Genome-wide association studies recently linked the locus for Na+,HCO3−-cotransporter NBCn1 (SLC4A7) to breast cancer susceptibility, yet functional insights have been lacking. To determine whether NBCn1, by transporting HCO3− into cells, may dispose of acid produced during high metabolic activity, we studied the expression of NBCn1 and the functional impact of Na+,HCO3−-cotransport in human breast cancer. We found that the plasmalemmal density of NBCn1 was 20–30% higher in primary breast carcinomas and metastases compared to matched normal breast tissue. The increase in NBCn1 density was similar in magnitude to that observed for Na+/H+-exchanger NHE1 (SLC9A1), a transporter previously implicated in cell migration, proliferation and malignancy. In primary breast carcinomas, the apparent molecular weight for NBCn1 was increased compared to normal tissue. Using pH-sensitive fluorophores, we showed that Na+,HCO3−-cotransport is the predominant mechanism of acid extrusion and is inhibited 34 ± 9% by 200 μM 4,4′-diisothiocyanatostilbene-2,2′-disulfonic acid in human primary breast carcinomas. At intracellular pH (pHi) levels >6.6, CO2/HCO3−-dependent mechanisms accounted for >90% of total net acid extrusion. Na+/H+-exchange activity was prominent only at lower pHi-values. Furthermore, steady-state pHi was 0.35 ± 0.06 units lower in the absence than in the presence of CO2/HCO3−. In conclusion, expression of NBCn1 is upregulated in human primary breast carcinomas and metastases compared to normal breast tissue. Na+,HCO3−-cotransport is a major determinant of pHi in breast cancer and the modest DIDS-sensitivity is consistent with NBCn1 being predominantly responsible. Hence, our results suggest a major pathophysiological role for NBCn1 that may be clinically relevant.
A high metabolic rate and insufficient blood supply are prominent characteristics of most malignant tumors promoting anaerobic metabolism.1–3 Furthermore, genetic and biochemical changes that take place during malignant transformation are associated with a metabolic shift—known as the Warburg effect—which favors glycolysis even under aerobic conditions.2, 3 Anaerobic metabolism is accompanied by lactic acid production, and aerobic metabolism can be followed by intracellular hydration of CO2, which will also lead to intracellular acid loading. As a consequence, the high metabolic rate of malignant tumors renders cancer cells vulnerable to developing intracellular acidification.1, 4, 5 Yet, all available evidence indicates that while the tumor microenvironment is acidic, the intracellular pH (pHi) of cancer cells within tumors is similar to, or even more alkaline than, that of normal cells.4–8 The inference from these observations is that very efficient mechanisms of acid extrusion operate in cancer cells to dispose of the excess acid to the point of actually decreasing cytosolic [H+] levels below those of normal cells.
Membrane acid-base transport represents an important pathway for disposing of cellular acid and can be mediated by a number of different membrane proteins. One protein in particular, the ubiquitously expressed Na+/H+-exchanger NHE1 (SLC9A1), has been extensively studied for a role in cancer development and suggested as a possible target for treatment.1, 7, 9, 10 The Na+,HCO3−-cotransporters comprise another prominent family of membrane proteins, which can contribute to net acid extrusion by transporting HCO3− and Na+ into cells.1 Na+-dependent HCO3−-transport has been proposed as an important mechanism for net acid extrusion in a number of cancer cell lines,9, 11–13 and based on studies in the human MCF-7 breast cancer cell line, the electroneutral Na+,HCO3−-cotransporter NBCn1 (SLC4A7) has been suggested as a molecular candidate for this transport.9 High NBCn1 activity could contribute to maintaining a relatively high pHi while keeping extracellular pH (pHo) low. This distribution of acidity is believed to favor cancer development because a high pHi is important for cancer cell growth and survival, while a low pHo is important for migration and invasiveness of the cancer cells.7
Na+,HCO3−-cotransport and polymorphisms in NBCn1 recently attracted attention when genome-wide association studies (GWAS)14–17 provided a link between NBCn1 and human breast cancer. In independent GWAS,14–17 the single nucleotide polymorphism (SNP) rs4973768 localized to the 3′ untranslated region (UTR) of NBCn1 was reported to be linked to breast cancer in women of European, Chinese, Korean and Japanese descent, with odds ratios for homozygosity between 1.2 and 1.3. We subsequently noted that the location of the rs4973768 SNP could confer a change in binding affinity to the microRNA miR-569, which may result in altered NBCn1 expression.1 Thus, several lines of evidence point to a major role for NBCn1 in breast cancer, although studies of NBCn1 expression and function in human breast tissue have been lacking.
In this study, we tested the hypothesis that NBCn1 expression and Na+,HCO3−-cotransport are prominent in human breast cancer and important for pHi regulation. We show that the membrane density of NBCn1 is increased in malignant human breast tumors relative to normal breast tissue and demonstrate that while both Na+/H+-exchange and Na+,HCO3−-cotransport contribute to cellular pH regulation, net acid extrusion by Na+,HCO3−-cotransport is the major determinant of pHi in human primary breast carcinomas.
CI: confidence interval; DCIS: ductal carcinoma in situ; EIPA: 5′-(N-ethyl-N-isopropyl)-amiloride; GWAS: genome-wide association study; NMDG+: N-methyl-D-glucammonium; pHi: intracellular pH; pHo: extracellular pH; SNP: single nucleotide polymorphism; UTR: untranslated region; VIS: Visiopharm Integrator System
Material and Methods
Tissue samples were provided by the Department of Pathology at Copenhagen University Hospital within the framework of the Danish Center for Translational Breast Cancer Research-DCTB. Written informed consent was obtained from all participants undergoing surgery and axillary dissection. Clinicopathological data for patients included in this part of the study are given in Supporting Information Table 1. Patients had not undergone previous surgery of the breast and had not received preoperative treatment. Samples of malignant tissue, matched nonmalignant tissue and lymph node metastases were collected. The project was approved (KF 01-069/03) by the Copenhagen and Frederiksberg regional division of the Danish National Committee on Biomedical Research Ethics.
Formalin-fixed, paraffin-embedded sections (5 μm thick) were cut from the tissue blocks and mounted on Super Frost Plus slides (Menzel-Gläser, Germany), baked at 60°C for 60 min, deparaffinized and rehydrated through graded alcohol rinses. Heat-induced antigen retrieval was performed by immersing the slides in Tris/EDTA pH 9.0 buffer (10 mM Tris, 1 mM EDTA) and microwaving at 750 W for 10 min. The slides were then cooled at room temperature for 20 min, rinsed thoroughly in tap water and blocked with 10% normal goat serum in phosphate buffered saline for 15 min. Endogenous peroxidase activity was quenched with 0.3% H2O2 in methanol for 30 min. Next, slides were incubated for 1 hr with previously characterized primary antibodies against NBCn1 (rabbit polyclonal anti NH2-terminal antibody,18 generously provided by Dr. Jeppe Praetorius, Aarhus University), NHE1 (Xb-17 rabbit polyclonal antibody, generously provided by Dr. Mark Musch, University of Chicago) or GLUT1 (DAKO, Denmark) followed by 30 min with a species-matched secondary antibody conjugated to a peroxidase complex (Envision+ poly-HRP system; DAKO, Denmark). Color development was done using DAB+ Chromogen (DAKO, Denmark). Slides were counterstained with hematoxylin. Incubation and development times were standardized to allow for accurate comparisons between slides.
Immunohistochemical staining intensities of tissue sections were digitized and quantified using an automated quantitative microscopy integrated system (Visiopharm Integrator System (VIS) version 22.214.171.124.; Visiopharm A/S, Denmark). The VisioMorph™ analysis software module was used to derive endpoints following segmentation of images. The mean staining intensity of stained cell membranes was calculated for each section using APP ID#10025 developed in VisiomorphDP (Supporting Information Figure 1), and following background subtraction, data were presented relative to the membrane density level of normal breast tissue. It should be noted that the resolution of the immunohistochemical images does not allow us to discriminate between proteins expressed in the plasma membrane and proteins expressed in vesicles in close proximity to the plasma membrane.
Indirect immunofluorescence analyses
Sections (5 μm thick) cut from paraffin blocks of breast tissue samples (obtained as described above for immunohistochemical analyses) were mounted on Super Frost Plus slides (Menzel-Gläser, Braunschweig, Germany), baked at 60°C for 60 min, deparaffinized and rehydrated through graded alcohol rinses. Heat-induced antigen retrieval was carried out as described above. Following antigen retrieval, sections were treated with Image-iT FX™ signal enhancer (Molecular Probes, OR, USA) to block nonspecific staining and subsequently incubated with the relevant primary antibodies at the appropriate dilutions. Detection of immune-complexes was done with species-specific secondary antibodies conjugated to Alexa Fluor® 488, Alexa Fluor® 568 or streptavidin-conjugated Alexa Fluor® 647 (Molecular Probes). Nuclear material was counterstained with DAPI. The sections were washed three times with cold phosphate-buffered saline between incubations. Normal rabbit or mouse sera instead of primary antibody were used for negative control experiments. Sections were imaged using a Zeiss LSM510META confocal laser scanning microscope (Carl Zeiss MicroImaging Gmbh, Germany).
Intracellular pH measurements
Freshly dissected human breast cancer samples for acute analysis of pHi were obtained from women undergoing breast conserving surgery at the Surgical Department P, Aarhus University Hospital, Denmark. Women with a tumor larger than 10 mm who were at least 18 years of age and presented with surgically operable primary breast cancer diagnosed by triple test (clinical examination, mammography and fine needle aspiration biopsy) or core needle biopsy were included. All participants gave written informed consent, and the procedure for obtaining biopsies was approved (M-20100288) by the Mid-Jutland regional division of the Danish National Committee on Biomedical Research Ethics. Upon excision, tumor samples were immediately placed in physiological salt solution and brought to the laboratory for further analyses. Clinicopathological characteristics of the patients included in this part of the study are provided in Supporting Information Table 2.
Tumor samples were cut to ∼0.5-mm-thick slices, placed on a glass coverslip in a custom-built tissue chamber and submerged in a continuously aerated salt solution. The tissue was loaded with 5 μM of the pH-sensitive fluorophore BCECF-AM in DMSO (final concentration of DMSO was 0.05%) for 20 min at 37°C. The loaded preparations were investigated on the stage of an Olympus IX70 microscope equipped with an Olympus LUCPlanFL N 20× objective (N.A. 0.45) and an EasyRatioPro fluorescence imaging system (Photon Technology International, NJ, USA). The preparations were excited alternately at 485 and 440 nm and emission light collected at 550 nm. The BCECF ratio between emission light collected after excitation at 485 and 440 nm, respectively, was calculated after subtraction of background fluorescence and calibrated to pH units using the high-K+ nigericin method as previously described.19 Intracellular intrinsic buffering capacity was calculated based on the acidification following washout of NH4Cl as described earlier.20 The intracellular intrinsic buffering capacity was ∼59 mM in the pHi range 6.55–6.70. We do not have sufficient data points to calculate intracellular intrinsic buffering capacity at lower pHi values than 6.55. It is important to note that an increase in intrinsic buffering capacity at low pHi—which has been reported in some cell types20—would not change our conclusion that Na+/H+-exchange is important primarily at low pHi values although it would increase the transport activities calculated at low pHi relative to those calculated at higher pHi and hence result in a steeper activation profile for the transporters with decreasing pHi. The contribution from CO2/HCO3− to intracellular buffering was calculated from the formula: βCO2/HCO3 = 2.3 [HCO3−]i, as previously described.21
The CO2/HCO3−-containing salt solution used for the functional experiments had the following composition (in mM): 139 Na+, 5.9 K+, 1.6 Ca2+, 1.2 Mg2+, 122 Cl−, 25 HCO3−, 1.2 SO42−, 1.2 H2PO43−, 10 HEPES, 5.5 glucose and 0.026 EDTA. In Na+-free solutions, Na+ was substituted with an equimolar amount of N-methyl-D-glucammonium (NMDG+), except for NaHCO3, which was replaced with choline-HCO3. In HCO3−-free solutions, HCO3− was replaced with an equimolar amount of Cl−. HCO3−-containing solutions were aerated with a gas mixture of 5% CO2 balance air, HCO3−-free solutions were gassed with atmospheric air (nominally CO2-free); pH was adjusted to 7.40 at 37°C. All solutions contained 5 mM probenecid to inhibit cellular extrusion of the liberated BCECF by the organic anion transporter.
RNA isolation, reverse transcription and polymerase chain reaction
NBCn1 and NHE1 expression at mRNA level was investigated using two-step RT-PCR analyses performed on human primary breast carcinomas and matched normal breast tissue. Tissue samples were obtained by the same procedure described above for pHi measurements. RNA isolation was performed using RNeasy mini kit (Qiagen, Germany), and the reverse transcription and PCR reactions performed as previously described.20 The primer sequences were as follows for NBCn1: forward, 5′ CAG ATG CAA GCA GCC TTG TGT G 3′; reverse, 5′ GGT CCA TGA TGA CCA CAA GCTG 3′; and NHE1: forward, 5′ TCC ACC GTC TCC ATG CAG AAC 3′; reverse, 5′ GTC TCC TTG CTC CGC ATC ATG 3′.
The quantitative reverse transcription and PCR analyses were performed as described earlier20 using TaqMan® real time PCR. The following primers and probes were used for amplification of total NBCn1: forward, 5′ GCA AGA AAC ATT CTG ACC CTC A 3′; reverse, 5′ GCT TCC ACC ACT TCC ATT ACC T 3′; probe, 5′ TCC TGG AAA CTT GGA CAA TAG TAA AAG TGG AG 3′. To specifically amplify cassette II containing NBCn1 transcripts, the forward primer was changed to 5′ TTC CCA CAG TAG TAA TTC ATC CGC 3′. The probes were modified with a 5′ FAM and a 3′ TAMRA.
Western blot analyses
NBCn1 and NHE1 protein expression was investigated by Western blot analyses on normal breast tissue and primary breast carcinomas. Tissue samples were obtained using the procedure described above for pHi measurements. The experimental protocol has previously been described.22 Previously characterized antibodies against NBCn1 (rabbit polyclonal anti NH2-terminal antibody18; kindly provided by Dr. Jeppe Praetorius, Aarhus University) and NHE1 (mouse monoclonal sc-136239; Santa Cruz, CA, USA) were used.
Unless otherwise specified, data are expressed as mean ± SEM and statistical analyses performed using paired, two-tailed Student's t-tests or repeated measures one-way or two-way ANOVA followed by Dunnet or Bonferroni post-tests. Details on the statistical analyses used for specific comparisons are given in the figure legends. p < 0.05 was considered statistically significant; n equals number of patients. Statistical analyses were performed using NCSS2007 (UT, USA) or GraphPad Prism 5.02 (CA, USA) software.
Expression of NBCn1 and NHE1 in human breast cancer and normal mammary tissue
NBCn1 and NHE1 mRNA (Fig. 1a) and protein (Fig. 1b) were detected in homogenates of human breast cancer samples and matched normal breast tissue. Interestingly, the apparent molecular weight of the NBCn1 protein was consistently higher in primary breast carcinomas than in normal breast tissue, while there was no detectable difference in the apparent molecular weight of the NHE1 protein in the same samples (Fig. 1b). Possible explanations for the difference in apparent molecular weight for NBCn1 include alternative splicing or altered post-translational modification(s) specifically in cancer tissue. Although several splice variants have been described for NBCn1, splice cassette II consisting of 124 amino acids is the only known splice cassette large enough to likely account for the observed change in apparent molecular weight.1, 23 This splice cassette is typically absent from epithelial tissue but present among others in vascular smooth muscle cells.1 The proportion of NBCn1 mRNA transcripts coding for cassette II in the cancer tissue was 68% (95% CI: 48% − 95%; n = 10; p < 0.05) of that in the normal breast tissue. The lower relative expression of cassette II containing NBCn1 transcripts is consistent with the increased epithelial cell abundance in the primary breast carcinomas compared to normal tissue and suggests that alternative splicing of cassette II does not account for the higher apparent molecular weight of NBCn1.
Breast cancer and normal breast tissue samples are highly complex structures, which in addition to epithelial cells include a vast connective tissue component containing among others adipocytes, fibroblasts, an abundant extracellular matrix component, and smaller and larger blood vessels. Considering these structural differences and the broad NBCn1 and NHE1 expression patterns,24, 25 expression levels in tissue homogenates cannot be directly interpreted as a measure of NBCn1 and NHE1 expression in carcinoma cells and normal breast epithelial cells. Hence, we performed immunohistochemical analyses to evaluate the density of these transporters in plasma membranes of normal and malignant epithelial cells. As shown in Figure 2, both NBCn1 and NHE1 were expressed in normal breast epithelium (Figs. 2a,d), ductal carcinoma in situ (DCIS) lesions (Figs. 2b,e), invasive primary breast carcinomas (Figs. 2c,f) and axillary lymph node metastases (data not shown). Although the expression pattern was heterogeneous with a considerable degree of variability between different tumor areas (Figs. 2c,f and 3a,b), the average plasmalemmal density of NBCn1 was increased in primary breast carcinomas and metastases compared to matched normal breast tissue (Fig. 2g). Importantly, the increase in NBCn1 density was similar in magnitude to that observed for NHE1 (Fig. 2h), a protein with a well-established role in pHi regulation and which has been directly associated with cellular transformation, invasion and metastasis. NHE1 and NBCn1 were very highly expressed in DCIS lesions (Figs. 2b,e). Crucially, in the primary breast carcinomas, both NBCn1 and NHE1 expression occurred predominantly in cytokeratin (CK)-19 positive cells (Fig. 4), which are generally assumed to be epithelium-derived.26 Expression in other cell types was very limited. Taken together, these findings support the hypothesis that NBCn1 and NHE1 may play a prominent role for pHi regulation in invasive cells of human primary breast carcinomas.
In addition to upregulation of NBCn1 and NHE1 expression, we also observed a trend in the patterns of immunostaining with NHE1 being most highly expressed in the peripheral and well-perfused areas of the tumors (Fig. 3a), whereas NBCn1 was expressed more uniformly throughout the tumors (Fig. 3b).
A recently proposed conceptual model of breast carcinogenesis suggested that adaptation to hypoxia and acidosis may represent key events in transition from carcinoma in situ to invasive cancer.27, 28 To address the issue of whether focal upregulation of NHE1 or NBCn1 occurred as a response to microenvironmental factors (such as regional hypoxia) or was a constitutive event that conferred a proliferative or survival advantage, we examined the patterns of expression of the hypoxia-responsive glucose transporter protein GLUT1 for possible correlations with NHE1 and NBCn1. However, we were unable to distinguish any consistent correspondence between the pattern of expression of GLUT1 and those of NHE1 and NBCn1 (illustrated in Figs. 3a–c).
Mechanisms of intracellular pH control in human breast cancer
Having established that the plasma membrane expression of NBCn1 and NHE1 is upregulated in breast cancer, we proceeded to evaluate their roles in the regulation of pHi in freshly dissected human primary breast carcinomas. The mechanisms of cellular acid extrusion during intracellular acidification were investigated using the NH4+-prepulse technique.21 An original recording is shown in Figure 5a. Addition of NH4+, which in aqueous solution is in chemical equilibrium with NH3, induces abrupt intracellular alkalinization as NH3 quickly penetrates the cell membranes and takes up protons from the intracellular environment. This is followed by a gradual acidification due to slower influx of NH4+ and activation of cellular base extrusion. Upon washout of NH4+, NH3 promptly leaves the cells and leaves behind the imported protons causing intracellular acidification. Pharmacological tools and ion replacement were used to assess the roles of NBCn1 and NHE1 in pHi recovery.
When experiments on tumor slices were performed in the presence of CO2/HCO3−, washout of NH4Cl into a Na+-free solution containing 10 μM of the Na+/H+-exchange inhibitor 5′-(N-ethyl-N-isopropyl)-amiloride (EIPA) caused a rapid intracellular acidification followed by a very stable pHi (Fig. 5b). On addition of Na+ in the continued presence of EIPA, pHi quickly recovered toward normal levels (Fig. 5b) consistent with activation of Na+,HCO3−-cotransport (Fig. 5c). As shown in Figure 5d, when samples of human primary breast carcinomas were preincubated with 200 μM of the anion transport inhibitor 4,4′-diisothiocyanatostilbene-2,2′-disulfonic acid (DIDS), the Na+-dependent and EIPA-insensitive pHi recovery rate was reduced by 34 ± 9% (n = 5, p < 0.05). NBCn1 heterologously expressed in Xenopus oocytes29 or HEK293 cells30 or natively expressed in epithelial tissues31–33 has a low sensitivity to DIDS in contrast to other known Na+,HCO3−-cotransporters of the SLC4-family.1, 34 Therefore, this modest effect of DIDS on the rate of pHi recovery is consistent with NBCn1 being the predominant Na+,HCO3−-cotransporter mediating cellular net acid extrusion from human primary breast carcinomas under these conditions.
When experiments were performed in the nominal absence of CO2/HCO3− to study HCO3−-independent mechanisms of acid extrusion, washout of NH4Cl into a Na+-free solution first produced a rapid acidification which was followed by a period with a slow pHi decrease (Fig. 6a). The slow phase of acidification would be consistent with reverse Na+/H+-exchange (i.e., Na+ moving out of cells in exchange for H+ moving in) as previously described for NHE1 in vascular endothelial cells22 and mouse neocortical neurons.35 When Na+ was returned to the bath solution in the continued absence of CO2/HCO3−, the slow acidification was replaced by a slow alkalinization (Figs. 6a,b) consistent with Na+/H+-exchange working in the forward mode (i.e., Na+ moving into cells in exchange for H+ moving out). To compare the activities of Na+/H+-exchange and Na+,HCO3−-cotransport, we next switched to a Na+- and CO2/HCO3−-containing solution and observed a strongly increased pHi recovery rate (Fig. 6a) supporting a predominant role of Na+,HCO3−-cotransport for pHi regulation in the breast cancer tissue. In fact, when compared before and after the switch from CO2/HCO3−-free to CO2/HCO3−-containing solutions (bars 3 and 4 in Fig. 6b), the rate of net acid extrusion due to Na+,HCO3−-cotransport even at this low level of pHi (∼6.6) appeared to be around 10 times greater than the rate of acid extrusion due to Na+/H+-exchange (Fig. 6b). At lower pHi levels, Na+/H+-exchange contributed significantly to acid extrusion, but at pHi levels around 6.6, Na+/H+-exchange activity was very low (Fig. 6a). Since both Na+/H+-exchange and Na+,HCO3−-cotransport activities decrease with increasing pHi, our data strongly imply that Na+/H+-exchange is important for acid extrusion primarily at very low pHi levels, while Na+,HCO3−-cotransport is the dominating mechanism of cellular acid extrusion at pHi levels higher than 6.6.
Finally, we evaluated whether the prominent Na+,HCO3−-cotransport activity of the human primary breast carcinomas is important for controlling resting steady-state pHi (Fig. 6c). In the nominal absence of CO2/HCO3−, pHi was 0.35 ± 0.06 units lower than in the presence of CO2/HCO3− (n = 5; p < 0.01), supporting a role for Na+,HCO3−-cotransport in maintaining a high steady-state pHi. The increased pHi in the presence of CO2/HCO3− suggests that HCO3− influx via Na+,HCO3−-cotransport in the resting steady-state is greater than HCO3− efflux via Cl−/HCO3−-exchange. In steady-state, this net HCO3− influx must be accompanied by net CO2 efflux additional to that generated by cellular respiration.
In this study, we investigated the cellular and molecular mechanisms responsible for disposing of intracellular acid in human breast cancer tissue. In contrast to previous studies, our functional data regarding pHi regulation are based on freshly isolated tissue slices, which—although more difficult to interpret than cell cultures—are likely to more directly reflect in vivo conditions. In these preparations, we show that both Na+,HCO3−-cotransporter NBCn1 and Na+/H+-exchanger NHE1 are almost exclusively expressed in the CK-19 positive (i.e., epithelium-derived) cells. The high abundance of CK-19 positive cells in the tumor slices ensures that the pHi signals recorded provide a good measure of the pHi regulatory function of the breast epithelium.
The most pertinent finding of this study is the very prominent role for NBCn1, which—along with NHE1—is generally expressed at higher levels in human primary breast carcinomas and metastases compared to normal breast epithelium. We have previously shown that expression of a constitutively active, amino-truncated ErbB2 receptor causes marked upregulation of NBCn1 in the human MCF-7 breast cancer cell line.9 In this study, only 20% of the samples used for expression analyses and 10% of the samples investigated for pHi regulatory function showed overexpression or gene amplification of the ErbB2 receptor (see Supporting Information Table 1 and 2) suggesting that NBCn1 and Na+,HCO3−-cotransport are of general importance in human breast cancer and not restricted to cancers influenced by increased ErbB2 signaling. Given the limited size of the sample set underlying this study, we were unable to determine the effect of ErbB2 status on NBCn1 expression and pHi regulatory function in human breast cancer.
Our findings are particularly striking in light of the series of recent GWAS demonstrating that a SNP in the 3′ UTR of NBCn1 appears to play a role for breast cancer susceptibility in women of both European16 and Asian origin.14, 15, 17 It will be important to address whether this SNP is associated with increased levels of NBCn1 expression, and consequently increased capacity for pHi regulation, favoring tumor development.
The only previous study, which to our knowledge has investigated NBCn1 expression in human breast cancer samples, found lower expression in invasive ductal carcinomas compared to matched normal breast tissue in patients of Asian origin.36 Furthermore, based on the MCF10AT progression model of breast proliferative disease, the authors of this study showed that NBCn1 expression decreased from low-grade to high-grade lesions.36 Notably, however, NBCn1 expression was approximately three times higher in low- and intermediate-grade lesions than in normal cells36 suggesting that at least early in breast cancer development, NBCn1 expression is upregulated. The reason for the conflicting findings is presently unclear but could be related to the high degree of regional heterogeneity associated with NBCn1 expression: in the specimens, which we examined, normal samples showed overall low levels of staining (Supporting Information Figure 2), but displayed occasional regional upregulation (inset, Supporting Information Figure 2; compare glandular cells displaying strong and weak expression of NBCn1, yellow and black arrows, respectively).
Although NBCn1 and NHE1 mediate ostensibly similar transport functions (i.e., exchange of one extracellular Na+ for the equivalent of one intracellular H+), our data suggest that Na+,HCO3−-cotransport and Na+/H+-exchange play different roles in pHi regulation. Hence, Na+/H+-exchange seems to contribute importantly to cellular acid extrusion in primary breast carcinomas at very low pHi values (below pHi 6.6), whereas Na+,HCO3−-cotransport is the major mechanism of cellular acid extrusion at higher pHi levels and consequently crucial for control of resting steady-state pHi. The particularly important role of Na+,HCO3−-cotransport in the near-physiological pHi range and considerable contribution from Na+/H+-exchange at low pHi is consistent with findings from other cell types (e.g., vascular smooth muscle and endothelial cells) where NBCn1 and NHE1 are also co-expressed.22, 37, 38 As such, even though there is currently no direct evidence that NBCn1 activity is upregulated in malignant tissue, our findings demonstrate that Na+,HCO3−-cotransport is critically important for pHi regulation in human breast cancer.
It should be noted that our present findings from freshly isolated human breast cancer slices differ in some important aspects from our previous observations based on the MCF-7 human breast cancer cell line expressing a constitutively active, amino-truncated ErbB2 receptor. In these cells, NBCn1 and NHE1 contributed approximately equally to net acid extrusion also in the near-physiological pHi range.9 The MCF-7 based cell studies also suggested that acid-base transporters may play important functions which are at least in part independent of pHi since NHE1—but not NBCn1—was found to be important for cell migration and chemotherapy sensitivity despite fairly similar effects on overall cytosolic pH.9, 10 Although our current findings suggest that Na+,HCO3−-cotransport plays a greater role than Na+/H+-exchange for cellular acid extrusion in human primary breast carcinomas, additional studies are required to further elucidate the importance of these respective transport mechanisms and pHi for integrated cellular responses such as cell migration, proliferation and invasiveness involved in cancer development and progression.
In addition to effects on pHi regulation, the increased NBCn1 and NHE1 membrane densities in primary breast carcinomas compared to normal breast tissue are also likely to affect tumor pHo. Notably, the consequence of increased transporter densities for the extracellular environment will be amplified by the higher abundance of epithelial cells in breast cancer tissue compared to normal breast tissue. Collectively, therefore, the high NBCn1 and NHE1 membrane densities and high proportions of epithelial cells in tumors have the potential to greatly intensify extracellular acid loading and thus may form the molecular basis for the acidic tumor microenvironment.
Although CO2 and HCO3− play a prominent role for pH regulation by supplying substrate for HCO3−-dependent transporters, they may also play a role as mobile extracellular buffers and thereby contribute to acid extrusion by reducing acid build-up in the extracellular space.11 To focus on the role of CO2/HCO3− as a transporter substrate, we minimized the inhibitory effect of removing CO2/HCO3− on extracellular diffusion by including 10 mM HEPES (which can serve as a mobile extracellular buffer11) in the bath solutions and recording pHi responses close to the periphery of the preparations where diffusion to and from the bulk bath solution was fast as judged from accessibility for the fluorescent dye, BCECF-AM.
Our immunohistochemical analyses indicated that NHE1 may localize predominantly to the periphery and well-perfused areas of the primary breast carcinomas, whereas NBCn1 is more broadly distributed throughout the tumors, although given the limited size of our sample set, this correlation is still speculative. Notably, our findings are in line with a recent report on the distribution of transporters in brain tumors39 in which NHE1 was found to localize predominantly to the periphery of the tumors (NBCn1 distribution was not analyzed). This spatial distribution of the transporters would further reinforce the recent proposal that Na+,HCO3−-cotransport may be quantitatively most important in poorly perfused tumor areas where the low pHo inhibits Na+/H+-exchange activity more than Na+, HCO3−cotransport activity.11
The increase in NHE1 and NBCn1 expression in lymph node metastases compared to normal breast epithelium is interesting in light of the known localization of NHE1 to invadopodia and its important role in regulation of cell adhesion, migration and invasion.10, 40, 41 Further studies are, however, required to establish the possible functional roles of the two transporters in breast cancer metastasis in vivo.
Intriguingly, we found that the apparent molecular weight of the NBCn1 protein expressed in human primary breast carcinomas is greater than in matched normal breast tissue. This finding suggests a difference in molecular structure, which could in principle be caused by splice variation (of which multiple physiological variants have been described23 and additional variants could be related to cancer-induced changes in RNA splicing42) or any of a number of different post-translational modifications including glycosylation,43 acetylation44 or phosphorylation.45 In addition to its potential functional importance, this difference between NBCn1 expressed in normal and transformed breast tissue may provide diagnostic information and a promising means to pharmacologically target NBCn1 in cancer tissue without affecting NBCn1 elsewhere. Considering the broad expression profile of NBCn1,24 the apparent change in molecular structure for NBCn1 may greatly improve its potential as a clinically useful target in cancer treatment.
In conclusion, we have shown here that expression of the Na+,HCO3−-cotransporter NBCn1 is upregulated in human primary breast carcinomas and metastases relative to normal breast tissue. Furthermore, while both Na+/H+-exchange and Na+,HCO3−-cotransport contribute to cellular pH regulation, net acid extrusion by Na+,HCO3−-cotransport is the major determinant of pHi in human primary breast carcinomas except under highly acidic pHi conditions. Given the increased need for net acid extrusion in all solid malignant tumors and the broad expression profile of NBCn1 and NHE11, 24, these findings are likely to be relevant also for other types of cancer.
The authors thank Susie Mogensen and Jane Rønn for expert technical assistance, Drs. Jeppe Prætorius, Aarhus University, and Mark Musch, University of Chicago, for the polyclonal NBCn1 and NHE1 antibodies, respectively, and Dr. Thomas Litman for 3′ UTR analyses. The Water and Salt Research Center was established and supported by the Danish National Research Foundation. This work was supported by the Danish Council for Independent Research (10-094816 to E.B.), the Lundbeck Foundation (R93-A8859 to E.B.), the Danish Cancer Society (A273 to S.F.P.), the Race Against Breast Cancer Foundation (to J.M.A.M.) and the Novo Nordisk Foundation (to S.F.P.).