuse glioma [9], colon cancer [11].As a result, P4HB has been reported by numerous researches to be a potential target for cancer therapy [12][13][14].But studies referred to GC were rather limited.Only one recent research showed P4HB positively correlated with hypoxia-associated biomarkers, overexpression of P4HB indicated poor prognosis in GC patients [15].
Glucose-regulated protein 78 (GRP78, also known as HSPA5 or BiP) is also an ER chaperone protein, will dissociate from three speci c sensors (PERK, ATF6 and IRE1) in response to ER stress, and subsequently actives UPR.6 [16].General overexpression of GRP78 in different cancers including GC has been widely accepted, also GRP78 could serve as a good indictor of poor prognosis and contribute a lot to therapy resistance [17,18].
Xia et al identi ed that P4HB through downregulating GRP78 promoted hepatocellular cancer tumorigenesis [19].However, in GC, single overexpression of the two ER chaperone proteins has been reported to represent poor prognosis, meaning P4HB may not negative correlate with GRP78.Due to rare relevant research, we aimed to investigate the correlation between P4HB and GRP78 expression and clinical features, as well as their prognostic implications in GC.Furthermore, we integrated the expression of P4HB and GRP78 with other prognostic variables to predict overall survival (OS) after surgery in GC patients.

Materials And Methods
Patients Amount to 150 GC patients who underwent gastrectomy from 2007 to 2013 at A liated Tumor Hospital of Nantong University (Nantong, China) were enrolled in our retrospective study.All patients con rmed histopathology diagnosis after gastrectomy, and none of them received preoperative chemotherapy or radiotherapy.Patients with incomplete follow-up data and histopathological diagnosed neuroendocrine or adenosquamous carcinoma were excluded from the cohort.The pathological tumor-node-metastasis (TNM) stage conformed to TNM Staging Manual (7th edition) from the AJCC (American Joint Committee on Cancer).Fluorouracil-based postoperative adjuvant chemotherapy were mainly performed in patients with advanced disease ( / / TNM stage).The clinicopathological parameters included were shown below: age, gender, Bormann type, tumor size, Lauren's classi cation, differentiation, histological type, depth of tumor invasion, lymph node metastasis, TNM stage, vessel invasion, perineural invasion and uorouracil-based postoperative adjuvant chemotherapy (sTable 1).The median follow-up time was 68.0 (range 2.5 -92.5) months.Overall survival (OS) was de ned as the period between the date of gastrectomy or the death or the last follow-up.If patients ended follow-up in keeping alive, data were censored.The study had achieved the approval from the Ethics Committee of A liated Tumor Hospital of Nantong University and the informed consent of all in cohort patients.

Immunohistochemistry (IHC)
The tissue microarrays (TMA) were constructed by duplicate 1.5 mm formalin-xed para n-embedded tissue cores of two different areas.We performed IHC according to the standard protocol for the immunostaining provided by a previous study [20].P4HB expression used a mouse anti-P4HB monoclonal antibody (1:100, Abcam, Cambridge, UK).While GRP78 IHC used a rabbit anti-GRP78 monoclonal antibody (1:100, Abcam, Cambridge, UK).The mi-quantitative H-score, ranged from 0 to 300, was derived from the staining intensities (0: negative, 1: weak staining, 2: moderate staining, 3: strong staining) and the proportion of stained (0-100%).The H-score was evaluated independently by two experienced pathologists who were blind to clinical information.The median H-score was used as the expression cutoff point (P4HB and GRP78 median H-score both = 200).Under the condition, for both proteins, H-scores less than 200 were considered to low expression, while H-scores on the opposite were considered to low expression, respectively.

Bioinformatics analyses
The Gene Expression Pro ling Interactive Analysis (GEPIA, http://gepia.cancer-pku.cn/index.html)[21], which is an interactive web server based on The Cancer Genome Atlas (TCGA), was used to nd the correlation of mRNAs.The mRNA expression data of totally 1168 cell lines, including 37 GC cell lines, was downloaded from the Broad Institute Cancer Cell Line Encyclopedia (CCLE, https://portals.broadinstitute.org/ccle)[22].Gene Ontology (GO) analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis were performed by R package "clusterPro ler".The correlation coe cients of genes visualized in KEGG pathway maps were performed by R package "pathview".

Statistical Analysis
SPSS version 25.0 (IBM, Armonk, NY) was applied to for the χ 2 test to explore the correlations among biomarker variables (P4HB and GRP78) and clinicopathological variables, the Spearman rank correlation to evaluate the correlation between H-scores of P4HB and GRP78, the univariate and multivariate analyses by the stepwise Cox proportional hazard regression model to estimate the crude hazard ratios (HRs) and 95% CI of HRs.MedCalc Software version 19.0.4 (MedCalc, Mariakerke, Belgium) was applied to compare two correlation coe cients.The R packages "rms", "survival", "foreign", "survivalROC" and "survminer" were used for ploting the nomogram, the calibration curves to evaluate the calibration ability of the nomogram, the ROC curves to assess the discrimination ability of the nomogram, Kaplan-Meier curves to estimate prognosis and calculate the Harrell's C-index also to evaluate the discrimination ability of the nomogram.Decision curve analysis (DCA) curves were plotted by R package "stdca", whose source le was from www.mskcc.org, to analyze the clinical usefulness of the nomogram.Two-sided P value less than 0.05 was considered statistically signi cant.

Bioinformatics analyses of P4HB and GRP78 expression in GC
We explored TCGA-STAD (stomach adenocarcinoma) data by GEPIA and analyzed of extracted GC cell lines data (amount to 37 lines) from CCLE successively.It turned out P4HB mRNA was also positively correlated with GRP78 in both GC tissue and cell databases (R = 0.61, P < 0.001, and R = 0.356, P = 0.031, respectively) (sFigure 1B and 1C), in accordance with the correlation of proteins.Then we performed GO and KEGG analysis of these two genes.The top 5 items of each GO term was shown in sFigure 1D.Only items involve with both two genes were taken into account.Finally, protein folding and response to ER stress were discovered in biological process (BP) term, ER chaperone complex, melanosome, pigment granule, ER-Golgi intermediate compartment and ER lumen were discovered in cellular component (CC) term, together with only cell adhesion molecule binding was discovered in molecular function (MF) term.
KEGG pathway analysis showed both two genes were enriched in the pathway "protein processing in ER" (sFigure 1E).As the occurrence and development of GC was most likely related to ER stress and UPR, we made further efforts to search correlation of P4HB or HSPA5 with genes within UPR, originated from the "protein processing in ER" KEGG pathway map (hsa04141), by explore TCGA-STAD cohort in GEPIA (sFigure 2-3).Except for P4HB and HSPA5 both positively correlated with typical genes within UPR from STAD (Table 2), correlation coe cients of most upstream genes were higher than downstream genes, as shown in sFigure 4.

Association of P4HB and GRP78 expression with OS in GC
We applied Kaplan-Meier survival analysis and log-rank test to compare OS according to P4HB and GRP78 expression.The results revealed that high expression of P4HB (Figure 2A) and GRP78 (Figure 2B) alone was related to an obvious shorter OS (P < 0.001 and P = 0.012, respectively).Further strati cation analysis showed that the prognosis was the best when P4HB and GRP78 were both low expressed, the worst prognosis was correlated with high expression of both P4HB and GRP78, while the prognosis was in between when one of P4HB and GRP78 was high expressed (Figure 2C, P = 0.002).When analyzing of co-expression of P4HB and GRP78, OS of both or one high expression was sharp lower than that of both low expression (Figure 2D, P = 0.002).
Since P4HB and GRP78 were both reported to be associated with chemotherapy, we also explored the prognosis of uorouracil-based postoperative adjuvant chemotherapy together with co-expression of P4HB and GRP78, which contained both variables might more precise.We observed the prognosis of group with postoperative adjuvant chemotherapy was better than the other group (Figure 2E, P = 0.011).High level of P4HB and/or GRP78 exhibited signi cant worse in the former group (Figure 2F, P = 0.007), but no prognostic signi cance in the latter (Figure 2G, P = 0.231).Once the group with postoperative adjuvant chemotherapy was further divided according to TNM stage, it exhibited no prognostic signi cance in TNM / stage (Figure 2H, P = 0.859), on the contrary, a signi cant worse OS in TNM / stage (Figure 2I, P = 0.004).This indicates that GC patients in TNM / stage may not bene t from postoperative adjuvant chemotherapy if P4HB and/or GRP78 high expressed.

Evaluate the nomogram and its clinical value with TNM stage
To evaluate the discrimination ability and clinical uesfulness of the nomogram, we compared it with widely clinical used TNM stage.From Figure 4A and 4B, we found areas under ROC (receiver operating characteristic) curve (AUROCs) of the 3-year and 5-year OS probability for nomogram model were 0.790 and 0.760, respectively, much better than that for TNM stage (0.684 and 0.671, respectively).Since DCA has superiorities over AUROC for estimating alternative prognostic strategies, we performed the 3-year and 5-year DCA curves individually for the nomogram and TNM stage.As shown in Figure 4C and 4D, compared to TNM stage, even at a very small threshold probability, the nomogram offered a high net bene t in 3-year and 5-year, respectively.

Discussion
ER is not only high in calcium-dependent molecular chaperones such as GRP78 for helping stabilize protein-folding intermediates, also has an oxidative environment for the formation of disulphide bonds with the aid of PDI and for the proper folding of proteins [23,24].Under non-stressed conditions, GRP78 combines with three UPR sensors (PERK, ATF6 and IRE1) keeping them inactive.Once ER stress occurs, GRP78 preferentially binds to the unfolded proteins, dissociates from the sensors, resulting in activation of the UPR transducers [6,25].The outcome of it involves in protein folding and transporting, as well as increasing unfolded protein clearance through pathways such as ERassociated degradation (ERAD) and autophagy.Upon unresolved stress, cells enter apoptosis [8].
PDI family is a part of the thioredoxin (TRX) superfamily, mainly act as catalysts for the formatting and rearranging S-S bonds, also can function as chaperones involved in ERAD [26], playing an important role in the maintenance and regulation of proteostasis.P4HB, also known as PDIA1 (usually referred as PDI), as the archetype PDI protein, the chaperone activity of who is independent of catalytic activity, while the former is vital for many client proteins [27].There is su cient evidence supporting that PDIA1 high expressed in a wide variety of cancers tissues [9][10][11]24] and cancer cells [28].And the upregulated PDIA1 is closely correlated with cancer metastasis and invasiveness [27].
GRP78 is encoded by HSPA5, as the most abundant protein of heat shock protein-70 (Hsp70) family, mainly exits in ER lumen, corrects the folding and assembly, as well as prevents the transport of misfolded proteins or protein subunits [29,30].GRP78 works as a molecular chaperone in cells, its high expression is crucial for proliferation, invasion, metastasis of many cancers, proves to be correlated with tumor resistance, recurrence, and a low OS [17,18,[31][32][33].With the help of GRP78, tumor cells can organize the stimulation of processes including macroautophagy, combat reactive oxygen species (ROS), and activate pro-survival signaling pathways [29].Thus, GRP78 will be a useful prognostic biomarker and a promising target for cancer.
P4HB and GRP78 are both ER chaperone proteins involved in UPR, and represent for poor prognosis of various cancers from the above mentioned.However, rare research pays attention to the correlation of them, especially in cancer.An early study shows that BiP and PDI perform synergistically in vitro in the oxidative folding of antibodies, BiP bind to the unfolded polypeptide chains to keep them in a conformation where cysteine residues are accessible for PDI [34].Different from P4HB negatively correlated with GRP78 in hepatocellular cancer in vitro [19], we con rmed that P4HB and GRP78 showed a positive correlation in protein (from our cohort) and mRNA (from the databases) levels, with disease progress, the two proteins might more closely related since the correlation coe cients were signi cantly elevated in tumor size ≥ 5 cm, TNM / stage and positive vessel invasion subgroups.Together with single or combined high expression of P4HB and GRP78 represented for poor prognosis in GC.The speci c mechanisms need further exploration.
A lot of studies have shown in a wide range of primary tumors, there exists a sustained and high-level activation of all three branches within UPR [35].From bioinformatics analyses, we also found P4HB and GRP78 might participate in UPR, since they positively correlated with all typical genes within UPR, as well as correlation coe cients of most upstream genes were higher than downstream genes.UPR may act as a key driver in resistance to chemotherapy [36,37].A new encouraging research nds that activation of the PERK branch with UPR is required for colon cancer cells surviving from treatment of 5-uorouracil, the usage of PERK inhibitor synergizes with 5-FU could suppress the growth of colon cancer cells in vivo [38].
It may explain in uorouracil-based postoperative adjuvant chemotherapy subgroup, GC patients with a high co-expression of P4HB and GRP78 had a shorter OS of our cohort.When strati ed by TNM stage, only patients in TNM / stage had a signi cant shorter OS, one reason might be that in early stage GC, the positive effects from chemotherapy could cancel out the negative effects from UPR, while in advanced GC, the negative effects probably exceeded.
Our nomogram integrated co-expression of P4HB and GRP78 with other prognostic variables, was a simple visualized tool for predicting OS after surgery of patients with OS.Comparing to TNM stage, the discrimination ability and clinical usefulness of the nomogram was much better through AUROCs and DCA curves of 3-year and 5-year, may be great valuable for clinic.
There still remain several limitations in our study.Firstly, this is a retrospective study of small sample group of GC patients, exists the statistical limitations.Secondly, further researches are required to con rm the correlation and regulation mechanisms of P4HB and GRP78 expression of GC in vitro and in vivo.
Finally, it needs an independent external cohort to verify our research ndings.
In conclusion, we revealed that P4HB expression is positively correlated with GRP78

Figure 1 Representative
Figure 1

Figure 3 Nomogram
Figure 3 expression, the correlationship may closer in advanced GC; P4HB and GRP78 may involve in UPR of GC from bioinformatics analyses; high co-expression of P4HB and GRP78 predicts a shorter OS with uorouracilbased postoperative adjuvant chemotherapy in the advanced stage; co-expression of P4HB and GRP78 can independently predict unfavorable OS for GC patients; the prognostic nomogram based on stepwise Cox regression model is much better than TNM stage in the discrimination ability and clinical usefulness.