CALM1 Promotes ESCC Progression and Dampens Chemosensitivity to EGFR Inhibitor

Background: Calmodulin1 (CALM1) has been identied as one of the overexpression genes in a variety of cancers and EGFR inhibitor have been widely used in clinical treatment but it is unknown whether CALM1 and epidermal growth factor receptor (EGFR) have a synergistic effect in esophageal squamous cell carcinoma (ESCC). The aim of the present study was to explore the synergistic effects of knock-out CALM1 combined with EGFR inhibitor (Afatinib) and to elucidate the role of CALM1 in sensitizing the resistance to Afatinib in ESCC. Method: Immunohistochemistry (IHC) and qRT-PCR were used to examine the expression of CALM1 and EGFR in ESCC tissues. Kaplan–Meier survival analysis was used to analyze the clinical and prognostic signicance of CALM1 and EGFR expression in ESCC. Furthermore, to evaluate the biological function of CALM1 in ESCC, the latest gene editing technique CRISPR/Cas9 (cid:0) Clustered regularly interspaced short palindromic repeats (cid:0) was applied to knockout CALM1 in ESCC cell lines KYSE150, Eca109 and TE-1. MTT, ow cytometry, Transwell migration, scratch wound-healing and colony formation assays were performed to assay the combined effect of knock-out CALM1 and EGFR inhibitor on ESCC cell proliferation and migration.. In addition, nude mice xenograft model was used to observe the synergistic inhibition of knock-out CALM1 and Afatinib. Results: Both CALM1 and EGFR were found to be signicantly over-expressed in ESCC compared with paired normal control. Over- expressed CALM1 and EGFR were signicantly associated with clinical stage, T classication and poor overall prognosis, respectively. In vitro, the combined effect of knock-out of CALM1 mediated by the lentivirus and EGFR inhibitor was shown to be capable of inhibiting the proliferation, inducing cell cycle arrest at G1/S stage and increasing apoptosis of KYSE-150 and Eca109 cells; invasion and migration were also suppressed. In vivo, the results of tumor weight and total uorescence were markedly reduced compared with the sgCtrl-infected group and sgCAML1 group. Conclusion: Our data demonstrated that knock-out of CALM1 could sensitize ESCC cells to EGFR inhibitor, and it may exert oncogenic role via promotion of EMT. Taken together, CALM1 may be a tempting target to overcome Afatinib resistance.


Introduction
Esophageal cancer (EC) is one of the most common gastrointestinal tract malignancies and ranks as the sixth most important cause of cancer mortalities globally, with an incidence of estimated 509,000 new deaths every year [1,2]. As one of the most common pathohistological subtypes of ESCC, ESCC usually composes over 90% of all EC cases in areas of Asia and Sub-Saharan Africa [1,3]. Despite the advances in diagnosis and treatment, ESCC still carries a poor prognosis [4],and the 5-year overall survival rate ranges from 15-25% [5].
CALM is a ubiquitous calcium ion (Ca 2+ ) receptor protein, mediating a large number of signaling processes; it is highly conserved from an evolutionary standpoint [6][7][8][9][10]. CALM with a sequence of 148 amino acids is present in all eukaryotic cells [11]. In humans, CALM is encoded by three different genes (CALM1, CALM2, CALM3), each of which has unique selective regulation, tissue speci city, and alternative splicing, but surprisingly, they all produce same protein [12,13]. However, although speci c cells can express these three genes, they do not necessarily all have the same functional roles because the three transcripts can be differentially processed by post-transcriptional regulation or subcellular distribution [14]. In this study we focus on CALM1. CALM is composed of Ca 2+ -binding EF-hands, and participates in signaling pathways that modulates proliferation, motility and differentiation [15]. Several studies found that the expression level of CALM1 was markedly associates with many kinds of cancer, including bladder cancer [16], prostate cancer [17] and nasopharyngeal carcinoma [18]. As far as CALM is concerned, numerous investigations have been carried out on mechanistic aspects, mainly in the cell proliferation, programmed cell death and autophagy. CALM/Ca 2+ binding to the SH2 domains of the p85 subunit of PI3Kα stimulates PI3Kα/Akt/mTOR signaling, and thereby regulating cell proliferation and growth [19,20]. CALM also regulated EGFR's tyrosine kinase activity [21] which activates Ras and PI3Kα and has essential roles in programmed cell death and autophagy [19]. However, the biological function of CALM1 and its regulatory mechanism in ESCC are rarely studied.
The EGFR gene encodes a membrane glycoprotein responsible for the upregulation of EGFR signaling. The success of EGFR tyrosine kinase inhibitor (TKIs) have provided a powerful validation for precision cancer medicine because the over-expression and mutations on EGFR plays an important carcinogenic role in a variety of solid tumors such as head and neck, breast, lung, and colorectal cancer, and numerous EGFR inhibitor have been widely used in clinical treatment [22][23][24][25]. Fumiyuki Sato et al [26] reported that EGFR inhibitor prevent induction of cancer stem like cells in ESCC by suppressing EMT. In view of these previous ndings, we hypothesized that CALM1 and EGFR may play a synergistic role in the development of ESCC. However, up to now, the relationship of CALM1 and EGFR in the progression of ESCC remains unknown. Herein, we undertake the study to present our results of characterization of CALM1 and EGFR and to analyze its clinical relevance in ESCC.

Cell culture
Two human ESCC cell lines, KYSE150 and TE-1, were obtained from the Chinese Academy of Sciences (Shanghai, China), and Eca109 cells were from Wuhan University(Wuhan, China).The three cell lines were cultured in RPMI-1640 (Invitrogen, Life Technologies, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum (Gibco, Carlsbad, CA, USA) and 1% penicillin-streptomycin (Gibco;Thermo Fisher Scientifc, Inc.). All the ESCC cell lines were cultured in a 5% CO 2 humidi ed incubator at 37 °C.

Tissue microarray
Tissue microarrays of clinical samples consisted of ESCC and paired normal adjacent tissues(NAT) .One tissue microarray included 34 paired cases of ESCC and matched NAT (catalog number: # HEsoS180Su08; Outdo Biotech, Shanghai, China), and another 50 additional independent, subjected to esophagectomy, obtained from the First A liated Hospital of Xinjiang Medical University. Tumor tissues and clinical data were collected after obtaining the relevant informed consent from each patient involved.
The study involving human tissue samples were approved by Medical Ethics Committee of the First A liated Hospital of Xinjiang Medical University.

Immunohistochemistry (IHC)
Tissue microarrays were de-waxed and hydrated, boiled in 0.01 M citrate buffer, and treated with 3% hydrogen peroxide after natural cooling. The primary polyclonal rabbit anti CALM1(1:400; Proteintech, Wuhan, China) and EGFR (1:600; Proteintech) was incubated overnight in 4 °C by adding drop of glass slide, followed by treatment with biotinylated antirabbit secondary antibody (CST) for 60 min at 37 °C.
The sections were evaluated by two pathologists under optical microscopy and cell localization of protein and immunostaining levels was assessed in each section. The intensity of staining was divided into four grades (0, none; 1,weak; 2, moderate; and 3, strong) and percentage of positive cells (0, < 10%; 1, 10%-25%; 2, 25%-50%; and 3, > 50%). According to the total score (staining intensity plus positive cell score), ESCC patients were divided into two groups, speci cally "low expression "(total score ,0-3) and" high expression "(total score ,4-6), which were used to analyze the prognostic signi cance of EGFR and CALM1 in ESCC.

Quantitative real-time polymerase chain reaction (qRT-PCR)
Total RNA was extracted with TRIzol reagent and then the RNA was reversely transcribed into cDNA using a Pria Revert Aid First Strand cDNA Synthesis Kit. Following the manufacturer's protocols, Real-time PCR was performed using a SYBR Green Premix PCR Master Mix. Relative mRNA expression of CALM1 and EGFR was calculated using the 2 −ΔΔCt method after being normalized to GAPDH. PCR was performed with the following primer sets: CALM1 forward, 5'GGTCAGAACCCAACAGAA3' and reverse, 5'AGACTCGGAATGCCTCA3'; and EGFR forward, 5'AGGCACGAGTAACAAGCTCAC3' and reverse, 5'ATGAGGACATAACCAGCCACC3'. GAPDH forward, 5'TGACTTCAACAGCGACACCCA3' and reverse, 5'CACCCTGTTGCTGTAGCCAAA3'.

Western blots
Cells were lysed with RIPA lysis buffer after CALM1-guide RNA transfection for 72 h, and protein concentration was detected with bicinchoninic acid protein assay (Thermo Fisher Scienti c)., 0.1 mg total protein were subjected to 10% SDS-PAGE separation under denaturing conditions and then transblotted to PVDF membranes (Millipore, Billerica, MA,USA). Target proteins were detected by using speci c antibody against CALM1 (1:800; Proteintech Group, Wuhan, China). GAPDH (1:500; Santa Cruz Biotechnology Inc,USA) was chosen as an internal control and the CALM1 and GAPDH dilutions were incubated at 4 °C with gentle shaking overnight. The blots were visualized with Pierce™ ECL Western Blotting Substrate (Thermo Fisher Scienti c, CA, USA), according to the manufacturer's protocol.

MTT assay
Cells were placed into the 96-well plates at the density of 4 × 10 3 /mL in RPMI-1640. At the designated time points, the cells were coated with 100 µL sterile MTT (Sigma-Aldrich) in an incubator with 5% CO 2 for 4 h at 37 °C. Afatinib was added with the desired drug treatment concentrations ranging from 0 to 20 µM and incubated for 72 h. The reaction waster was performed by removing the culture medium and then adding 100 µL of dimethyl sulfoxide (Sigma-Aldrich) for 0.5 h to dissolve the formaldehyde. Finally, absorbance values were measured at 490 nm. The IC50 (half-maximum inhibitory concentration) was used as the measure of relative cytotoxicity.

Apoptosis assay and cell cycle
After transfection with sgCtrl and sgCAML1-1 with or without EGFR inhibitor for 72 h, ESCC were collected after washing twice with PBS. For cell cycle, cells were xed in 70% ice-cold ethanol overnight, then washed twice with PBS and stained with 10 µg/mL RNase A in the dark for 15 min at room temperature. The analysis was performed by a owcytometer (BD FACS Calibur; BD Biosciences, Brea, CA, USA). For analysis of apoptotic cells, it was analyzed by ow cytometry using an FITC Annexin V Apoptosis Detection Kit (Thermo Fisher Scienti c) according to the manufacturer's instructions after harvesting cells.
Cells at a density of 1 × 10 5 cells per well were placed into the upper chambers in 600 µL serum-free RPMI 1640. After incubating at 37 °C with 5% CO 2 for 24 h, cells invading the lower surface of the lter membrane were scraped off with swabs; the number of invaded cells was counted using Image J software (NIH, Bethesda, MA, USA).

Wound healing assay
The migratory variation of ESCC cells was determined by Wound healing assay. The extent of cell motility was quanti ed by measuring the distance area between migrating cell boundaries. At 0 and 24 h, Photographs were captured by using amicroscope at × 40 magni cations. At least, four wound areas were photographed on each plate and counted under an Olympus inverted uorescence phase-contrast microscope (Tokyo, Japan) Colony formation assay KYSE150 and Eca109 cells transfected with sgCtrl, sgCALM1-1 were plated in 6well plates (1000 cells/well) and incubated at 37 °C for 14 days to allow colony formation. In the drug treatment group, the medium was changed with fresh medium containing Afatinib or vehicle (DMSO) every 2 days. The cell medium was subsequently removed. Cells were washed using PBS and xed with 4% paraformaldehyde for 10 min at room temperature. The cells were stained with crystal violet kit (Beyotime Institute of Biotechnology, Shanghai, China) for 15 min at room temperature. The colonies were washed, photographed by camera and counted using ImageJ software.

Tumorigenesis in nude mice and in vivo imaging
Nude mice (4 weeks old) were purchased from Shanghai Lingchang Biological Technology Co., Ltd. All animals (22 ± 1.5 g) were handled according to the Guide for the Care and Use of Laboratory Animals and were housed at a controlled temperature (22-28˚C) and humidity (50%) under a 12-h light/dark cycle. All mice were randomly divided into three groups: sgCtrl group, sgCAML1-1 group, sgCAML1-1 plus EGFR inhibitor group. Then, the stably cells (4 × 10 6 for each side) were suspended in PBS and implanted subcutaneously into male BALB/c nude mice. Animals in the sgCAML1-1plus EGFR inhibitor group treated with 20 mg/kg paclitaxel every 3 days once when tumor size reached about 100 mm 3 intraperitoneally (i.p.). After 7 days, the tumor weight was measured every 3 days for 4 weeks. After 27 days of monitoring, in vivo imaging of animals before they are sacri ced and the tumors were dissected and weighted.

Statistical analysis
Data were expressed as the mean ± standard deviation (SD), using SPSS for Windows version 19.0 (SPSS, Inc., Chicago, IL, USA). Student t test or one-way analysis of variance (ANOVA) was used to evaluate the differences among groups, and chi-square and Fisher's exact tests were applied to analyze correlation between CALM1/EGFR expression and clinicopathological characteristics. The Kaplan-Meier survival curve and log rank test were used to plot the survival curves and estimate survival rates. A twotailed P < 0.05 was taken as signi cant in all tests.

Results
High CALM1 and EGFR expression were signi cantly associated with metastasis and poor prognosis in ESCC To understand the pathological signi cance of CALM1 and EGFR, we rst detected the expression levels of CALM1 and EGFR in 84 para n-embedded human ESCC and paired NAT tissue blocks, by immunohistochemistry (IHC). Apparently, IHC analysis revealed that various ESCC tissues show higher expression of CALM1 and EGFR in ESCC tissues compared to the NAT tissues (Fig. 1A, Table 1). The relationship between the levels of these two proteins in the ESCC tissue and the clinicopathological parameters of the 84 ESCC patients was analyzed in Table 1. Expression of CALM1 and EGFR was not found to be correlated with gender, age, Tumor diameter (cm) but closely related to clinical stage and T classi cation (Table 1). Kaplan-Meier survival analyses revealed a signi cantly shorter overall survival time for patients with high CALM1 and EGFR expression relative to patients with low CALM1 and EGFR expression (Fig. 1C). Notably, the CALM1 expression was positively correlated with EGFR in clinical tissues of ESCC (Fig. 1B, Table 2).
Knockout of CALM1 and treated with EGFR inhibitor markedly impaired the proliferation, cell cycle and increased apoptosis of ESCC cells.
Having understood the clinicopathological signi cance of the CALM1 and EGFR in vivo in ESCC, therefore we hypothesized that knockout of CALM1 and treatment with EGFR inhibitor (Afatinib) could markedly impair the proliferation and apoptosis in vitro in ESCC cell lines. To test the hypothesis, rstly, the basal level of CALM1 and EGFR on mRNA was evaluated using qRT-PCR, in a panel of human ESCC cell lines-KYSE150, Eca109 and TE-1. Results showed these three ESCC cell lines, the basal level of CALM1 and EGFR was higher in KYSE150 and Eca109 cell lines than that in TE-1 cells lines ( Fig. 2A). On the basis, the two extreme cases were selected as cell model to further investigate the synergistic reaction of CALM1 and EGFR in ESCC cells. qRT-PCR data showed that sgCALM1-1 successfully achieved signi cant depletion of CALM1 in two cell line (Fig. 2B). To investigate the effect that synergistic reaction of CALM1 and EGFR exerted over proliferation and apoptotic variation of ESCC cells, we carried out MTT assay and ow cytometry after KYSE150 and Eca109 cell lines were transfected with lentiviral-based knockdown of CALM1. The IC50 value of Afatinib for KYSE150 and Eca109 cells were 8.80 µM and 4.01 µM, respectively (Supplementary: Fig S1). It was exhibited that depletion of CALM1 moderately inhibited the proliferation (Fig. 2C) and increased apoptotic (Fig. 2D). More important, treatment with Afatinib can markedly slow down the proliferation (Fig. 2C) and increase apoptotic (Fig. 2D), compared with control and sgCALM1-1 group, strongly suggesting the tumor-promoting role of CALM1 and EGFR in ESCC cells. Cell cycle analysis revealed that KYSE150 and Eca109 cells with CALM1 knockout arrested in G1 and S phase after EGFR inhibitors treatment than the sgCtrl group and sgCALM1-1 group (Fig. 2E). These ndings indicated that reducing CALM1 and EGFR expression inhibits the G1/S phase transition.

Knockout of CALM1 and treated with EGFR inhibitors markedly inhibited the invasion and migration of ESCC via EMT
Next, to investigate the combined effect of knock-out of CALM1 and EGFR inhibitor that exerted over proliferation of ESCC cells, we carried out Transwell assay and Wound-healing assays after KYSE150 and Eca109 were transfected with lentiviral-based knockout of CALM1. As shown in Fig. 3A-B, The results showed that compared with the control, the cell proliferation of ESCC signi cantly suppressed in the sgCALM1-1 group compared with control group cells; however, a stronger increase was observed in the sgCALM1-1 group plus EGFR inhibitors group. In clonogenic assay, we also found that silencing of CALM1 in combination with Afatinib caused a marked inhibition of proliferation in two cell lines, which is consistent with our previous results (Fig. 3C). We further focused on the mechanisms underlying CALM1 and EGFR activity in ESCC by examining the levels of FN1, the markers of EMT that are critical in cell invasion and migration. Notably, the CALM1 and EGFR expression was negatively correlated with that of FN1 in ESCC (Fig. 3D). Based on the collective results, we suggest that CALM1 and EGFR contribute to tumor cell migration and invasion through promoting EMT.
Knockout of CALM1 and treated with EGFR inhibitors markedly impaired the tumorigenesis in nude mice in vivo.
To con rm the results of knockout of CALM1 and application of EGFR inhibitor in vivo mouse tumorigenesis model, where mice were injected with KYSE150 cells from the sgCtrl or sgCALM1-1 groups with vehicle (saline) or Afatinib, was generated. The results of tumor weight analysis revealed that sgCALM1-1 cells with Afatinib generated markedly smaller subcutaneous xenograft tumors in nude mice compared with NC and sgCALM1-1 cells group (Fig. 4A-C). In addition, in order to further con rm that knockout of CALM1 and treatment of EGFR inhibitor was directly associated with the observed effects on tumor growth, a uorescence imaging test was also conducted using a small animal live imaging system, which monitors the uorescence signals emitted from tissues. The sgCtrl-infected and sgCAML1-infected KYSE150 cells were also transduced with GFP; therefore, tumor xenografts in three groups emit uorescence signals when triggered by speci c uorescence in the live imaging system in vivo. The uorescence imaging results demonstrated that the total radiant e ciency of mice in the sgCAML1infected group with treatment of EGFR inhibitor was markedly reduced compared with in the sgCtrl-infected group and sgCAML1 group ( Fig. 4D-E). These results clearly demonstrated that inhibition of CAML1 increased Afatinib sensitivity in vivo.

Discussion
In the present investigation, we found that CALM1 and EGFR were remarkably up-regulated in ESCC,compared with paired NAT and that over-expression of CALM1 and EGFR in ESCC was signi cantly associated with tumor progression and poor overall prognosis. Furthermore, to functionally analyze the role of CALM1 in ESCC cell lines in vitro, KYSE150 and Eca109 cells were employed, whose endogenous CALM1 was down-regulated, respectively, by using lentiviral-based transfection. The combined effect of knock-out of CALM1 mediated by the lentivirus and EGFR inhibitors was shown to be capable of inhibiting the proliferation, cell cycle and increasing apoptosis of KYSE150 and Eca109 cells in vitro; invasion and migration were also depressed and enhancing epithelial-mesenchymal transition (EMT). In addition, to investigate the synergistic effect of CALM1 and EGFR plays in cell proliferation, nude mice were xenografted with ESCC cells whose CALM1 was stably knowdown in vivo.
While extensive research has shown that synergy between CALM and EGFR promotes gene transcription and cell proliferation in different cancer types, including human breast cancer, lung cancer, and astrocytic gliomas [27][28][29] but there are rare data in regard to its role in ESCC, especially CALM1. Kobayashi H et al. found that only CALM 1 played a role in the migration of mouse precerebellar neurons (PCNs) in vivo, while Calm2 Calm 3 genes did not functionally replace CALM1. When the CALM1 is knocked down with the shRNA, the radial and tangential migration of the cells is inhibited, and the nal goal failded to reach during the development, but there is no harmful effect after knocking down the CALM2 CALM3 [30]. San Jose, E et al [31] obtained the rst experimental evidence for CALM binding to the EGFR in a Ca 2+dependent manner in rat liver. In addition, the occurrence of CALM/EGFR complexes in living cells was established and the possible functional effects of this interaction on ligand-dependent activation were identi ed [32][33][34]. Based on these studies, it has rstly con rmed the expression of CALM1 and EGFR using IHC with ESCC tissue array. CALM1 and EGFR was upregulated in ESCC relative to NAT, and signi cantly correlated with poor overall prognosis, in the present study. By expanding the quantities of samples, further results were obtained showing that CALM1 and EGFR-positive staining is positively correlated with tumor progression and poor overall prognosis. Unlike CALM1 that has been seldom reported in the setting of ESCC, studies of EGFR in tumor are relatively extensive.
Although the overexpression or mutation of EGFR levels has proven to be a valid predictor of treatment outcome, the response rates in selected patients remain chemoresistance or poor prognosis in squamous cell carcinomas as well as other malignancies [35][36][37]. Therefore, future research should focus on exploring more biomarkers to optimize the therapeutic effect on EGFR inhibitors. CALM inhibitors plays an essential role in cell proliferation and/or reverse multiple drug resistance tendencies in many tumor cells [38,39], so it has been thought it has potential therapeutic effects in cancer [40,41]. Based on many studies which mostly performed in vitro point to a potential bene t of treating cancers with CALM antagonists. A, V et al [27] found that the site(s) of action of CALM in speci c CALM-dependent systems that are upregulated in tumor cells interacting with EGFR. Ca 2+ -CALM binding to the CALM binding domain (CALM-BD) of cytosolic juxtamembrane region of the receptor plays an important trigger role in ligand-dependent activation EGFR in living cells [42][43][44]. A further study was shown that nonphosphorylated CaM only interacts with the EGFR when is not phosphorylated at Tyr 1173 (tyrosine 1173 ) [45].Herein, We found here for the rst time that treatment with knock-out of CALM1 and EGFR inhibitors have signi cant effects against tumors in vivo and in vitro in ESCC. Recent studies have pointed to the potential for combinations of EGFR inhibitors with TKIs to overcome a certain degree of resistance for EGFR mutations [46]. However, such strategies may be limited for special resistant population. In the current observation, knock-out of CALM1 mediated by the lentivirus turns out to be able to slow down the growth and motility of KYSE150 and Eca109 cells, preliminarily de ning the oncogenic roles of CALM1 in ESCC cells. Further, our combined use of EGFR inhibitors signi cantly reduced cell proliferation, invasion and migration in vitro and in vivo. Our study provided a molecular phenotype for ESCC, suggesting that CALM1 and EGFR inhibitors might be used as a potential therapeutic target for patients with ESCC. It remains unknown what was the mechanism of CALM1 and EGFR in ESCC, and further study is warranted.
In conclusion, the combined effect of CALM1 and EGFR was observed to be able to remarkably inhibit tumor development in KYSE150 and Eca109 cells, suggesting that the combined effect of CALM1 and EGFR may assist in the development of new therapeutic strategies to enhance treatment e cacy of EGFR-targeted therapy.

Declarations Ethics approval and consent to participate
The study got approved by the Medical Ethics Committee of the First A liated Hospital of Xinjiang Medical University, with the written informed consents being obtained from all participants involved.

Consent for publication
All authors involved in the authorship are consent for publication in the current form.

Availability of data and material
Source data and materials can be available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests.    Knockout of CALM1 and application of EGFR inhibitor can synergistically inhibit the invasion and migration of ESCC. A, B, variation of invasive and migration ability was assessed by Transwell assay and Wound-healing assays in KYSE150 and Eca109 cell lines. C, Colony-formation assay to quantify the combined effect of CALM1 and EGFR on ESCC viability. D, Correspondingly, expression variation of biomarkers of ESCC cell lines KYSE150 and Eca109 related to EMT on protein level using immunoblotting. CALM1, calmodulin1; EGFR, epidermal growth factor receptor. ESCC, esophageal squamous cell carcinoma; EMT, epithelial-mesenchymal transition. *P < 0.05,**p < 0.01, ***p < 0.001 (Student's t-test).

Figure 4
Effects of Knockout of CALM1 and application of EGFR inhibitor on tumorigenesis in nude mice in vivo. KYSE150 cells that were infected with CALM1 or scramble lentivirus were injected s.c. into nude mice. Supplementary Files