Lung cancer is the leading cause of cancer death in the industrial nations [1]. Despite recent advances, therapeutic regimens support quality of life but frequently fail to increase long term survival. One of the main reasons for the failure of therapeutic regimens is the fact that cancer cells originate from normal cells and therefore possess similar characteristics. This means that anti-cancer therapies inevitably affect the normal cell population and these side effects often hinder more effective treatments. Thus, knowledge of the differences in the cellular physiology between malignant and non-malignant cells is crucial for the development of more successful treatments.

Calcium is a ubiquitous signal molecule that is involved in almost all cellular pathways [2, 3]. Elevation of the cytoplasmic Ca²⁺-concentration ([Ca²⁺]_c) can result either from Ca²⁺-influx from the extracellular space or from Ca²⁺-release from internal Ca²⁺-stores, primarily the ER. Proteins involved in the Ca²⁺-release from the ER are the inositol-1,4,5-trisphosphate receptor (IP₃R) and the ryanodine receptor (RyR) (Figure 1). Sarco/endoplasmic reticulum Ca²⁺-ATPases (SERCA) force calcium against the concentration gradient from the cytoplasm into the ER. Within the ER, calcium is buffered by calreticulin [2, 3]. Calcium is particularly important for the regulation of proliferation and apoptosis and the imbalance of cell growth and cell death finally leads to cancer. The aim of this study was therefore to evaluate whether the ER Ca²⁺-homeostasis is altered in lung cancer cell lines compared to normal bronchial epithelium.

In this study, we showed that the contribution of Ca²⁺-influx from the extracellular space to intracellular Ca²⁺-homeostasis varied between lung cancer cell lines. However, in those cell lines in which Ca²⁺-influx played a minor role (H1339 and HCC) the ER Ca²⁺-content was reduced compared to NHBE cells. The reduced Ca²⁺-content in H1339 and HCC cells correlated with a reduced expression of SERCA 2 pumping calcium into the ER, an increased expression of IP₃R releasing calcium from the ER, and a reduced expression of calreticulin buffering calcium within the ER. Reducing the ER Ca²⁺-content with CPA for 24 h led to an increased cell number.

The origin of the various lung carcinomas is still controversially being discussed. While squamous cell lung carcinomas are believed to origin from metaplastic bronchial epithelium, many authors believe small cell lung carcinomas to origin from neuro-epithelial bodies. But, the origin of large cell carcinomas and adeno carcinomas is less clear. However, being forced to choose a "normal" tissue to compare the malignant cell lines with, we decided to use normal human bronchial epithelial cells as a reference knowing that this choice constitutes a compromise.

SERCA is an ER transmembrane protein, consisting of a single polypeptide chain folded into four major domains, and is encoded by the genes ATP2A1-3 [5]. So far, 3 isoforms have been identified. As SERCA serves to maintain the concentration gradient between the cytoplasm and the ER by pumping calcium into the ER, SERCA has been regarded as a potential mediator of alterations of the ER Ca²⁺-content. In heart failure, the ER Ca²⁺-content of cardiac myocytes has been found to be reduced due to altered expression of SERCA [6]. In our laboratory, bronchial hyperreactivity in an asthma model was correlated with increased Ca²⁺-content in the sarcoplasmic reticulum of airway smooth muscle cells [4]. Further, in an interleukin based asthma model, the increased Ca²⁺-content was at least partially caused by increased expression of SERCA [7]. Several studies investigated the expression of SERCA in normal and tumor tissue reporting downregulation of this ATPase in cancer [8–11]. But, in colorectal cancer, Chung et al. reported that SERCA 2 mRNA was increased compared to normal tissue [12]. Moreover, increased SERCA 2 protein levels were correlated with serosal invasion, lymph node metastasis, advanced tumor stage and poorer survival-rate. Hence, an altered Ca²⁺-content of the ER might not only be involved in the early steps of carcinogenesis but may also cause further malignant transformation towards an invasive and aggressive phenotype.

Investigating the correlation of SERCA expression, [Ca²⁺]_ER and proliferation, Legrand et al. showed that in prostate cancer cells an increased growth rate was correlated with higher [Ca²⁺]_ER and increased SERCA 2 expression [13]. A decreased growth rate was correlated with decreased [Ca²⁺]_ER and decreased expression of SERCA 2b. Neuroendocrine differentiation of prostate cancer cells is considered to mark increased aggressiveness of cancer growth. Vanverberghe et al. showed that neuroendocrine differentiation in these cells was associated with apoptosis resistance probably due to decreased filling of the ER Ca²⁺-store caused by under-expression of SERCA 2 and calreticulin [14]. But, Crepin at al. reported that prolactin stimulated proliferation in immortalized prostate cells through increased [Ca²⁺]_ER due to increased SERCA 2 expression [15]. In a comprehensive review, Lipskaia proposed that proliferation is associated with a sustained increase in [Ca²⁺]_c or sustained Ca²⁺-oscillations, decreased refilling of the ER because of SERCA inhibition, and enhanced store operated Ca²⁺-entry from the extracellular space [16]. Apparently, the relationship between [Ca²⁺]_ER, SERCA expression and tumor growth varies between studies, cell types and differentiation status. However, an altered ER Ca²⁺-homeostasis is obviously involved in malignant transformation. To our knowledge, this is the first report showing an altered ER Ca²⁺-homeostasis in lung cancer cells.

The IP₃R is a Ca²⁺-channel composed of 4 subunits, which releases calcium upon the binding of IP₃ [17]. Sakakura and colleagues showed that the IP₃R was overexpressed in a gastric cancer cell line established from peritoneal dissemination, but weakly expressed in a gastric cancer cell line established from a primary tumor as well as in normal gastric epithelial cells [18]. The authors therefore suggest a role for the IP₃R in the transition to a metastatic phenotype. Our finding of increased IP₃R expression in H1339 and HCC cells is in agreement with in vivo data obtained from patients with resectable NSCLC, where Heighway et al. found amplification of the IP₃R gene in the tumor tissue compared to normal tissue [19].

Calreticulin is a 46-kDa chaperone that binds calcium in the lumen of the ER with high capacity [20]. It also participates in the folding of newly synthesized proteins. Recently, a role for calreticulin in immunogenic cell death has been proposed [21]. The authors reported that anthracyclines and γ-irradiation induced translocation of calreticulin to the plasma membrane thereby stimulating immunogenic cell death. In this context, our finding of reduced calreticulin expression in lung cancer cells could be of particular importance.

A decreased [Ca²⁺]_ER is regarded as a pathophysiological mechanism in heart failure [6]. Istaroxime is a SERCA activator that has been successfully tested in a clinical phase 1–2 trial and found to be well tolerated and to improve cardiac function [22]. As substances altering the intracellular Ca²⁺-homeostasis become available for clinical use, the altered Ca²⁺-homeostasis of cancer cells may become a valuable target to improve therapeutic options in lung cancer.

In our study, we showed that in H1339 and HCC cells the ER Ca²⁺-content was reduced compared to NHBE cells. The reduced Ca²⁺-content correlated with a reduced expression of SERCA 2 pumping calcium into the ER, an increased expression of IP₃R releasing calcium from the ER, and a reduced expression of calreticulin buffering calcium within the ER. The differences in the intracellular Ca²⁺-homeostasis between lung cancer and normal bronchial epithelial cells may lay the basis for new diagnostic or therapeutical approaches.