Background
Clinical studies and experimental evidence have established that zinc levels are consistently decreased in prostate adenocarcinoma as compared to normal prostate glands. The Zip1 zinc uptake transporter, in addition to other members of the zinc transporter genes, is responsible for the uptake and accumulation of zinc from the circulation in the normal glands against the concentration gradient. The malignant prostatic glands exhibit a down regulation of Zip1, normally a ubiquitously expressed protein, expression that will eliminate the uptake and accumulation of zinc intracellularly [
1‐
4].
In contrast to the aforementioned role of zinc in the prostate, the role of hZip1 expression in adenocarcinomas of other organs has not previously been comprehensively examined. No such conclusive background exists for epithelial ovarian cancer for example which is considered one of the least understood of all major human malignancies [
5]. Identification of the most important initial alterations in ovarian cancer may facilitate the development of better methods for early diagnosis, through the discovery of epithelial ovarian cancer biomarkers, and for the development of more optimal therapeutic approaches that target key molecular pathways [
6].
There is evidence that relates zinc as a potential tumor suppressor agent and
ZIP1 transporter as a tumor suppressor gene in some epithelial ovarian cancers. Lightman et al. showed a highly significant decrease in the zinc concentration of malignant ovarian tissue as compared to their non-neoplastic counterparts [
7]. Bae et al. showed that zinc treatment of OVCAR-3 cells exerts tumor suppressor effects with increased apoptosis [
8]. The OVCAR-3 is a human epithelial ovarian cancer cell line which was established from the malignant ascites of a patient with poorly differentiated “papillary adenocarcinoma” of the ovary [
9]. A significant correlation between a decrease in blood and scalp-hair zinc levels with an increase in the incidence of ovarian cancer has been reported [
7,
10]. These preliminary data make it reasonable to expect that there is a difference in the zinc levels between malignant and non-malignant cells which may be related to alteration of the zinc uptake transporter; presumably Zip1.
In the current classifications of ovarian neoplasms, mucinous tumors are classified as surface epithelial tumors. Primary mucinous tumors of the ovary are classified into benign, borderline, or malignant categories depending on their histopathologic features. The neoplastic cells may be of gastrointestinal type, endocervical type, or mixtures of both cell types that exist in the seromucinous tumors [
11] . Primary invasive mucinous carcinomas of the ovary are very uncommon, accounting for only 3-10 % of all ovarian epithelial tumors, and metastatic tumors, most commonly from the gastrointestinal tract (GIT), are more common [
12,
13]. Therefore, the possibility that a mucinous tumor in the ovary is metastatic rather than primary ovarian carcinoma always must be entertained. Common primary sites are GIT, especially the large intestine [
14], appendix [
15,
16], and pancreas [
17] and less commonly from stomach [
18]. Endocervical adenocarcinomas occasionally spread to the ovary [
19].
We hypothesized that mucinous carcinomas, irrespective of different organ site, might share similarities at the molecular and biochemical levels, which are manifested by similar morphology at different anatomical sites. The anatomical site obviously will dictate the surgical approach; however, the molecular and biochemical characteristics might be more useful in the development and composition of adjuvant treatment protocols.
In this report, we show, for the first time, a consistent pattern of low expression of the hZip1 protein in mucinous carcinomas from different organs, including the ovary, colon, stomach and lung. The evidence presented herein indirectly supports the likelihood that ZIP1 gene is an essential step in the development of mucinous neoplasms.
Discussion
As a general rule, mucinous carcinomas represent a small fraction of neoplasms in different organs with about 3-10 % among ovarian neoplasms, 7 % of all colorectal cancers, 1.5 % of gastric carcinomas, and very rarely in the lung [
12,
20]. Compared with conventional adenocarcinomas, mucinous tumors tend to be associated with young age, advanced tumor stage, and distinct molecular patterns, such as microsatellite instability and mutations of the
BRAF and KRAS genes [
21,
22].
Differentiation of primary ovarian mucinous carcinoma from metastases of other organs mostly appendix and colorectum can be challenging [
13,
23]. This differentiation is pivotal and of clinical significance given that treatment protocols tend to be tailored to each organ site. For example platinum-based taxane agents are used in the treatment of primary ovarian carcinomas in contrast to fluorouracil which is the main chemotherapeutic agent used for metastatic carcinoma at the ovary originating from the colorectum [
24,
25].
For a tumor to be considered mucinous carcinoma rather than a conventional adenocarcinoma depends on organ site and classification systems. Ovarian mucinous carcinomas show conspicuous amounts of intracellular mucin in more than 90 % of tumor cells [
11]. In contrast, colorectal mucinous carcinoma is defined if at least 50 % of the tumor’s volume is composed of extracellular mucin [
26]. Mucinous carcinomas are distinct from conventional and signet ring carcinomas in different organs with specific molecular alterations and clinical outcomes [
25]. Common features between ovarian and colorectal mucinous carcinomas include higher prevalence at younger age and larger tumor size at presentation. In contrast, a higher proportion of ovarian mucinous carcinomas compared with serous carcinomas are diagnosed at a low stage, while colorectal mucinous carcinomas seem equally likely as non-mucinous carcinomas to present at a low stage [
23,
27,
28].
More than 2 decades ago, Lightman et al. [
7] reported low serum and tumor tissue zinc in patients with “malignant ovarian tumors”. However, in this publication, no mention of which kind of ovarian tumors were included in the study and not much clinical information was provided [
7]. Relatively more recently, Bae et al. [
8] reported that treatment of OVCAR-3 (malignant ovarian cell line) and NOSE (normal ovarian cell line) with zinc–citrate compound lead to induced zinc accumulation more in the former cell line than the latter. This resulted in a decrease in cell number and activity of M-aconitase in OVCAR-3 with increased apoptosis. The authors concluded that exposure to high concentration of zinc prevents the proliferation of OVCAR-3 cells [
8]. The OVCAR-3 cell line used by Bae et al. is a human epithelial ovarian cancer cell line which was established from the malignant ascites of a patient with poorly differentiated “papillary adenocarcinoma” rather than mucinous carcinoma of the ovary [
8,
9]. More mechanistic studies preferably with mucinous ovarian carcinoma cell lines e.g., RMUG-L, RMUG-S, MN-1, OMC-1 and MCAS [
29] are recommended to further explore the role of zinc and zinc transporters in the pathogenesis and at least theoretically, influence management of mucinous carcinomas.
We and others conclusively established that zinc levels and hZip1 transporter are significantly decreased in prostate adenocarcinomas. The background and supporting evidence are fully and extensively presented in several reviews and reports [
1‐
4]. This encouraged us to extend the study in the challenging mucinous carcinomas in different organs and to compare the hZip1 expression in mucinous versus non mucinous carcinomas in some of these organs.
Our novel finding in this report is that hZip1 showed consistently low expression in mucinous carcinomas of the ovary, colon, stomach and lung with no association between clinical stage and hZip1 scores in any of the mucinous tumors. Interestingly, we identified low expression of hZip1 in mucinous carcinomas versus non mucinous ones in three organs we performed the comparison in, namely; ovary, colon and stomach.
Specifically, hZip1 is strongly expressed in all cases of normal ovarian tissue and was moderately to strongly positive in 91 % of serous carcinomas. In contrast, 60 % of ovarian mucinous carcinomas were negative to weak positive for hZip1 expression. The second interesting finding is that there was a statistically significant lower expression of hZip1 in low grade ovarian serous compared to high grade serous carcinoma. Although it was not statistically significant, still hZip1 expression was lower in low grade compared to high grade mucinous carcinomas. These observations may indicate that the higher grade ovarian tumors may acquire different mechanisms to accumulate more zinc by switching off the inhibitory mechanisms which down express the zinc transporter hZIP1 gene, or perhaps other members of the zinc transporter gene family.
The scenario was the same in colonic and gastric carcinomas with low expression of hZip1 in mucinous carcinoma versus high expression in conventional adenocarcinomas in those two organs. In the colon and stomach, 55 % and 22 % of mucinous carcinomas were negative to weakly positive for hZip1, respectively. In contrast, hZip1 expression was moderately to strongly positive in 96 % and 100 % of conventional adenocarcinomas, respectively in the colon and stomach. The number of conventional gastric adenocarcinomas included in the TMA was small (n = 45). In contrast to ovarian carcinomas, hZip1 expression was decreased as a function of increasing grade in both mucinous and non-mucinous tumors of the colon and stomach. This observation may point to different mechanisms involved in tumor progression regarding regulation of zinc concentration which need more mechanistic studies. In the lung, we also identified low expression of hZip1 in pulmonary mucinous carcinomas. Unfortunately, and due to un-availability of conventional lung adenocarcinoma TMA slides, we did not compare the expression of hZip1 in mucinous and conventional pulmonary carcinomas.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
MMD and OF participated in the study design, and data interpretation. MMD, RBF and LC participated in the data creation. MMD conceived of the study, carried out the analysis of the immunohistochemical markers and performed statistical analysis. All authors drafted, read and approved the final manuscript.