G-Protein Coupled Receptor Family C, Group 5, Member A (GPRC5A) Expression Is Decreased in the Adjacent Field and Normal Bronchial Epithelia of Patients with Chronic Obstructive Pulmonary Disease and Non–Small-Cell Lung Cancer

Abstract

Introduction

Understanding oncogenes and tumor suppressor genes expression patterns is essential for characterizing lung cancer pathogenesis. We have previously demonstrated that mGprc5a/hGPRC5A is a lung-specific tumor suppressor evidenced by inflammation-mediated tumorigenesis in Gprc5a-knockout mice. The implication of GPRC5A in human lung cancer pathogenesis, including that associated with inflammatory chronic obstructive pulmonary disease (COPD), a risk factor for the malignancy, remains elusive.

Methods

We sought to examine GPRC5A immunohistochemical expression in histologically normal bronchial epithelia (NBE) from lung disease-free never- and ever-smokers (n = 13 and n = 18, respectively), from COPD patients with (n = 26) and without cancer (n = 24) and in non-small cell lung cancers (NSCLCs) (n = 474). Quantitative assessment of GPRC5A transcript expression in airways (n = 6), adjacent NBEs (n = 29) and corresponding tumors (n = 6) from 6 NSCLC patients was also performed.

Results

GPRC5A immunohistochemical expression was significantly lower in tumors compared to uninvolved NBE (p < 0.0001) and was positively associated with adenocarcinoma histology (p < 0.001). GPRC5A airway expression was highest in lung disease-free NBE, decreased and intermediate in NBE of cancer-free COPD patients (p = 0.004) and further attenuated and lowest in epithelia of COPD patients with adenocarcinoma and SCC (p < 0.0001). Furthermore, GPRC5A mRNA was significantly decreased in NSCLCs and corresponding NBE compared to uninvolved normal lung (p = 0.03).

Conclusions

Our findings highlight decreased GPRC5A expression in the field cancerization of NSCLC, including that associated with lung inflammation. Assessment of the use of GPRC5A expression as a risk factor for NSCLC development in COPD patients is warranted.

Lung cancer, the majority of which is non–small cell lung cancer (NSCLC), is the leading cause of deaths in the United States and worldwide.¹ The high mortality associated with lung cancer is in part because of late diagnosis after regional or distant spread of the disease.² Improved clinical management of NSCLC is tightly linked to the identification of new effective early biomarkers that can spear novel strategies for early detection, prevention, and treatment.^2,3

The majority (85%) of diagnosed NSCLC cases are attributable to cigarette smoking.^4,5 Auerbach et al.⁶ earlier showed that tobacco carcinogen exposure causes multifocal and clonal histopathological changes in the airway epithelia of smokers, suggestive of a field cancerization in the lung. In addition, cigarette smoking perpetuates inflammation throughout the smoking-exposed airway epithelia of heavy smokers,⁷ which was suggested, to lead to onset of lung cancer development.⁸ Chronic obstructive pulmonary disease (COPD) of the lung is an inflammatory condition that is, like lung cancer, causally linked to cigarette smoking^9,10 and is a major cause of mortality in the United States.¹¹ Moreover, preinvasive lung cancer lesions are common (approximately 50%) in airways of COPD patients.¹² Importantly, although phenotypically healthy smokers comprise a significant population at risk for lung cancer,^5,7 COPD increases lung cancer risk by 4.5-fold, more than either age or quantity of smoking.^11,13,14 In addition, 33% of patients with mild-to-moderate COPD eventually die because of lung cancer burden.¹¹ Although smoking is the main cause of both diseases, only 10% to 15% of smokers develop COPD and/or lung cancer, suggesting that other factors, including genetic variation and expression, may differ across individuals in response to cigarette smoke.^4,5,9,10 It is plausible to assume that understanding common expression patterns between both diseases for subsequent identification of biological markers that explain risk of disease onset and/or progression will favorably promote the clinical management of COPD-associated lung cancer.¹¹

Our group has previously demonstrated that loss of G-protein coupled receptor, family C, group 5, member A (mGprc5a/hGPRC5A), exemplified by the Gprc5a-knockout mouse model, leads to chronic inflammation and spontaneous lung tumor development.^15,16 However, the expression of this tumor suppressor in human lung NSCLCs, including those associated with inflammatory conditions such as COPD, is unknown. We investigated the immunohistochemical expression of GPRC5A in a large series of NSCLC histological tissue specimens and in normal bronchial epithelia (NBE) from lung-disease-free individuals, COPD patients and from COPD patients with lung cancer and its transcript expression in independent resected normal epithelia and corresponding tumors. Our findings reveal marked reduced GPRC5A expression in human NSCLC and in the histologically normal field cancerization associated with COPD and lung cancer. Our study pinpoints to a potential role of this tumor suppressor in the progression of COPD-associated NSCLC, which warrants further studies to assess its use as a risk marker for this disease.

MATERIALS AND METHODS

NSCLC Tissue mMcroarrays and Normal Bronchial Epithelial Specimens

Tissue microarrays (TMAs) used in this study comprised 474 surgically resected NSCLC tumor specimens (308 adenocarcinomas, 166 squamous cell carcinomas [SCCs]) collected under an Institutional Review Board protocol and archived as formalin-fixed paraffin-embedded (FFPE) specimens in the University of Texas Specialized Program of Research Excellence thoracic tissue bank at the University of Texas M.D. Anderson Cancer Center. Clinicopathological features of the NSCLC patients examined are summarized in Table 1. TMAs were prepared with a manual tissue arrayer (Advanced Tissue Arrayer ATA100, Chemicon International, Temecula, CA) using 1-mm diameter cores in triplicate for tumors, as described previously.¹⁷ Histological sections 4 μm in thickness were then prepared for subsequent immunohistochemistry analysis. FFPE specimens of NBE from 50 patients with COPD with forced expiratory volume in 1 second/forced vital capacity ratio of 70% or less, a smoking history of 30 or more pack-years, and collected at Baylor college of Medicine (Houston, TX) were included in the study, including 24 cancer-free cases and 26 cases with NSCLC. In addition, NBE specimens from never smokers (n = 13) and ever smokers (n = 18) without lung cancer and with no or mild airway obstruction status were analyzed.

TABLE 1.

Clinicopathological Features of NSCLC Patients in Tissue Microarray Sets Used in the Study

Covariate	Levels	n (%)
Sex	Female	240 (50.6)
	Male	234 (49.4)
Histology	Adenocarcinoma	308 (65.0)
	Squamous cell carcinoma	166 (35.0)
Stage	I	302 (63.7)
	II	86 (18.1)
	III or IV	86 (18.1)
Grade*	Well	36 (10.4)
	Moderate	193 (55.9)
	Poor	116 (33.6)
Smoking status	Current	178 (37.6)
	Former	229 (48.3)
	Never	67 (14.1)
Tobacco history	No	67 (14.1)
	Yes	407 (85.9)

^*Information on differentiation grade was available in a subset of NSCLCs analyzed. NSCLC, non-small-cell lung cancer.

Immunohistochemical Analysis

Polyclonal antirabbit antibodies raised against human GPRC5A were generated by peptide synthesis and site-directed carrier conjugation using keyhole limpet hemocyanin as a custom service by Zymed Laboratories Inc. (South San Francisco, CA). The synthetic peptide (cysteine)-PSPYKDYEVKKEGS-COOH, corresponding to amino acids 344–357 in the human GPRC5A C-terminus, was covalently linked to keyhole limpet hemocyanin via an added cysteine residue, and polyclonal antipeptide antibodies were prepared in rabbits. Sera were confirmed to contain high-titer antibodies against the specific peptide using enzyme-linked immunosorbent assays. TMAs and the histologic sections (4 μm thick) of surgical resected specimens were deparaffinized and hydrated, and antigen retrieval was performed using a decloaker with Dako target retrieval system at a pH of 6.0 (Dako North America, Inc., Carpinteria, CA). Intrinsic peroxidase activity was blocked by 3% methanol and hydrogen peroxide for 12 minutes and serum-free protein block (Dako) was used for 30 minutes for blocking nonspecific antibody binding. Slides were then incubated with the antibodies against human GPRC5A (1:200 dilution) at 4°C overnight. After three washes in Tris-buffered saline, slides were incubated for 30 minutes with Dako Envision + Dual Link at room temperature. After three additional washes, slides were incubated with Dako chromogen substrate for 5 minutes and were counterstained with hemotoxylin for another 5 minutes. FFPE whole-section specimens, except for the omission of the primary antibodies, were used as negative controls. The intensity and extent of cytoplasmic and nuclear GPRC5A immunostaining were evaluated using a light microscope (magnification, × 20) independently by two pathologists (JF and IIW). GPRC5A immunoreactivity was mainly cytoplasmic, which was quantified using a four-value intensity score (0, none; 1, weak; 2, moderate; and 3, strong) and the percentage (0–100%) of the extent of reactivity. A final cytoplasmic expression score was obtained by multiplying the intensity and reactivity extension values (range, 0–300).

GPRC5A Transcript Expression in Resected NSCLC Specimens and Adjacent Airway Epithelia

GPRC5A mRNA expression was assessed in ever-smoker NSCLCs and their uninvolved normal lung parenchyma tissues as well as in NBE collected by brushing of multiple corresponding airways from the same lobectomy- and pneumonectomy-resected specimens. Tumor tissues from six ever-smoker NSCLCs and normal lung specimens (n = 6) as well as NBE by brushings (n = 29) from the same patients were obtained under an Institutional Review Board approved protocol in which specimens were collected from patients who had signed letters of consent. NBE were collected from multiple adjacent to tumor airways by brushing each site independently using Cytosoft cytology brushes (Cardinal Health, Dublin, OH). Parallel brushes were used for touch-prep for cytological assessment by pan-cytokeratin staining, which revealed epithelial content more than 90%. Normal histology was determined by hematoxylin and eosin staining. Brushes were immediately placed in Qiazol lysis reagent (Qiagen, Valencia, CA) and stored in –80°C until further processing. NSCLC and normal lung specimens were shaved for histological assessment of percentage of tumor content and malignant cells and for corresponding RNA isolation. Total RNA from all samples was purified using the miRNeasy kit (Qiagen) according to the manufacturer's instructions. RNA quantity was determined using the NanoDrop spectrophotometer, (Thermo Scientific, Wilmington, DE), and quality was determined by analysis of RNA integrity with Agilent Bioanalyzer 2000 (Agilent Technologies, Santa Clara, CA).

Quantitative-Real Time Polymerase Chain Reaction

Total RNA (150 ng) was reverse-transcribed using the high-capacity RNA-to-cDNA kit (Life Technologies, Carlsbad, CA) according to the manufacturer's instructions. Quantitative real-time polymerase chain reaction (PCR) was performed using TaqMan® (Applied Biosystems, Foster City, CA) gene expression assays for GPRC5A (Hs00173681_m1) and beta-actin (ACTB) (Hs99999903_m1) primers (Life Technologies) on a 7900HT Fast-Real-Time PCR System (Life Technologies) according to the manufacturer's instructions. All samples were done in triplicates and normalized to ACTB. Relative quantification was calculated using the comparative cycle threshold method as previously described.¹⁸

Statistical Analysis

Summary statistics, including frequency tabulation, means, standard deviations, median, and range, were given to describe patient characteristics. Wilcoxon rank sum test or Kruskal–Wallis test was used to compare GPRC5A immunohistochemical expression between two levels or among more than two levels, when appropriate. General linear model was applied to test the difference of GPRC5A expression across NBE from COPD- and cancer-free never and ever smokers, cancer-free COPD patients, as well as NBE from patients with both COPD and cancer. Repeated measures analysis was used to determine significance of GPRC5A transcript variation across matched NSCLCs, airways, and uninvolved normal lung. All statistical tests were two-sided, and p values of 0.05 or less were considered to be statistically significant. Statistical analysis was performed with standard statistical software, including SAS Release 9.1.3 for Windows and S-Plus 8.0 for Windows.

RESULTS

GPRC5A Immunohistochemical Expression in Human NSCLC

We have previously demonstrated that mGprc5a/hGPRC5A is a lung-specific tumor suppressor evidenced by spontaneous tumor development in mice with knockout of this gene as well as loss of the transformed phenotype in malignant NSCLC cell lines after GPRC5A overexpression.¹⁶ We sought to examine the immunohistochemical expression patterns of GPRC5A tumor suppressor in NSCLC, which is unknown, and its association with clinicopathological features of the disease, including the two major histologic types examined, adenocarcinoma and SCC. The clinicopathological characteristics of the NSCLC patients from which FFPE tumor specimens were analyzed are detailed in Table 1. Representative photomicrographs of GPRC5A immunohistochemical expression in human lung adenocarcinoma (left) and SCC (right) are depicted in Figure 1A. GPRC5A immunoreactivity was mainly cytoplasmic. GPRC5A expres sion was significantly lower in the adenocarcinomas (n = 308; mean, 48.91 ± 36; median, 41.67; minimum, 0; maximum, 180) and SCCs (n = 166; mean, 23.47 ± 22.79; median, 18.33; minimum, 0; maximum, 110) examined compared with the uninvolved normal bronchial epithelia (mean, 164.52 ± 34.04; median, 160; minimum, 100; maximum, 230) from lung-disease–free smokers and never smokers (p < 0.0001) (Fig. 1B). Moreover, GPRC5A immunohistochemical expression was significantly lower in SCCs compared with that in adenocarcinomas (p < 0.0001) (Fig. 1B). We then correlated GPRC5A with other clinicopathological features for all NSCLCs. Notably, GPRC5A expression was significantly increased in NSCLC tumor specimens from never smokers (n = 67) (mean, 53.94 ± 39.94; median, 40; minimum, 0; maximum, 163.33) compared with ever smokers (n = 407) (mean, 37.71 ± 32.65; median, 30; minimum, 0; maximum, 180) (p = 0.001) (Fig. 1C). GPRC5A expression was also significantly different among never, former and current smoker NSCLCs (p = 0.003) (Fig. 1D). In addition, GPRC5A expression was significantly positively associated with well-differentiated tumor grade in a subset of NSCLC tumors (n = 346) examined with available grade information (p = 0.004) (Fig. 1E). It is important to note that when we examined each histology type separately, there were no statistically significant correlations between GPRC5A expression and patient's smoking history and status, and tumor-differentiation grade.

FIGURE 1.

GPRC5A immunohistochemical expression in NSCLC histological tissue specimens. A, Representative photomicrographs of GPRC5A immunohistochemical expression in lung adenocarcinoma (left) and SCC (right) observed under a light microscope at a magnification of 20 ×. Bars, 100 μm. B, Box-plots depicting differences in GPRC5A immunohistochemical expression among NBE, Adeno, and SCCs. C, Box-plots depicting GPRC5A expression based on tobacco history (never versus ever smokers), (D), smoking status (never, former and current), and (E), differentiation grade (well, moderate, and poor). p values based on the Wilcoxon rank sum test and Kruskal–Wallis test. GPRC5A, G-protein coupled receptor family C, group 5, member A; NSCLC, non–small-cell lung cancer; SCC, squamous cell carcinomas; NBE, normal bronchial epithelia; Adeno, adenocarcinomas.

GPRC5A Immunohistochemical Expression in NBE from COPD Patients

We have previously shown that loss of mGprc5a/hGPRC5A tumor suppressor leads to chronic-inflammation–mediated prosurvival signaling and transformation of nonmalignant lung epithelial cells.¹⁵ The relevance of GPRC5A expression to human NSCLC pathogenesis, including that associated with COPD, a chronic inflammatory condition and risk factor for lung cancer, remains elusive. In light of the strong association between inflammation, COPD and NSCLC,^10,11 we sought to examine the expression of GPRC5A NBE in COPD-associated adenocarcinoma and SCC. GPRC5A expression was assessed in histological tissue specimens of NSCLC and NBE, including those from lung cancer-free COPD and from NSCLC patients. Representative photomicrographs of GPRC5A immunohistochemical staining in the different NBE are depicted in Figure 2A. GPRC5A immunoreactivity seemed to be highest in NBE from lung-disease–free never and ever smokers, successively significantly decreased in NBE of COPD patients who are cancer-free, and lowest in NBE of patients with both COPD and adenocarcinoma or SCC (Fig. 2B). A general linear model demonstrated a significant gradual decrease of GPRC5A expression from NBE of disease-free never smokers to NBE from patients with COPD and adenocarcinoma or SCC (p < 0.0001) (Fig. 2B). The mean (177.31 ± 30.66) and median (195; minimum, 130; maximum, 220) of GPRC5A expression in NBE of lung-disease–free never smokers were higher than that of smokers (mean, 155.28 ± 34.15, median, 147.5; minimum, 100; maximum, 230) although the differences were not statistically significant (Fig. 2B). Moreover, GPRC5A expression was significantly higher in NBE of lung-disease–free smokers relative to NBE in cancer-free COPD patients (mean, 85.48 ± 53.55, median, 72.88; minimum, 20; maximum, 211) (p = 0.007), which in turn was significantly higher compared with the expression in NBE from COPD patients with adenocarcinoma (mean, 48.15 ± 52.84, median, 35; minimum, 0; maximum, 170) (p < 0.0001) or SCC (mean, 32.24 ± 34.9, median, 12.5; minimum, 0; maximum, 101.69) (p < 0.0001) (Fig. 2B). Furthermore, although GPRC5A expression was lower in NBE of COPD patients with SCC compared with those with COPD and adenocarcinoma, the difference was not statistically significant (Fig. 2B). These findings suggest that reduced expression of the GPRC5A tumor suppressor may be implicated in the pathogenesis of NSCLC associated with inflammatory COPD.

FIGURE 2.

Reduced GPRC5A expression in the sequential pathogenesis of COPD-associated NSCLC. A, Representative photomicrographs of GPRC5A immunohistochemical expression in NBE from disease-free never and ever smokers (upper left and upper right, respectively) and in NBE from cancer-free COPD patients (bottom left) and NSCLC patients (bottom right) with COPD observed under a light microscope at a magnification of × 20. Bars, 100 μm. B, Box-plots depicting quantification of GPRC5A immunohistochemical expression represented in (A). Main p value signifies statistical significance of reduced GPRC5A expression in the sequential pathogenesis of COPD-associated NSCLC and among the five groups by the general linear model and Kruskal–Wallis test. p values signifying the indicated pair wise comparisons among the groups were obtained by the Wilcoxon rank sum test. GPRC5A, G-protein coupled receptor family C, group 5, member A; NBE, normal bronchial epithelia; NSCLC, non–small-cell lung cancer; COPD, chronic obstructive pulmonary disease; SCC, squamous cell carcinomas; NS, not significant.

We then examined GPRC5A expression in epithelial cells from different compartments of the lung (bronchial, bronchiolar and alveolar epithelia and cells). GPRC5A expression was statistically significantly lower in bronchi and bronchioles from cancer-free COPD patients (p < 0.005) and patients with both COPD and SCCs or adenocarcinomas (p < 0.0001) compared with lung-disease–free never smokers (Supplementary Table 1, Supplemental Digital Content 1, http://links.lww.com/JTO/A334). In contrast, GPRC5A expression in alveolar compartment remained high (score average 172.2–261.5) in all groups of cases examined, and there were no statistically significant differences based on COPD status (Supplementary Table 1, Supplemental Digital Content 1, http://links.lww.com/JTO/A334).

GPRC5A Expression in Smoker NSCLC Field Cancerization

Our previous findings on GPRC5A tumor-suppressive function^15,16,19 as well as our current observation on significant and progressive reduced expression of this tumor suppressor in NBE from lung cancer-free COPD patients and patients with NSCLC, prompted us to probe whether GPRC5A expression exhibits a field cancerization pattern in NSCLC. We sought to assess GPRC5A transcript expression in multiple bronchial sites (n = 29) as well as in the tumors (n = 6) and uninvolved normal lung parenchyma (n = 6) in six NSCLC cases (3 SCCs and 3 adenocarcinomas). Quantitative real-time PCR analysis demonstrated progressively reduced GPRC5A expression from normal uninvolved lung parenchyma (relative quantification [RQ] mean, 4.18 ± 0.62; median, 4.16; minimum, 3.51; maximum, 5.04) to histologically normal and adjacent bronchial epithelia (RQ mean, 3.12 ± 1.94; median, 2.86,; minimum, 0.59; maximum, 10.12) to NSCLC tumors (RQ mean, 1.35 ± 0.71; median 1.31; minimum 0.24; maximum 2.30), which exhibited on an average the lowest expression of the gene (Fig. 3A and B) (p = 0.03). These findings suggest that GPRC5A expression is reduced in the smoking-injured field cancerization in NSCLC highlighting a potentially strong implication of this tumor suppressor in the pathogenesis of this malignancy.

FIGURE 3.

Decreased GPRC5A mRNA expression in histologically NBE and corresponding NSCLC tumors from resected specimens compared with matched uninvolved normal lung. Total RNA was isolated from brushings of NBE and frozen sections of matched NSCLC (T) and uninvolved NL parenchyma obtained from resected lobectomy or pneumonectomy specimens performed on smoker early-stage patients. A, GPRC5A expression was analyzed in all samples (in triplicates) by QRTPCR and normalized against that of ACTB to yield a relative expression by the 2^−△△CT method depicted in the graphs. NSCLC tumors (squamous cell carcinoma, cases 1–3, and adenocarcinomas, cases 4–6), uninvolved NL parenchyma and bronchial structures are labeled by the indicated colors. B, GPRC5A transcript expression among uninvolved NL tissue, NBE, and NSCLC tumors was assessed statistically by repeated measures analysis. NBE, normal bronchial epithelia; NSCLC, non–small-cell lung cancer; NL, normal lung; CT, comparative cycle threshold method.; QRTPCR, quantitative real-time polymerase chain reaction.

DISCUSSION

In this study, we assessed the expression of GPRC5A, which we had found to exhibit lung-specific tumor-suppressor properties in mice,^15,16,19 in human NSCLC histological tissue specimens, as well as in NBE from COPD and lung adenocarcinoma and SCC patients. We found that the protein product of human GPRC5A was lower in lung tumors compared with uninvolved histologically NBE, and was significantly associated with adenocarcinoma histology. GPRC5A airway expression seemed to exhibit a progressive decrease in the field cancerization of COPD-related NSCLC with highest expression in bronchial epithelia obtained from disease-free individuals, intermediate immunoreactivity in normal epithelia from cancer-free COPD patients, and lowest in normal epithelia from patients with both COPD and adenocarcinoma and SCC histologies. Furthermore, GPRC5A transcript expression was also lower in histologically NBE as well as in adjacent and corresponding smoker NSCLC tumors, irrespective of COPD status, compared with uninvolved normal lung suggestive of a field cancerization-mediated expression modulation of this G-protein coupled receptor. In light of the relevance of the field cancerization phenomenon to NSCLC pathogenesis,^5–7,20,21 our findings pinpoint to a potential tumor-suppressive role, similar to that established in mice, of GPRC5A in the sequential development of human NSCLC, in particular those associated with inflammatory chronic obstructive disease. NSCLC and COPD are both mainly attributable to cigarette smoke^9,10,13 and are leading causes of deaths in the United States and worldwide.^1,2,11 Given that COPD is a major risk factor for lung cancer and shares various pathogenic features with lung tumors,^9,10,22–25 better molecular markers are needed to identify which COPD patients will continue to develop lung malignancies.¹¹ Thus, the results herein raise the intriguing possibility that GPRC5A loss may be a useful biomarker in assessing the risk of NSCLC development in COPD patients.

COPD and infections as well as inflammatory disorders of the respiratory tract may be linked to an increased risk of lung cancer.^26,27 We have previously demonstrated significantly increased macrophage infiltration into lungs of Gprc5a-knockout mice along with their direct association with adenocarcinomas, which was accompanied by higher constitutive levels of proinflammatory cytokines and chemokines and increased susceptibility to stimulation of nuclear factor-kappa B (NF-κB) activation in vivo.¹⁵ We also showed that loss of Gprc5a-mediated activation of NF-κB was causally linked to macrophage recruitment and enhanced inflammation toward the creation of a tumor-promoting microenvironment.¹⁵ It is noteworthy that lungs of patients with severe COPD exhibit macrophage/CD8⁺ T cell and neutrophil infiltration based on histopathological studies and bronchial-biopsies.²⁸ Moreover, it is worthwhile to mention that we attempted to statistically assess the association between expression of GPRC5A and that of NF-κB (nuclear p65 immunoreactivity) in NSCLCs and found a significant inverse correlation between both proteins in particular in adenocarcinomas (data not shown). In addition, it was recently shown that the incidence of hyperplastic lesions in lungs of Gprc5a-knockout mice was increased after exposure to nontypeable Haemophilus influenzae.²⁹ Notably, bacterial colonization, particularly with nontypeable Haemophilus influenzae, has been implicated as a cause of airway inflammation in COPD besides cigarette smoke.³⁰ In this context, the commonalities between inflammatory-mediated and histopathological mechanisms in Gprc5a-knockout mice and those evident in lungs of COPD patients along with our current findings on reduced GPRC5A expression in normal cells from COPD patients pinpoint to a tumor-suppressive role of this gene in COPD-associated human lung tumorigenesis.

We previously cloned the retinoid-regulated mGprc5a/hGPRC5A and found that it was preferentially expressed in fetal and adult mouse and human lung tissue compared with normal specimens from other anatomically distinct organs.³¹ The functional relevance of this lung-specific expression was highlighted by our earlier study demonstrating spontaneous development of adenomas and adenocarcinomas in Gprc5a-nockout mice, which was not observed in wild-type littermates. In addition, spontaneous tumor development in Gprc5a-knockout mice was characterized by late tumor onset (12–16-month-old mice) and low multiplicity.¹⁶ Notably, we reported that exposure to the tobacco-specific carcinogen 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (nicotine-derived nitrosamine ketone) augmented tumorigenesis evidenced by 6-month earlier development of lesions, two- to fourfold increased tumor incidence and multiplicity, as well as a dramatic increase in lesion size.¹⁹ It is worthy to note that, in this study, we found significantly reduced expression of GPRC5A in human smoking-injured histologically NBE adjacent to lung tumors and in the localized field cancerization of smoker NSCLC patients, compared with uninvolved normal lung. Moreover, reduced GPRC5A expression was concomitant among airways and corresponding NSCLC compared with matched normal lung tissue. It is plausible to assume that smoking-mediated reduced airway expression of GPRC5A in patients may explain our previous observations on augmented tumorigenesis in Gprc5a-knockout mice after exposure to tobacco carcinogens. Because COPD is causally linked to smoking,^9–11,22,23 it cannot be neglected that reduced GPRC5A expression in NBE from COPD patients may be because of the molecular impact of smoking on the airway transcriptome. However, it is important to mention that we found significantly decreased GPRC5A expression in normal epithelia from COPD patients compared with epithelia obtained from disease-free smokers. Moreover, GPRC5A expression was further decreased and lowest in normal airway epithelia from COPD patients with NSCLC compared with those from cancer-free COPD patients. Thus in this context, our findings raise the possibility that GPRC5A expression may be impacted by cigarette smoking and may be both a cause and consequence of increased inflammation in tumor-promoting microenvironment in the lung as well as NSCLC development, which may be better delineated in future warranted studies.

Our group has previously shown that human lung and head and neck cancer cell lines exhibit reduced GPRC5A expression compared with their normal counterparts.³¹ It is worthwhile to mention that treatment of cancer cells with all-trans retinoic acid restored GPRC5A expression in cells devoid of the tumor suppressor in part by a retinoic acid receptor-dependent manner.^31,32 In addition and in a separate study, we have previously noted reduced retinoic acid receptor beta (RARβ) in approximately 50% of adenocarcinomas and SCCs as well as loss of RARγ and retinoid × receptorβ in a significant fraction of the tumors.³³ Moreover, RARβ expression is decreased in bronchial epithelia and preneoplastic lesions representing the sequence of lung SCC pathogenesis.³⁴ More recently, we demonstrated, using gene expression profiling, that loss of murine Gprc5a in lung epithelial cells reduced expression of markers of squamous differentiation concomitantly with an increase in mediators of the inflammatory process (e.g., NF-κB signaling). In light of the biological connection between retinoid signaling and GPRC5A expression as well as our current findings in this revised article on the significant reduced expression of the tumor suppressor in NBE of COPD patients including those with NSCLC, it is plausible to suggest that retinoid signaling may be aberrantly regulated in pathogenesis of COPD-associated NSCLC. It is important to note that retinoids are currently being tested for their lung regenerative properties and effects in patients with COPD.³⁵

In conclusion, we report herein decreased expression of GPRC5A in human NSCLC tissue compared with uninvolved NBE and in the field cancerization of smoker NSCLCs pinpointing to a tumor-suppressor role of this G-protein coupled receptor in the pathogenesis of this leading cause of cancer-related deaths. Moreover, our findings highlight a progressive decrease in GPRC5A expression in the sequential pathogenesis of NSCLCs arising in COPD patients, warranting future studies, including the analysis of lung cancer preneoplastic lesions, to assess the potential of the utility of this gene as a biomarker for lung cancer risk in COPD patients.