Thirteen BC cell lines (BT-20, BT-474, BT-549, HCC1419, HCC1954, Hs578T, MCF7, MDA-MB-231, MDA-MB-361, MDA-MB-415, MDA-MB-468, SK-BR-3 and ZR-75-1) and two non-cancerous breast epithelial cell lines (MCF-10A and MCF-12A) were used in the present study. BT-549, HCC1419, HCC1954 and Hs578T cell lines were purchased from the Japanese Collection of Research Bioresources Cell Bank (Osaka, Japan), and BT-474, MCF-7 and MCF-12A were gifted by Professor David Sidransky of Johns Hopkins University (Baltimore, MD, USA). All other cell lines were purchased from the American Type Culture Collection. All cell lines were cultured in RPMI-1640 medium (Sigma-Aldrich; Merck KGaA), supplemented with 10% fetal bovine serum and incubated in an atmosphere with 5% CO₂ at 37°C (13).

Human samples were resected from 114 patients with BC who had undergone surgery at Nagoya University Hospital between March 2002 and May 2007. The selected patients were those whose surveillance data for more than five years after surgery were available. All patients were females, and the median age was 53 years (range, 32–78 years). Primary BC and non-cancerous specimens and clinical data were collected from these patients. Clinical specimens were resected to ~1.5 mm in diameter and frozen immediately at −80°C. Non-cancerous specimens were resected >3 cm away from the edge of the tumor (12). The resected BC specimens were diagnosed histologically as BC and classified using the Union for International Cancer Control (UICC) staging system for BC (8th edition) (14). Adjuvant medication therapy was determined by physician discretion considering each patient's general condition, pathological features, and subtype (12,13).

The present study complied with the Declaration of Helsinki and was approved by the Nagoya University Hospital Institutional Review Board (approval number: 2019–0028). Participants provided written informed consent for the use of their clinical samples and data.

MZB1, DNA damage inducible transcript 3 (DDIT3), DERL3, and XBP1 mRNA expression levels were evaluated using RT-qPCR. RNA was extracted from cell lines (8.0×10⁶ cells per cell line) and from 114 patient BC and non-cancerous specimens using the RNeasy Mini kit (Qiagen GmbH). cDNA was synthesized as previously described (12,13,15). GAPDH mRNA levels were evaluated for normalizing MZB1, DDIT3, DERL3, and XBP1 mRNA expression levels. The primers specific for each gene were as follows: MZB1: Forward, 5′-CTCACAGGCCCAGGACTTAG-3′ and reverse, 5′-TGTGGCTGACACCTTCTCTG-3′, which generated a 219-bp product; DDIT3: Forward, 5′-AGCGACAGAGCCAAAATCAG-3′ and reverse, 5′-TGCTTTCAGGTGTGGTGATG-3′, which generated a 88-bp product; DERL3: Forward, 5′-CTCACTTTCCAGGCACCGT-3′ and reverse, 5′-TAGTAGATATGGCCCACCGC-3′, which generated a 110-bp product (12); XBP1: Forward, 5′-CAGACTACGTGCGCCTCTGC-3′ and reverse, 5′-CTTCTGGGTAGACCTCTGGG-3′, which generated a 208-bp product (12); and GAPDH: Forward, 5′-GAAGGTGAAGGTCGGAGTC-3′ and reverse, 5′-GAAGATGGTGATGGGATTTC-3′, which generated a 226-bp product (3). SYBR Green PCR core reagent kit (Thermo Fisher Scientific, Inc.) was used for RT-qPCR with these cycling conditions: One cycle at 95°C for 10 min, followed by 40 cycles at 95°C for 5 sec and 60°C for 60 sec, using an ABI StepOnePlus real-time PCR System (Thermo Fisher Scientific, Inc.). The 2^−ΔΔCt method was used for PCR quantification (16). All samples were assayed in triplicate. The mRNA expression levels of MZB1, DDIT3, DERL3, and XBP1 in each sample were obtained from the value divided by the GAPDH value for normalization (12,13,15).

Formalin-fixed, paraffin-embedded sections (4-µm thick) were constructed from blocks of resected specimens from 114 BC patients. The resected specimens were fixed with 10% formalin for 48 h. The slides were heated for 2 min for antigen retrieval with 1 mM EDTA buffer. The blocking procedure was not conducted. The MZB1 rabbit polyclonal antibody (cat. no. 11454-1-AP; ProteinTech, Inc.), which was diluted at 1:100, was used for immunohistochemistry, and sections were incubated for 1 h at room temperature (3). Then, SignalStain Boost IHC Detection reagent (cat. no. 8114; Cell Signaling Technology, Inc.) was used for the secondary antibody, and sections were incubated for 30 min at room temperature. The entire cancerous area of each section was observed using an upright light microscope (Olympus Corporation; ×100 and ×400 magnification). The staining intensity of the cytoplasm in cancer cells was evaluated and the intensity was divided into three groups: ‘Negative’, ‘weak’, and ‘strong’. Subsequently, ‘weak’ and ‘strong’ were combined and defined as ‘positive’.

Differences in the levels of MZB1 mRNA between two groups were evaluated using a Mann-Whitney test. Correlations between MZB1, DDIT3, DERL3, and XBP1 mRNA levels were analyzed using Spearman's rank correlation test. The associations between mRNA or protein expression levels of MZB1 and patient clinicopathological factors were analyzed using the χ² test. The Kaplan-Meier method was utilized for evaluating disease-free survival (DFS) and overall survival (OS) rates, and the survival curves were compared using the log-rank test. For multivariate regression analysis, the Cox proportional hazards model was utilized to identify prognostic factors. Other than MZB1 positivity, the variables were those considered to affect breast cancer prognosis, including age, tumor size, lymph node metastasis, and biological statuses. Next, variables for which P<0.05 were entered into the final model. JMP 12 (SAS Institute, Inc.) was employed for the statistical analysis, and P<0.05 was considered to indicate a statistically significant difference.

MZB1 mRNA expression in 13 BC cell lines and two non-cancerous cell lines from the mammary gland were evaluated (Fig. 1). The estrogen receptor (ER), progesterone receptor (PgR), and HER2 statuses of the cell lines have been evaluated in previous studies (17,18). MZB1 mRNA was detectable in four ER-positive BC cell lines, but not in the other nine BC and both non-cancerous cell lines. When MZB1 mRNA expression levels were compared between ER-positive and ER-negative BC cell lines, MZB1 mRNA expression levels in ER-positive cell lines were significantly higher than those in ER-negative BC cell lines (P=0.009).

The UICC stage distribution of 114 patients was as follows: stage 0, six patients; stage I, 29 patients; stage II, 56 patients; and stage III, 23 patients. T stage was distributed as follows: Tis (ductal carcinoma in situ), six patients; T1, 43 patients; T2, 54 patients; T3, six patients; and T4, five patients. Half of the patients had lymph node metastasis. The median follow-up duration was 123 months (range, 8–191 months) or until death. ER, PgR, and HER2 statuses, determined from immunohistochemistry tests in primary tumors, were as follows: ER-positive, n=86; ER-negative, n=28; PgR-positive, n=76; PgR-negative, n=38; HER2-positive, n=25; HER2-negative, n=80 (data missing for nine patients); triple-negative, n=12; and non-triple-negative, n=101 (data missing for one patient). Patients whose tumor expressed at least one of ER, PgR, or HER2 were defined as ‘non-triple-negative’. As eight patients out of nine whose HER2 statuses were unknown showed ER-positivity, they were categorized as non-triple-negative.

MZB1 is expressed, not only in BC cells, but also in the cellular components of non-cancerous specimens (e.g. mammary cells and lymphocytes). To evaluate MZB1 mRNA expression levels in clinical BC samples, the ‘MZB1 C/N ratio’, a ratio of MZB1 mRNA expression levels between BC and adjacent non-cancerous specimens, was adopted aiming to reduce the effects of MZB1 expression on non-cancerous tissues. There were no significant differences between Tis/T1 (n=49) and T2/T3/T4 patients (n=65; P=0.134). However, in patients with lymph node metastasis (n=57), the MZB1 C/N ratio tended to be higher than in lymph node-negative patients (n=57; P=0.064), and stage III patients (n=23) had a significantly higher MZB1 C/N ratio compared with stage 0/I/II patients (n=91; P=0.009; Fig. 2A). Patients whose MZB1 C/N ratios were higher than three were placed into a ‘high MZB1 group’ (n=33), and the remaining patients were designated as ‘others’ (n=81). The high MZB1 group was associated with lymph node metastasis (P=0.007) and a more advanced UICC pathological stage (P=0.006; Table I). As MZB1 mRNA expression levels were detected in only ER-positive cells among BC cell lines, MZB1 mRNA expression levels were evaluated in ER-positive patients (n=86). Stage III patients (n=13) had a significantly higher MZB1 C/N ratio than stage 0/I/II patients (n=73; P=0.003). Furthermore, the high MZB1 group (n=28) was associated with lymph node metastasis (P=0.004) and a more advanced UICC pathological stage (P=0.002), compared with others (n=58). In summary, MZB1 mRNA expression was associated with lymph node metastasis and a more advanced stage in patients with ER-positive BC.

As MZB1 has been reported to exist in the endoplasmic reticulum of cancer cells (5), the correlations between mRNA expression levels of MZB1, XBP1, and DDIT3, UPR-related molecules (11,19), were evaluated. Although there was no significant correlation between mRNA expression levels of MZB1 and XBP1 (correlation coefficient, 0.052; P=0.584; Fig. 2B), MZB1 mRNA expression was significantly correlated with that of DDIT3 (correlation coefficient, 0.421; P<0.0001; Fig. 2B). Furthermore, when MZB1 and DERL3 mRNA levels were evaluated, there was a significant correlation (correlation coefficient, 0.669; P<0.0001; Fig. 2B). These results suggested that MZB1 serves some role associated with UPR pathways.

As mRNA extracted from cancer specimens may contain mRNA derived from stromal cells, immunohistochemistry of MZB1 was conducted to assess the significance of its expression in BC cells. Among 114 patients with breast cancer, BC specimens from patients did not express MZB1 (determined as negative), and those from 35 and 29 patients expressed MZB1 weakly and strongly, respectively (determined as positive, Fig. 3A). There was no significant association between MZB1 positivity and T status (P=0.181), node status (P=0.706), or pathological stage (P=0.326). However, patients' age (≤60-year-old; P=0.009), ER-positivity (P=0.003), PgR-positivity (P=0.003), and non-triple-negativity (P=0.023) were significantly associated with MZB1 positivity (Table II). There were no differences between MZB1 positive and negative groups regarding DFS (P=0.478) or OS (P=0.996).

MZB1 mRNA expression levels were detected in ER-positive BC cell lines, and high MZB1 C/N levels were associated with a more advanced pathological stage in specimens from patients with BC. Furthermore, MZB1 exhibited increased expression in ER-positive patients (Table II). These results implied that there is an association between MZB1 and ER-positive BC. In ER-positive patients (n=86), DFS rates was significantly poorer in MZB1-positive patients (n=55) compared with MZB1-negative patients (n=31; 5-year DFS rates: MZB1-positive group, 80.0%; negative group, 93.4%; P=0.026; Fig. 3B). By contrast, OS in MZB1-positive patients did not differ from that in MZB1-negative patients (5-year OS rates: MZB1-positive group, 94.5%; negative group, 96.8%; P=0.260; Fig. 3C). Multivariate analysis of DFS identified ‘lymph node metastasis’ (HR, 7.81; 95% CI, 2.10–50.9; P=0.001) and ‘MZB1 positivity’ (HR, 4.30; 95% CI, 1.21–27.4; P=0.022) as independent prognostic factors (Table III).