Galanin Receptor 1 Has Anti-proliferative Effects in Oral Squamous Cell Carcinoma*

In the United States, oral cancer accounts for more deaths annually than cervical cancer, leukemias, or Hodgkin's lymphoma. Studies have shown that aberrations of chromosome 18q develop with tumor progression and are associated with significantly decreased survival in head and neck cancer patients. The G-protein-coupled receptor, galanin receptor 1 (GALR1), maps to this region of chromosome 18q. Although the role of GALR1 has been well characterized in neuronal cells, little is known regarding this receptor in non-neuronal cells. In this study, the expression, mitogenic function, and signaling mechanism of GALR1 are investigated in normal and malignant oral epithelial cells. mRNA expression was determined via reverse transcriptase-PCR. Protein quantification was done via immunoblot analysis and enzyme-linked immunosorbent assay. For functional and signaling studies, an inhibitory antibody was generated to the N-terminal ligand binding domain of GALR1. GALR1 protein and mRNA expression and GAL secretion were detected at variable levels in immortalized human oral keratinocytes and human oropharyngeal squamous cell carcinoma cell lines. Upon competitive inhibition of GALR1, proliferation was up-regulated in immortalized and malignant keratinocytes. Furthermore, studies with the inhibitory antibody and U0126, the MAPK inhibitor, show that GALR1 inhibits proliferation in immortalized and malignant keratinocytes by inactivating the MAPK pathway. GALR1s inhibitory effects on proliferation in epithelial cells raises the possibility that inactivation or disregulation of this receptor can lead to uncontrolled proliferation and neoplastic transformation.

In the United States, oral cancer accounts for more deaths annually than cervical cancer, leukemias, or Hodgkin's lymphoma. Studies have shown that aberrations of chromosome 18q develop with tumor progression and are associated with significantly decreased survival in head and neck cancer patients. The G-protein-coupled receptor, galanin receptor 1 (GALR1), maps to this region of chromosome 18q. Although the role of GALR1 has been well characterized in neuronal cells, little is known regarding this receptor in nonneuronal cells. In this study, the expression, mitogenic function, and signaling mechanism of GALR1 are investigated in normal and malignant oral epithelial cells. mRNA expression was determined via reverse transcriptase-PCR. Protein quantification was done via immunoblot analysis and enzyme-linked immunosorbent assay. For functional and signaling studies, an inhibitory antibody was generated to the N-terminal ligand binding domain of GALR1. GALR1 protein and mRNA expression and GAL secretion were detected at variable levels in immortalized human oral keratinocytes and human oropharyngeal squamous cell carcinoma cell lines. Upon competitive inhibition of GALR1, proliferation was up-regulated in immortalized and malignant keratinocytes. Furthermore, studies with the inhibitory antibody and U0126, the MAPK inhibitor, show that GALR1 inhibits proliferation in immortalized and malignant keratinocytes by inactivating the MAPK pathway. GALR1s inhibitory effects on proliferation in epithelial cells raises the possibility that inactivation or disregulation of this receptor can lead to uncontrolled proliferation and neoplastic transformation.
In the United States, oral cancer accounts for more deaths annually than cervical cancer, leukemias, or Hodgkin's lymphoma (1,2). Aggressive surgery and radiation treatments are physically and emotionally debilitating and are selected on the basis of tumor size and spread rather than on the biology of individual lesions. However, two tumors that receive the same treatment may vary dramatically in biologic behavior. Therefore, identification of the molecular events that predict the biologic behavior of individual head and neck squamous cell carcinomas (SCC) 1 will facilitate treatment selection, thereby reducing the morbidity of the disease.
One such prognostic biomarker has been identified on the long arm of chromosome 18q in head and neck SCC. Studies from Carey and others (3,4) have shown that aberrations of chromosome 18q develop with tumor progression and are associated with significantly decreased survival in head and neck cancer patients. The G-protein-coupled receptor (GPCR), galanin receptor 1 (GALR1), maps to this region of chromosome 18q (5,6). GPCRs activate GTP-binding proteins to trigger signaling cascades such as the mitogen-activated protein kinase (MAPK) pathway, a well established mitogenic pathway (7,8). Hence, GALR1 is a candidate protein for regulation of cell proliferation in head and neck SCC.
Galanin (GAL), a 30 amino acid neuropeptide, binds to galanin receptors to induce several regulatory functions in neuronal cells, including neuroregeneration, control of endocrine and exocrine secretions, and modulation of sensory and behavioral functions (9 -11). GAL has also been implicated in neuronal growth and development (12)(13)(14). For example, transgenic mice overexpressing GAL, develop pituitary adenomas (14). In contrast, mice with a loss of function mutation in the GAL gene show enhanced apoptosis of ganglion cells upon sectioning of the nerve (12). Moreover, GAL knock-out mice have a loss of one-third of the cholinergic neurons in the basal forebrain by 7 days of age (13). In the gut, GAL has an inhibitory role on circular muscle contraction (15) and exerts anti-proliferative effects on regenerating rat adrenal glands and immature rat thymocytes (16,17). Although these studies increase our understanding of galanin receptor function because of the complexity of the biological systems used, they do not provide mechanistic information of these exciting biological outcomes. The studies presented here investigate the intracellular signaling events responsible for these effects. GAL-mediated events occur via GALR1 as well as two other galanin receptors, GALR2 and GALR3. The latter receptors are products of two different genes on chromosomes 17q25.3 and 22q13.1, respec-tively (18). Although the role of GALR1 has been well characterized in neuronal cells, the expression of this GPCR in nonneuronal cells is poorly understood. In fact, few studies describe this receptor outside the neuronal system (19) and none investigates the mitogenic role of GALR1. Furthermore, previous studies were exclusively correlational or performed with exogenously expressed receptor.
In the current study, the expression, mitogenic function, and signaling mechanism of endogenous and exogenously expressed GALR1 are investigated in normal and malignant oral epithelial cells.
Galanin ELISA-Conditioned medium was collected from HOK16B, OSCC3 (23), and a panel of well characterized oropharyngeal UM-SCC cell lines (UM-SCC-11A, -11B, -14A, -14B, -17B, -74A, -81B, -22A, and  -22B). HOK and SCC cells were cultured to ϳ60 -70% confluence in 100-mm dishes. The culture medium was changed and harvested 24 h later. The medium was centrifuged at 10,000 rpm for 10 min to remove cellular debris, and the supernatant was designated the conditioned medium. The total cell number was determined by trypan blue enumer-ation. GAL secreted into the medium was measured by a competitive ELISA using a previously described protocol with modification (25). GAL (Sigma) was gently cross-linked with bovine serum albumin, a carrier protein, using glutaraldehyde. The coupled peptide and carrier were then used to coat 96-well plates (Corning). The conditioned medium from SCC culture plates and eight standard samples ranging from 0.05 to 500 ng/ml human GAL (Sigma) were combined with 1:50,000 rabbit anti-GAL polyclonal antibody (Chemicon International, Temecula, CA) and preincubated at 4°C for 2 h prior to being added to the GAL coated 96-well plates. The specificity of the GAL antibody used for the ELISA was verified previously (26). Following an overnight incubation, the plates were washed, secondary antibodies were added for 2 h, and color development was accomplished using o-phenylenediamine (Sigma) as a substrate. Finally, plates were read at 490 nm with a microplate reader. The principle of this assay was that GAL in the conditioned medium would competitively inhibit the binding of GAL antibody to the GAL substrate.
Generation of N-terminal GALR1 Antibody-An N-terminal peptide from GALR1 was selected based on its location in the first extracellular domain and high antigenicity score. The peptide was synthesized in The University of Michigan Protein Core, and two rabbits were immunized four times. Sera were assessed for specific binding to the peptide using a non-competitive ELISA.
Proliferation Assay-Rabbit anti-GAL polyclonal antibody or rabbit anti-GALR1 polyclonal antibody was used to investigate the role of GAL and GALR1, respectively, on cell proliferation. HOK16B, UM-SCC-22A, and UM-SCC-17B cells (1.5-3.0 ϫ 10 4 ) were cultured overnight in a 24-well plate in KGM or DMEM. The following day, these cells were treated with medium containing anti-GAL or anti-GALR1 antibodies or vehicle control for 24 h. Each antibody dilution was normalized to 1:50 rabbit serum concentration. Each treatment was performed in triplicate. Cells in three wells were counted prior to the incubation. After 24 h, total cell number was determined by the trypan blue enumeration assay.
Western Blot Analysis-The cell lines were washed once with ice-cold phosphate-buffered saline and lysed in 0.2% Nonidet-P40 lysis buffer containing protease inhibitors on ice for 10 min, as described previously (21). Cell lysates were scraped into Microfuge tubes and left for an additional 10 min with occasional vortexing, and particulate material was pelleted by centrifugation at 15,700 ϫ g for 10 min at 4°C. The supernatant was collected, and protein content was measured by the Bio-Rad protein assay. Equal amounts of protein (30 g) were electrophoresed on SDS-12% PAGE gels and transferred to nitrocellulose membranes (Schleicher & Schuell). The membranes were incubated in 5% nonfat dry milk in Tris-buffered saline for 1 h at room temperature to block nonspecific binding. Membranes were incubated with the primary antibody for 2 h at room temperature or overnight at 4°C. Primary antibody concentrations were as follows: rabbit anti-GALR1 polyclonal antibody (1:1000, generated against the N terminus of GALR1);
MAPK activation was evaluated following incubation of immortalized keratinocytes (HOK16B cells) and UM-SCC-11A cells in serum-free medium with 1:50 concentration of GALR1 polyclonal antibody or with 1:50 normal rabbit serum. For MAPK inhibition experiments, cells were preincubated in serum-free medium for 6 h prior to treatment with 10 nM U0126 for 24 h (Sigma). For experiments involving GAL stimulation, cells were stimulated with 150 nM GAL for 10 min following a 6-h preincubation in serum-free medium. Following standard Western blotting techniques as described above, the activation of the MAPK pathway was evaluated using anti-active MAPK (ERK1/2) (1:2000, Promega).

FIG. 4. An antibody generated to the N-terminal ligand binding domain of GALR1 is a competitive inhibitor of ligand binding. A,
293T cells were transfected with 5.0 g of pcDNA or 2.5 or 5.0 g of pcDNA-GALR1. Whole cell protein lysates were separated by PAGE and blotted with N-terminal GALR1, commercial GALR2, or GALR3 antibodies. B, HOK16B cell lysate was electrophoresed, transferred, and blotted with anti-GALR1 antibody (left panel) or anti-GALR1 antibody preincubated with GALR1 peptide (10 or 20 g/ml) prior to Western blotting (middle and right panels, respectively). C, 293T cells were transfected with pcDNA or pcDNA-GALR1 and treated with 150 nM galanin in the presence or absence of N-terminal GALR1 antibody at 24 h post-transfection. Cell lysates were electrophoresed and blotted with anti-GALR1 antibody. A duplicate filter was blotted with anti-active ERK and subsequently with anti-total ERK monoclonal antibodies. Ab, antibody.
Total RNA Isolation-RNA was isolated using the TRIzol reagent according to the manufacturer's instructions. cDNA was synthesized from 1 g of RNA using the reverse transcription system kit (Promega).
Transfections-The pcDNA 3.1-GALR1 construct was obtained from The University of Missouri-Rolla cDNA Resource Center. The pcDNA 3.1 empty vector was used as a control for transfection effects on endogenous gene expression. UM-SCC-11A or 293T cells were seeded at 1 ϫ 10 6 cells/60-mm dish and transfected with pcDNA 3.1-GALR1 or pcDNA 3.1 using Lipofectamine 2000 (Invitrogen) according to the manufacturer's instructions. At 24 h post-transfection, cells were treated as indicated in figure legends and whole cell lysates were prepared.
Statistical Analysis-All of the statistical comparisons were performed using the Student's t test, and experiments were done in triplicate. A p value of Յ0.05 was used as a measure of significance.

Galanin Is Secreted by Human Oropharyngeal
Keratinocytes-Although the growth effects of GAL have been studied in neuronal and non-neuronal cells (12)(13)(14), the role of GAL and its receptors in epithelial cell proliferation has not been explored. Hence, in this study, initial experiments investigated the expression and secretion of GAL by normal and malignant oral epithelial cells. Using reverse transcriptase-PCR, GAL mRNA was identified in immortalized oral keratinocytes (HOK16B) and oropharyngeal SCC cell lines including UM-SCC-17B, UM-SCC-11A, UM-SCC-22B, and UM-SCC-14A (Fig. 1A). As expected, the human brain, the positive control, showed a signal of similar molecular mass. The water control was appropriately negative. In order for GAL to stimulate the GALRs, it must be secreted. Secretion of the 3-kDa GAL neuropeptide from immortalized and malignant keratinocytes was investigated using a competitive ELISA (Fig. 1B). HOK16B and all of the SCC cell lines secreted GAL. Secretion was statistically greater in five SCC cell lines (17B and 14A (p ϭ 0.05) and 11A, 14B, and 74A (p ϭ 0.01)) when compared with the non-malignant HOK16B cells. In conditioned medium collected at 24 h, GAL secretion in UM-SCC-11A, UM-SCC-11B, UM-SCC-14A, UM-SCC-14B, UM-SCC-17B, UM-SCC-74A, and UM-SCC-81B consisted of a minimum of 175 ng/ml/million cells, whereas HOK16B, UM-SCC-22A, UM-SCC-22B, and OSCC3 secreted less than 50 ng/ml (Fig. 1B). Thus, GAL is transcribed in and secreted by non-malignant and malignant keratinocytes.
Galanin Up-regulates Proliferation in Immortalized and Malignant Keratinocytes-Several studies have shown that GAL has an autocrine mitogenic effect on neuronal cells and modulates regeneration and survival in these cells (12-14, 27). Hence, we hypothesized that if GAL has an autocrine mitogenic effect in oral epithelial cells, an anti-GAL antibody should

FIG. 5. Proliferation and MAPK activation are increased in HOK16B cells at 24 h following treatment with anti-GALR1 antibody.
A, HOK16B cells were incubated with anti-GALR1 antibody, and cell number was determined by the trypan blue enumeration assay. rbt, rabbit; ctrl, control. B, HOK16B cells were treated with 1:50 concentration of anti-GALR1 antibody in the absence or presence of GALR1 peptide (10 g/ml) for 24 h. C, anti-GALR1-treated HOK16B cells were lysed and electrophoresed. MAPK activation was evaluated by blotting with anti-active ERK and anti-total ERK monoclonal antibodies (Ab). compete for secreted GAL, thereby inhibiting proliferation. To test this hypothesis, HOK16B and two cancer cell lines, UM-SCC-22A and UM-SCC-17B, were incubated with varying concentrations of anti-GAL antibody for 24 h. Cell proliferation was determined by the trypan blue enumeration assay. These cell lines were selected, because they represent low and intermediate levels of GAL secretion, i.e. Ͻ50 ng/ml and Ͼ175 ng/ml GAL, respectively (Fig. 1B). All three cell lines exhibited a reduction in cell number with anti-GAL antibody; however, compared with control-untreated cells, HOK16B proliferation at 24 h was not significantly inhibited at any of the antibody concentrations tested (Fig. 2). In contrast, both of the cancer cell lines exhibited a dose-dependent inhibition of proliferation that was also significantly different from control at the highest antibody concentration (1:50,000). At this concentration, UM-SCC-22A exhibited a 38% inhibition (Ϯ3.9), whereas UM-SCC-17B, which secretes more GAL, exhibited a 63% (Ϯ3.0) inhibition. Thus, we conclude that GAL up-regulates proliferation in malignant keratinocytes.
GALR1, GALR2, and GALR3 Are Expressed in Immortalized and Malignant Keratinocytes-Three galanin receptor isoforms, GALR1, GALR2, and GALR3, have been identified in neuronal cells. The pro-proliferative effects of GAL on human keratinocytes suggested that these cells express at least one of the three receptors. To verify this hypothesis, the expression of these three receptors in immortalized and malignant oral keratinocytes was investigated by reverse transcriptase-PCR and immunoblot analysis. GALR1, GALR2, and GALR3 mRNAs were identified in non-malignant (HOK16B) and malignant (UM-SCC-17B, -11A, -22B, and -14A) oral keratinocyte cell lines (Fig. 3A, top, middle, and bottom panels, respectively). Signals of corresponding molecular mass were detected in the human brain, which was used as a positive control. GAPDH served as an internal control. Protein expression was confirmed using isoform-specific galanin receptor antibodies (Fig. 3, B-D). GALR1-immunoreactive peptides were detected at a molecular mass of ϳ70 kDa in all of the cell lines tested (Fig. 3B, top  panels). However, when compared with the corresponding GAPDH signals that served as loading controls, it is apparent that HOK16B and UMSCC17B strongly express GALR1, whereas signals of lesser intensity were detected in the other SCC cell lines tested (Fig. 3B, top panels). GALR2-immunoreactive peptides were detected at a molecular mass of ϳ60 kDa, primarily in HOK16B, and to a lesser extent in UM-SCC-22B and UM-SCC-11A (Fig. 3C, top panels). GALR3 (ϳ55 kDa) was strongly expressed in UM-SCC-17B as well as UM-SCC-11A and UM-SCC-22B (Fig. 3D, top panels). Thus, GAL and all three of its receptors are variably expressed in non-malignant and malignant human oropharyngeal keratinocytes.
An Antibody to the N Terminus of GALR1 Acts as a Competitive Inhibitor of Receptor Activation-The GALR1 gene has been mapped to the chromosome 18q23 region, which has been implicated in the progression of head and neck SCC. Thus, to investigate GALR1 function and signaling in immortalized and malignant oral keratinocytes, an antibody was generated to the N terminus, the ligand binding domain of GALR1, to serve as a competitive inhibitor of ligand binding. This strategy was adopted, because GALR1-specific agonists or inhibitors were unavailable. A peptide representing the ligand binding domain of the extracellular N-terminal region was synthesized and inoculated into rabbits to generate a GALR1-specific polyclonal antibody. Antibody specificity was verified by immunoblot analysis of whole cell lysates of 293T cells transfected with full-length GALR1 cDNA and HOK16B cells (Fig. 4). As expected, in 293T cells, the antibody detected a peptide in cells transfected with increasing concentrations of GALR1 cDNA, whereas no signal was detected in the vector control (pcDNA). This signal is specific for GALR1 and was not detected by antibodies to GALR2 or GALR3 (Fig. 4A). In HOK16B cells, the N-terminal antibody detected a single band at 70 kDa (Fig. 4B,  left panel), which was not detected when the antibody was preincubated with increasing concentrations of the immunogenic peptide prior to Western blotting (Fig. 4B, middle and  right panels).
Antibodies to receptors may have a stimulatory or inhibitory effect on receptor activation (28). To investigate whether the N-terminal GALR1 antibody inhibited receptor signaling, 293T cells were transfected with GALR1 and stimulated with GAL in the presence or absence of GALR1 antibody. Because stimulation of GALR1 facilitates MAPK activation (29 -32), we investigated whether the GAL-induced MAPK effects in 293T cells were blocked by preincubation with the N-terminal GALR1 antibody (Fig. 4C). In 293T cells transfected with GALR1 as a model for GALR1 signaling, GAL stimulated ERK activation (Fig. 4C), whereas in empty vector (pcDNA) transfected cells, GAL did not induce ERK activation. Furthermore, MAPK activation was blocked when GALR1 cells were preincubated with the N-terminal GALR1 antibody prior to GAL treatment (Fig. 4C). Thus, the N-terminal GALR1 antibody specifically detects GALR1 and inhibits receptor activation.
GALR1 Inhibits Proliferation in HOK16B and UMSCC11A-Oropharyngeal SCC cells secrete GAL, which binds to GALR1, GALR2, and GALR3, all of which are variously expressed in SCC cell lines (Fig. 3). To identify the GALR1-specific effects, the antibody generated to the N-terminal ligand binding domain of GALR1 (Fig. 4) was used as a competitive inhibitor to investigate the functional role of this receptor in human keratinocytes. Because GALR1 has a proliferative role in neuronal cells (12)(13)(14), subsequent studies focused on a possible proliferative function for this receptor in SCC cells. The treatment of HOK16B with increasing concentrations of GALR1 rabbit polyclonal antiserum showed a dose-dependent increase of total cell number when compared with the preimmune rabbit serum control (Fig. 5A). This increase was significantly different from the control at a 1:50 antibody dilution. Because varying serum concentrations in the different treatment groups can contribute to pro-mitogenic effects, serum content in all of the treatment groups was normalized with preimmune serum. Preincubation of the antibody with the immunogenic peptide abrogated the increase in the cell number observed in cells treated with GALR1 antibody alone (Fig.  5B). In cells treated with non-immune rabbit serum and peptide, the GALR1 N-terminal peptide also induced a slight but statistically insignificant increase in cell number. We postulate that the peptide may exert some effect by competitively binding endogenously secreted GAL. The MAPK pathway is a significant mitogenic pathway in oropharyngeal keratinocytes (33). Hence, we investigated whether the pro-proliferative effects of GALR1 antibodies correlated with MAPK activation. HOK16B were treated as above with GALR1 antibody (1:50), and whole cell lysates were prepared. As shown in Fig. 5C, in the presence of anti-GALR1 antibody, there was an increase in active phospho-ERK when compared with the rabbit serum control. In contrast, total ERK and GAPDH, which were used as loading controls, were unchanged. It is unlikely that ERK activation was due to a disparity in serum concentration, because the serum concentration was equivalent in both treatment groups. Thus, GALR1 inhibits proliferation by inhibition of MAPK activation.
Incubation of UM-SCC-11A, a cell line with very high GAL secretion but low GALR1 expression, behaved similarly. It exhibited a dose-dependent increase in cell number with increasing concentrations of GALR1 antibody (Fig. 6A) and ERK activation (Fig. 6B) when compared with the serum control. This effect plateaued at a 1:50 antibody dilution. Thus, in both HOK16B (Fig. 5B) and UMSCC11A cells (Fig. 6B), the MAPK pathway was activated following inhibition of GALR1, consistent with the anti-mitogenic function for GALR1 observed in human oral keratinocytes (Figs. 5A and 6A).
Chinese Hamster Ovary (CHO) Cells Are Unaffected by GALR1 Inhibition-Previous studies have shown that CHO cells do not express GALR1 (34). These findings were confirmed using the anti-GALR1 antibody, which detected no signal in CHO cell lysates (Fig. 7A). to verify the specificity of the inhibitory effects of the N-terminal antibody on proliferation, CHO cells were treated using the same protocol described above for HOK16B and UMSCC11A. As expected, in cells that do not express GALR1, the anti-GALR1 antibody did not affect proliferation (Fig. 7B). Thus, the N-terminal GALR1 antibody specifically inhibits the GALR1 receptor in GALR1-expressing cells.
GALR1 Inhibits Keratinocyte Proliferation via the MAPK Pathway-We observed that oral keratinocyte cell proliferation is stimulated by GALR1 inhibition and that this correlates with the activation of the MAPK pathway in these same cells. To directly investigate whether GALR1 inhibits MAPK activation in SCC, UM-SCC-11A cells were transfected with GALR1 or pcDNA control vector. Transfected cells were subsequently treated with GAL in the presence or absence or anti-GALR1 antibody. ERK activation was determined by immunoblot analysis (Fig. 8A) and quantified by densitometry. ERK activation was significantly decreased in untreated or GAL-treated GALR1-transfected cells (Fig. 8, A and B). As predicted based on the results in Fig. 6, when GALR1 cells were preincubated with the anti-GALR1 antibody prior to GAL treatment, ERK was significantly activated compared with vector controls (Fig.  8, A and B). These findings are consistent with an inhibitory effect of GALR1 on MAPK activation (Fig. 8, A and B). To verify the role of the MAPK pathway in GALR1 signaling and regulation of cell growth in these tumor cells, a specific MAPK inhibitor, U0126, was used in anti-GALR1 antibody-treated HOK16B and UM-SCC-11A cells. Proliferation and MAPK activation for both cell lines were investigated. Following a 24-h anti-GALR1 antibody incubation, HOK16B and UMSCC11A cell numbers were increased (Fig. 9, A and C, respectively). However, preincubation with U0126 abrogated this up-regulation of proliferation. ERK activation was increased in HOK16B cells and UM-SCC-11A cells as described above; however, when U0126 was added, ERK activation (Fig. 9, B and D, respectively) returned to control levels. DISCUSSION In this study, we show that GAL is secreted by oropharyngeal keratinocytes and inhibition of GAL binding in both nonmalignant and malignant (SCC) keratinocytes decreases proliferation. This is the first report showing GAL and galanin receptor expression and function in epithelial cells of the oropharyngeal region and one of a very few reports characterizing the effects exerted by these important proteins at the cellular level. GAL distribution has been fairly well characterized in the central nervous system (10). High concentrations of GAL (1-5 pmol/mg protein) have been detected in the hippocampus, amygdala, nucleus accumbens, and hypothalamus (10,35). GAL has also been shown in a number of peripheral organs, such as the pituitary, pancreas, respiratory tract, and gastrointestinal tract (10,35). Variable GAL secretion by individual SCC cells may play a role in the regulation of proliferation in these cells. Consistent with this hypothesis, we observed that anti-GAL antibody treatment resulted in the inhibition of proliferation in GAL-secreting head and neck SCC cells. In the gastrointestinal tract, GAL is secreted by enteric nerves where it exerts a paracrine regulation of intestinal smooth muscle contraction and relaxation as well as regulation of pancreatic secretions (19,36,37). Because keratinocytes secrete GAL, it is likely that normal cells utilize GAL in an autocrine loop to regulate cell proliferation. This regulation may also be mediated by GAL secretion from neuronal tissue that lies in proximity to the oral epithelium. Further characterization of GAL secretion and the functional impact of varying endogenous levels may help us to better understand disregulated proliferation in malignant keratinocytes.
Growth regulation may be disrupted in SCC cells as a result of differential galanin receptor expression. A number of reports have implicated 18q loss of heterozygosity (LOH) in the progression of head and neck SCC (3, 4). Tumor progression following a chromosomal loss would be consistent with a tumor suppressor hypothesis where mutation of one allele and a loss of another may lead to the loss of function of a potential tumor suppressor gene. This loss of function could have a profound impact on disregulated proliferation in head and neck SCC. Takebayashi et al. (38) identified the important 18q loci lost in head and neck cancer and speculated that tumor suppressor gene(s) were located in this region. Therefore, the identification of this gene(s) would help elucidate the mechanism by which head and neck SCC development and progression occur. GALR1 was identified as a gene within the minimally lost region with the highest LOH frequency in head and neck SCC (38). Because GALR1 regulates neuronal growth and development, it is an excellent candidate for growth regulation in oropharyngeal epithelial cells.
Our results in non-malignant and malignant keratinocytes consistently point to a growth suppressive function for GALR1. In UM-SCC-11A, a cell line that does not exhibit 18q LOH, and in non-malignant keratinocytes, GALR1 inhibits proliferation. Wang et al. (29) show that GALR1 induces the G i family of heterotrimeric G-proteins to inhibit cAMP and stimulate MAPK activity, effects that are sensitive to pertussis toxin. However, GALR1 may have pro-or anti-proliferative effects that vary by cell type (31,32). Furthermore, previous studies were correlational or performed exclusively with exogenously expressed receptor. For example, Berger et al. (32) found that, in neuroblastoma cells, GALR1 is anti-mitogenic. In contrast, McDonald et al. (31) correlated GAL secretion with GALR1 expression in cartilage growth plates and fracture repair sites.
In the studies presented here, we used a GALR1-specific antagonist to target endogenous GALR1 and exogenously expressed GALR1. We show that GALR1 inhibits proliferation in non-malignant and malignant keratinocytes via inhibition of MAPK. To verify that the GALR1 antibody inhibited GALR1 signaling, 293T cells were transfected with wild type GALR1 and stimulated with GAL in the presence or absence of GALR1 antibody. In this setting, GAL stimulated MAPK activation and the GALR1 antibody blocked this effect of GAL on exogenously expressed GALR1. This disparity in response between 293T cells and keratinocytes in which GALR1 inhibits MAPK supports the concept of highly specialized alternative mechanisms to activate the MAPK pathway by GPCRs in distinct cellular settings (7).
GAL engages GALR1, GALR2, and GALR3, all of which are expressed in SCC (Fig. 3). Therefore competitive inhibition of GAL binding with the anti-GAL antibody reveals the net effect of all three receptors. In contrast, the GALR1 antibody specifically inhibits GALR1, showing that GALR1 has anti-proliferative effects in keratinocytes (Figs. 5 and 6). Stimulation of all three receptors by GAL leads to proliferation, suggesting that GALR2 and/or GALR3 is/are pro-proliferative (Fig. 10). Consistent with these findings, previous studies with GALR2 (39) and recent findings in keratinocytes 2 show that GALR2 is pro-proliferative. Therefore, GAL-induced proliferation may be delicately regulated by the ratio of galanin receptors expressed in a given cell. Overexpression of GALR2 in the presence of normal or reduced GALR1 expression may lead to unchecked cell proliferation (Fig. 10). Alternatively, normal GALR2 function and a loss of GALR1 function could generate the same effect (Fig. 10).
Additionally, disregulation of signaling proteins downstream from individual receptors may also affect this regulation. Consistent with this speculation, Wittau et. al. (39) show that activation of GALR2 stimulates G 12/13 to induce MAPK activation and clonal growth of small cell lung cancer cells. Thus, GAL effects may be regulated at the level of the receptor or by differential intracellular signaling mechanisms (29).
Future studies will focus on the outcomes of differential galanin receptor activation as well as the initiation of downstream signaling molecules following binding of GAL to GALR1 and GALR2.