K-ras activation generates an inflammatory response in lung tumors

Article metrics

Abstract

Activating mutations in K-ras are one of the most common genetic alterations in human lung cancer. To dissect the role of K-ras activation in bronchial epithelial cells during lung tumorigenesis, we created a model of lung adenocarcinoma by generating a conditional mutant mouse with both Clara cell secretory protein (CC10)-Cre recombinase and the Lox-Stop-Lox K-rasG12D alleles. The activation of K-ras mutant allele in CC10 positive cells resulted in a progressive phenotype characterized by cellular atypia, adenoma and ultimately adenocarcinoma. Surprisingly, K-ras activation in the bronchiolar epithelium is associated with a robust inflammatory response characterized by an abundant infiltration of alveolar macrophages and neutrophils. These mice displayed early mortality in the setting of this pulmonary inflammatory response with a median survival of 8 weeks. Bronchoalveolar lavage fluid from these mutant mice contained the MIP-2, KC, MCP-1 and LIX chemokines that increased significantly with age. Cell lines derived from these tumors directly produced MIP-2, LIX and KC. This model demonstrates that K-ras activation in the lung induces the elaboration of inflammatory chemokines and provides an excellent means to further study the complex interactions between inflammatory cells, chemokines and tumor progression.

Introduction

Activating mutations of the K-ras oncogenes, found in 20 to 40% of nonsmall cell lung carcinomas (NSCLC), are one of the most common genetic alterations in human lung cancer, especially for those associated with tobacco cigarette smoke exposure (Rodenhuis and Slebos, 1990). The majority of K-ras genetic alterations are guanine to thymidine point mutations (Rodenhuis and Slebos, 1990) in codon 12 (i.e. G12D), resulting in the change of the encoded amino acid from glycine (G) to aspartic acid (D). Activated Ras protein is known to directly interact with a number of distinct effectors to trigger downstream signaling pathways, including activation of the RAF/MEK/ERK, PI3K/AKT and RAL-GTP pathways, to impact cellular proliferation, apoptosis and neoplastic transformation (Vojtek and Der, 1998; Adjei, 2001). Despite years of intense effort, therapeutics targeted against oncogenic K-ras have failed to prolong patient survival in tumors harboring activated K-ras mutations (Karp et al., 2001; Doll et al., 2004). Thus, a more detailed understanding of the downstream effectors of oncogenic K-ras is required to identify additional targets for the development of novel therapeutics.

The role of activated K-ras in tumorigenesis is unlikely to be confined to cell-autonomous effects (Karin, 2005). Recent data implicate activated K-ras in the elaboration of cytokines from tumor cells, thereby generating a proinflammatory tumor microenvironment, which may promote tumor growth and invasion. For example, IL8 has been shown to be a transcriptional target for ras signaling and its secretion is required for tumor-associated inflammation and angiogenesis in a human tumor xenograft model (Sparmann and Bar-Sagi, 2004). Ras has also been shown to mediate endothelial cell-dependent tumor angiogenesis via transcriptional upregulation of vascular endothelial growth factor (VEGF) and repression of thrombospondin (TSP-1) (Okada et al., 1998; Watnick et al., 2003). Ras alters the tumor microenvironment by inducing the expression of members of the matrix metalloproteinase (MMP) family (Ballin et al., 1988), which are capable of degrading basement membrane and other extracellular matrix (ECM) structures, allowing for tumor growth and invasion.

The roles of inflammation, tumor microenvironment and ECM remodeling during tumorigenesis are complex, as multiple cell types are involved in intricate crosstalk that is difficult to recapitulate in vitro. Thus, to understand the role of oncogenic ras in the process of tumorigenesis and tumor associated inflammation, we generated a cohort of conditional mutant mice in which the G12D oncogenic allele of K-ras is expressed specifically in CC10-positive bronchiolar epithelial cells. As expected, these conditional K-ras mutant mice developed lung tumors. Additionally, we have demonstrated that activated K-ras tumor cells produce and secrete chemotactic substances resulting in a profound inflammatory response within the lung that is associated with early mortality.

Results

CC10-Cre/LSL-K-rasG12D conditional mutant mice develop lung tumors

We generated a cohort of conditional mutant mice by crossing Lox-Stop-Lox (LSL)-K-rasG12D mice (Jackson et al., 2001) to CC10-Cre mice in which Cre recombinase expression is under control of the Clara cell secretory protein promoter. As with other transgenes driven by CC10 (Perl et al., 2002), Cre expression was limited to the lung bronchiolar epithelium as demonstrated by analysis of the CC10-Cre strain crossed with the ROSA-Stop-LacZ reporter strain (data not shown), which demonstrated the complete absence of LacZ staining beyond the bronchoalveolar duct junction. We analysed serial histological sections at high power from all lobes of a cohort of these mice, and were unable to detect LacZ positive cells beyond the bronchoalveolar duct junction (data not shown). Thus, in the CC10-Cre/LSL-K-ras mice, the endogenous promoter driven expression of the oncogenic K-ras occurred in the bronchiolar epithelial cells. These mutant K-ras expressing mice are viable at birth but display a dramatically shortened life span, with a median survival of 8 weeks (n=35), when compared to either of the corresponding single mutant (CC10-Cre or LSL-K-rasG12D) littermate controls (n=31) (Figure 1). Serial histological analyses of the lungs from CC10-Cre/LSL-K-rasG12D mice as a function of age showed the appearance of airway epithelial cell hyperplasia by 3 weeks of age, progression to adenoma by 6 weeks, and less commonly to adenocarcinoma at time points beyond 15 weeks of age (Figure 2a–d) CC10-Cre/LSL-K-ras tumors lose CC10 expression (Figure 2g), which has been proposed to be a marker of poorly differentiated tumors (Szabo et al., 1998). Additionally, these tumors express SPC (Figure 2h). It is likely that SPC expression by CC10-Cre/LSL-K-ras tumors is reflective of the fact that these lung tumors in mice arise from stem cells located at the bronchoalveolar duct junction, which are known to express both CC10 and SPC (Kim et al., 2005) and/or that there is active downregulation of the CC10 promoter as CC10 expression has been associated with tumor suppression (Linnoila et al., 2000).

Figure 1
figure1

CC10-Cre/LSL-K-rasG12D mice display early mortality. Kaplan–Meier survival curves are shown for CC10-Cre-LSL-K-rasG12D mice (n=35) and the single transgenic LSL-K-rasG12D control (n=31). Median survival for the CC10-Cre/LSL-K-rasG12D group is 8.0 weeks. The survival curves were generated using Prism data analysis software (P<0.0001).

Figure 2
figure2

CC10-Cre-LSL-K-rasG12D mice display lung tumorigenesis. Normal airway and alveolar structures from an LSL-K-ras single mutant control mouse are depicted in (a). CC10-Cre/LSL-K-ras mice chronologically develop airway epithelial cell hyperplasia (b) by 3 weeks, adenomas (c) by 6 weeks and occasionally features of adenocarcinoma (d), such as prominent nucleoli and high N/C ratio (arrows), but only at timepoints beyond 15 weeks. These mice also develop pronounced pulmonary inflammation (e and f), in which macrophages occupy a large proportion of the alveolar space, shown here at 10 weeks. Immunostaining for CC10 (g) demonstrates airway specific staining. The adenoma shown has lost CC10 expression. In contrast, SPC (h) staining spares the airway structures but does demonstrate SPC expression by adenomas.

CC10-Cre/LSL-K-rasG12D mice develop pronounced pulmonary inflammation

Surprisingly, detailed analyses of the lungs from CC10-Cre/LSL-K-rasG12D mice revealed only modest tumor burdens at the time of death. The tumors were typically multifocal but restricted in size to <0.1 to 0.2 mm and did not show significant bronchiolar invasion, airway obstruction or distant metastasis. The major histological abnormality in these mice was a profound inflammatory response within the lung in which macrophages and neutrophils occupied much of the alveolar airspace, to an extent that may prevent effective gas exchange (Figure 2e and f). Detailed histological of all other major organs including the spleen, brain, liver, intestines and muscle did not reveal any other site of inflammation in the CC10-Cre/LSL-K-rasG12D mice. To quantify the inflammatory infiltration into the lung, BAL was performed on the conditional mutant K-ras mice and the single (CC10-Cre or LSL-K-rasG12D) mutant littermate controls at 4 and 8 weeks of age. Control BAL total cell counts were 6.4 × 104 cells (98% mac, <1% PMN) at 4 weeks and 8.8 × 104 cells (97%, mac <1% PMN) at 8 weeks. The total cell counts were elevated in CC10-Cre/LSL-K-ras mice both at 4 weeks (total count=45.2 × 104 cells, 97% macs, 0% PMN) and 8 weeks (total count=165.4 × 104 cells, 73% macs, 25% PMN). More specifically, lung macrophages in CC10-Cre/LSL-K-rasG12D mice were elevated fivefold when compared to LSL-K-ras controls at 4 weeks, and continued to increase with age, such that by 8 weeks, a 14-fold increase in macrophage content in the K-ras mutant vs single mutant controls was found (Figure 3a). In contrast, BAL neutrophil counts were normal at 4 weeks but displayed a several 100-fold increase by 8 weeks in CC10-Cre/LSL-K-rasG12D mice when compared to LSL-K-ras controls (Figure 3b). Immunohistochemistry using the macrophage specific mac-3 stain (Flotte et al., 1983) demonstrated that although there were some tumor-associated macrophages; the majority were located within the alveolar space (Figure 3c and d), and not in close proximity to tumors. However, the neutrophil specific GR-1 immunostain (Lagasse and Weissman, 1996) demonstrated that the neutrophils were predominantly colocalized with small tumors and to a lesser extent with the larger tumors (Figure 3e and f). One potential explanation of this finding is that the larger tumors may produce excessively high concentrations of CXC chemokines. If the concentration is sufficiently elevated, the gradient may repel cells that are normally recruited by a given chemokine, a property known as chemofugetaxis (Poznansky et al., 2002).

Figure 3
figure3

CC10-Cre/LSL-K-ras mice accumulate macrophages and neutrophils within the lungs. BAL was performed at 4 and 8 weeks of age to determine the macrophage (a) and Neutrophil (b) content of the lungs. N=6 for each group at each time point. * denotes P<0.01. To determine the location of these inflammatory cells within the lung, mac-3 and GR-1 immunostaining was performed to identify macrophages and neutrophils, respectively. Resident alveolar macrophages in single transgenic LSL-K-ras control mice (c) are shown in contrast to the abundant accumulation of macrophages throughout the alveolar space in CC10-Cre/LSL-K-ras mice at 8 weeks of age (d). GR-1 staining in CC10-Cre/LSL-K-ras mice demonstrates the abundance of neutrophils within a small adenoma (e) and the more modest association of neutrophils with larger adenomas (f), both at 8 weeks.

To determine which chemokines were involved in the recruitment of these inflammatory cells into the lungs of the CC10-Cre/LSL-K-rasG12D mice, we measured the levels of several C-C and C-X-C chemokines, molecules known for their monocyte and neutrophil chemotactic properties, respectively, in BAL fluid from these mice (Walz et al., 1991; Baggiolini et al., 1994; Baggiolini et al., 1995) (Figure 4). None of these chemokines were detectable above control values at 3 weeks of age by ELISA. Of the C-C chemokines, MCP-1 was elevated by 7 weeks of age and MIP-1α was elevated by 10 weeks in CC10-Cre/LSL-K-rasG12D mice when compared to single mutant controls. These findings correlated with the early increase and continued rise in BAL macrophages. The neutrophil chemokines, MIP-2 and KC (both orthologues of IL-8) were both modestly elevated in the BAL fluid of CC10-Cre/LSL-K-rasG12D mice relative to single mutant controls by 7 weeks and dramatically elevated at the 10 week time point, in agreement with the later rise in neutrophils relative to macrophages. In addition, at 10 weeks, the level of VEGF, a cytokine that is associated with angiogenesis, was dramatically elevated in CC10-Cre/LSL-K-rasG12D mice.

Figure 4
figure4

Chemokine expression in BAL fluid. BAL fluid was obtained from CC10-Cre/LSL-K-ras mice and LSL-K-ras controls at 4, 7 and 10 weeks of age. Unconcentrated BALF was normalized for total protein content and subjected to ELISA testing for MIP-1α, MCP-1, MIP-2, KC, VEGF, and TNF-α. N=8 samples per group; each sample performed in triplicate. Results are expressed in pg/ml. * denotes P<0.05.

K-ras activated cells produce inflammatory chemokines

The abundance of alveolar macrophages, neutrophils and chemokines within the lung prompted us to address whether the tumor cells themselves contributed to the inflammatory phenotype. In order to assess the capacity of the tumor cells to produce these chemokines, we sought to establish tumor-derived cell lines. Since lung tumor cell lines from the CC10-Cre/LSL-K-rasG12D mice were not readily cultured, we crossed these mice onto a p53 mutant background. CC10-Cre/LSL-K-rasG12D/p53−/− mice developed lung tumors with kinetics and histologic phenotypes comparable to the CC10-Cre/LSL-K-rasG12D strain. Importantly, tumors from the p53-deficient animals were amenable to culture in vitro (data not shown). Using RNA prepared from three different tumor cell lines, we demonstrated, via real time quantitative PCR analyses, that these cell lines secrete the neutrophil chemokines KC, MIP-2 and LIX (CXCL5), at levels significantly above that observed in the single mutant lung (Table 1). Interestingly, these tumor cell lines produced only modest amounts of chemokines that are chemotactic for macrophages, such as MCP-1 and MIP-1α. ELISA assays for the various chemokines performed on the cell culture supernatants further confirmed the real time quantitative PCR data (data not shown).

Table 1 Chemokine expression from tumor cell lines

These results demonstrate that K-ras activated tumor cells produce neutrophil chemokines and recruit neutrophils to the site of tumorigenesis. The source of monocyte chemokine production remains uncertain but clearly does not originate from tumor cells. There are several capable sources of MCP-1 and MIP-1α residing within the lung, including pulmonary vascular smooth muscle cells (Lukacs et al., 1995) and airway epithelial cells, both likely sources.

Comparison between two different models of K-ras activated tumors

The LSL-K-rasG12D mutant mouse has previously been shown to develop progressively less differentiated lung tumors after the administration of adenoviral Cre. Although extensive pulmonary inflammation was not a major component of the phenotype in this model, we wished to more closely examine whether comparable inflammatory responses may in fact be operative. We obtained BAL fluid from of cohort of these mice 12 weeks after the administration of a high titer of adenoviral Cre. Histological analyses of the 12-week adeno-Cre infected lungs from these LSL-K-rasG12D mutant mice showed the presence of atypical adenomatous hyperplasia and adenomas with modest infiltration of inflammatory cells (data not shown). ELISA analysis of these adeno-Cre infected LSL-K-rasG12D mutant mouse BAL fluids for MCP-1, MIP-1α, KC and MIP-2 showed that these chemokines are present but at a much lower level than those from the CC10-Cre/LSL-K-rasG12D mice (Figure 5). These results suggested that pulmonary inflammation may be a common component of K-ras directed transformation in the lung epithelium.

Figure 5
figure5

Comparison of chemokine expression from two models of lung tumorigenesis. BAL fluid was obtained from CC10-Cre-LSL-K-rasG12D (8 weeks old) and adenoviral Cre treated (5 × 106 PFU) LSL-K-ras mice (8 weeks post-AdenoCre) in the same way. The unconcentrated BAL fluid was normalized for protein concentration and subjected to ELISA testing for MIP-1α, MCP-1, MIP-2, and KC. N=8 samples per group; each sample performed in triplicate. Results are expressed in pg/ml. Lower limit of detection is 1.5 pg/ml. * denotes P<0.01.

Human lung tumors with activating K-ras mutation have higher levels of inflammation

To extend these findings to human lung cancer, we performed blinded histological analyses on a well-annotated lung adenocarcinoma set (Bhattacharjee et al., 2001) to assess if any association was present between tumor-associated inflammation and K-ras mutational status. To this end, 76 human lung adenocarcinoma frozen tissue samples stained with hematoxylin and eosin (H&E) were analysed by a pathologist (RP) who was not informed of the K-ras mutational status of the specimens. The samples were then divided into quartiles based on the degree of inflammation present within the tumors. The proportion of samples demonstrating K-ras mutations was significantly higher in the quartile showing the highest degree of inflammation, when compared to the proportion of samples with K-ras mutation present in the lowest inflammation quartile (P<0.05, χ2 test) (Figure 6).

Figure 6
figure6

K-ras mutations are associated with increased inflammation in human lung adenocarcinoma. H&E stained sections from a previously described dataset of 76 human lung adenocarcinomas (Bhattacharjee et al., 2001) were scored for severity of inflammation by a pathologist (RP) who was blinded as to mutational status of the tumors. The samples were placed into quartiles based on the inflammation score. Mutations in K-ras were present in the adenocarcinomas scored in the highest quartile of inflammation significantly more than in the lowest quartile as determined by the χ2 method, P<0.05. Representative examples of two lesions graded in the highest quartile (a and b) and the lowest quartile (c and d) of inflammation are shown.

Discussion

In the current study, we have shown that K-ras activation in CC10-positive bronchiolar epithelial cells causes the development of lung tumors. Additionally, we showed that expression of K-rasG12D within the bronchiolar epithelium induces the production of chemokines that leads to the accumulation of neutrophils and macrophages within the lung. We are able to extend our findings to the human disease by showing that human lung adenocarcinomas that harbor K-ras mutations are associated with a higher degree of inflammation.

With respect to tumorigenesis, the findings presented here are comparable to other recently published models using either adenoviral Cre or transgenic systems to induce the expression of mutant K-ras within the lung (Fisher et al., 2001; Jackson et al., 2001; Johnson et al., 2001; Meuwissen et al., 2001). However, the robust nature of inflammation in the CC10-Cre/LSL-K-rasG12D model and the high concentrations of chemotactic substances elaborated by these tumor cells have not been previously reported and therefore provide a unique opportunity to dissect the means by which inflammatory cells are recruited to sites of tumorigenesis, and the roles that these cells play in tumor progression vs suppression. The adenoviral Cre activated LSL-K-rasG12D model displays a readily detectable, although more modest level of inflammation, suggesting that this phenotype is a general response to K-ras activation in the lung epithelium. The elevated responses seen in the CC10-Cre/LSL-K-rasG12D model may relate to the greatly increased proportion of lung epithelial cells harboring activated K-ras in these mice. The production of C-X-C chemokines by K-ras mutant tumor cell lines establishes a direct link between activated K-ras and neutrophil recruitment. Furthermore, the elevation of MIP-2, KC and LIX observed here correlates with published reports that human NSCLC produces the human counterparts of these factors (IL-8 and ENA-78 (CXCL-5)) (Arenberg et al., 1996, 1998).

In addition to their chemotactic properties for neutrophils, several members of the C-X-C family of chemokines are known to be directly angiogenic if they display the ELR motif (Strieter et al., 1995). Both IL-8 and ENA-78, the human orthologues of KC/MIP-2 and LIX, respectively, have recently been implicated in fueling tumor angiogenesis in a xenograft model (Arenberg et al., 1998; Sparmann and Bar-Sagi, 2004). One of these reports used K-ras transfected cell lines to demonstrate that K-ras mutations drive the expression of IL-8 (Sparmann and Bar-Sagi, 2004). Our results, which demonstrate elevations in MIP-2 and KC, are consistent with these results. Whether the tumor microenvironment is influenced directly by these angiogenic chemokines or by the neutrophils that they recruit requires further investigation.

Although the association between inflammation and cancer has been long recognized, only recently have specific inflammatory cells and their bioactive substances been implicated in promoting tumorigenesis. Experimental studies have demonstrated that tumor neutrophils and macrophages can be an intrinsic and tumor-promoting component of tumorigenesis, specifically influencing tumor behavior via activation or degradation of angiogenic, angiostatic and other growth factors, as well as their ability to remodel the ECM using their vast arsenal of MMPs and serine proteinases (Coussens and Werb, 2002). As examples, macrophages (Ono et al., 1999), mast cells (Coussens et al., 1999), and bone marrow-derived MMP-9 (Coussens et al., 2000) have all been demonstrated to play important roles in tumor angiogenesis. Neutrophils are capable sources of oxidases that generate reactive oxygen species that have the potential to interact with tumor cells to attenuate their apoptotic cascade and increase their mutational rate (Haqqani et al., 2000). The data presented here establishes a link between the presence of activated K-ras mutations and macrophage and neutrophil predominant inflammation in both murine and human lung tumors. Future work will be required to determine the exact roles of macrophages and neutrophils and the means by which they affect processes such as angiogenesis and tissue invasion in lung cancers.

Materials and methods

Mice

The CC10-Cre mouse was generated using the previously described 2. 4 kb upstream promoter fragment of the rat CC10 gene capable of driving airway-specific heterologous gene expression in transgenic mice (Perl et al., 2002). This promoter fragment was generated by PCR from rat genomic DNA and was inserted upstream of the coding sequence for a modified Cre recombinase containing a nuclear localization sequence (gift of TJ Ley, St Louis). The complete characterization of this mouse will be published elsewhere. The LSL-K-rasG12D allele was generously provided by Dr Tyler Jacks' laboratory and has been thoroughly described elsewhere (Jackson et al., 2001). All experiments utilized age and sex-matched single mutant controls, either CC10-Cre or LSL-K-rasG12D (data shown in figures represents LSL-K-rasG12D). CC10-Cre mice were on a mixed 129SvJ-C57BL/6 background, while the LSL-K-ras mice were on a mixed FVB-129SvJ background. All of these mice are maintained on approved protocols in a pathogen free environment at Dana-Farber Cancer Institute.

In a subset of the experiments, mutant K-ras activation was achieved via intratracheal administration of adenoviral Cre to LSL-K-ras mice as previously described (Jackson et al., 2001). In brief, the mice were anesthetized with avertin, intubated with an i.v. catheter and administered a titer 5 × 106 PFU adenoviral Cre in total volume 40 μl.

BAL fluid analysis

At the appropriate time points, mice were killed before performing BAL via a 22 g i.v. catheter inserted in to the trachea. The lungs were lavaged with 0.75 ml saline 4 × to obtain the BAL fluid. The BAL fluid was then centrifuged at 3000 r.p.m. for 3 min. The supernatant was retained for further analysis. The red blood cells were lysed, and the BAL was resuspended in 1.0 ml normal saline to evaluate cellular content. Macrophage, neutrophil, lymphocyte and total cell counts were determined using a hemocytometer and cytospins stained with Hema3 (Fisher).

ELISA

BAL fluid was obtained as above. Each sample of unconcentrated BAL fluid was normalized for total protein content using the BCA Protein Assay Kit (Pierce). ELISA testing was also performed on cell culture supernatants. Primary lung tumor cell lines were established as described below. The supernatants were normalized for total protein content as described for BAL. The levels of MIP-2, MCP-1, MIP-1α, KC, TNF-α, VEGF were determined using the commercially available Quantakine kits (R&D Systems) as per the manufacturer's instructions. Each sample was performed in triplicate, n=8 samples.

Lung tissue preparation

At the appropriate timepoints, the lungs were inflated to allow for further analysis. The animals were killed, the right ventricle (RV) was perfused with saline to remove blood, and the lungs were inflated at 25 cm H2O pressure with 10% buffered formalin for 10 min via of an intratracheal catheter. The lungs were then removed and fixed in 10% buffered formalin for 24 h before embedding in paraffin. Serial mid-sagittal sections (5 μ M thickness) were obtained for histological analysis. A subset of the lungs were inflated and fixed with zinc (BD Biosciences) to allow for specific immunostains.

Immunohistochemistry

Lung content of macrophages and neutrophils were analysed using mac-3 and GR-1 (rat anti-mouse antibodies at 1:1000 dilution and 1:100 dilution, respectively; Pharmingen) immunostaining on mid-sagittal sections using the avidin-biotin horseradish peroxidase technique in which 3,3″-diaminobenzidine was the chromogenic substrate. Zinc-fixed sections were utilized for GR-1 immunostaining.

Generation of primary lung cancer cell lines

Single lung tumors were excised from tumor bearing CC10-Cre/LSL-K-rasG12D mice that were crossed with p53 null animals on a mixed FVB-129SvJ background. The tumors were washed four times in sterile PBS, minced with sterile blades and incubated in RPMI 1640 media containing 4% collagenase/dispase for 30 min at 37°C. Following digestion, the cells were washed with RPMI 1640 containing 10% fetal bovine serum and plated on fibronectin/vitrogen coated plates, which were used for the first three passages. Regular cell culture plates were utilized thereafter. Routine microscopic examination ensured that these cells were epithelial in origin.

Real-time quantitative RT-PCR

Tumor cell lines were generated as above. RNA was isolated from these cells and single mutant lung tissue using the trizol method as per the manufacturer's instructions (Ambion). The samples were reverse-transcribed using RT–PCR beads with ply (dT) oligos as per manufacturer's instructions (Amersham). Real-time PCR analysis was performed using the GeneAmp 5700 Sequence Detection System (Applied Biosystems). The comparative threshold method was employed using GAPDH as an endogenous reference housekeeping gene. Serial dilutions of a control sample of cDNA were used as the standard curve for each reaction. SYBR green buffer (Applied Biosystems) was used as the fluorphore. All experiments were performed in triplicate from an n=4 different biologic specimens. The primers sequences (Invitrogen) were as follows: MCP-1, forward primer 5′-IndexTermTTCTGGGCCTGCTGTTCAC-3′; reverse primer 5′-IndexTermGAGCCAACACGTGGATGCT-3′; MIP-1α, forward primer, 5′-IndexTermCTGCCTGCTGCTTCTCCTACA-3′; reverse primer, 5′-IndexTermCAACGATGAATTGGCGTGG; MIP-1β, forward primer, 5′-IndexTermTCTCTCCTCTTGCTCGTGGC-3′; reverse primer, 5′-IndexTermTGGTGCTGAGAACCCTGGA-3′; CXCL1 forward primer, 5′-IndexTermGCACCCAAACCGAAGTCATA; reverse primer, 5′-IndexTermTGGGGACACCTTTTAGCATC; CXCL2, forward primer, 5′-IndexTermAGCCTGGATCGTACCTGATG; reverse primer, 5′-IndexTermTAACAACATCTGGGCAATGG; CXCL5, forward primer, 5′-IndexTermTTCATGAGAAGGCAATGCTG; reverse primer, 5′-IndexTermCCCAGGCTCAGACGTAAGAA.

Pathologic analyses of human tumors

Frozen section slides of primary lung tumors taken at the time of tissue banking from one of the author's (MM) lung adenocarcinoma set (Bhattacharjee et al., 2001) were reviewed by a pulmonary pathologist (RP), who was blinded to the mutational status of the tumors. The slides were initially screened to ensure a technically acceptable slide and the presence of adenocarcinoma on the section. A total of 76 tumors were further evaluated for the degree of inflammation present within the tumor. Only viable areas of the tumor were evaluated; inflammation associated with areas of necrosis was ignored. The inflammatory infiltrate was subjectively graded on a 1–4 scale where 1 represented minimal, 2 represented mild, 3 represented moderate and 4 represented severe to give a preliminary ranking. The slides were then bubble sorted to put them into a rank order from 1 (least inflammation) to 76 (most severe inflammation), and divided into quartiles. Images were acquired using a QColor 3 digital camera mounted on an Olympus BX41 microscope.

Statistics

Data are expressed at the mean value ±SEM. Statistical significance was determined using the student's t-test (two tailed distribution with a two sample equal variance) unless otherwise indicated. The Kaplan–Meier survival curve analysis was performed using the Prism data analysis software.

References

  1. Adjei A . (2001). J Natl Cancer Inst 93: 1062–1074.

  2. Arenberg DA, Keane MP, DiGiovine B, Kunkel SL, Morris SB, Xue YY et al. (1998). J Clin Invest 102: 465–472.

  3. Arenberg DA, Kunkel SL, Polverini PJ, Glass M, Burdick MD, Strieter R . (1996). J Clin Invest 97: 2792–2802.

  4. Baggiolini M, Loetscher P, Moser B . (1995). Int J Immunopharmacol 17: 103–108.

  5. Baggiolini M, Moser B, Clark-Lewis I . (1994). Chest 105: 95S–98S.

  6. Ballin M, Gomez DE, Sinha CC, Thorgeirsson UP . (1988). Biochem Biophys Res Commun 154: 832–838.

  7. Bhattacharjee A, Richards WG, Staunton J, Li C, Monti S, Vasa P et al. (2001). Proc Natl Acad Sci 98: 13790–13795.

  8. Coussens LM, Raymond WW, Berger G, Laig-Webster M, Behrendtsen O, Werb Z et al. (1999). Genes Dev 13: 1382–1397.

  9. Coussens LM, Tinkle CL, Hanahan D, Werb Z . (2000). Cell 103: 481–490.

  10. Coussens LM, Werb Z . (2002). Nature 420: 860–867.

  11. Doll RJ, Kirschmeier P, Bishop WR . (2004). Curr Opin Drug Discov Devel 7: 478–486.

  12. Fisher GH, Wellen SL, Klimstra D, Lenczowshi JM, Tichelaar JW, Lizak MJ et al. (2001). Genes Dev 15: 3249–3262.

  13. Flotte TJ, Springer TA, Thorbecke GJ . (1983). Am J Pathol 111: 112–124.

  14. Haqqani AS, Sandhu JK, Birnboim HC . (2000). Neoplasia 2: 561–568.

  15. Jackson EL, Willis N, Mercer K, Bronson RT, Crowley D, Montoya R et al. (2001). Genes Dev 15: 3243–3248.

  16. Johnson S, Mercer K, Greenbaum D, Bronson RT, Crowley D, Tuveson DA et al. (2001). Nature 410: 1111–1116.

  17. Karin M . (2005). Nat Med 11: 20–21.

  18. Karp JE, Kaufmann SH, Adjei AA, Lancet JE, Wright JJ, End DW . (2001). Curr Opin Oncol 13: 470–476.

  19. Kim CF, Jackson EL, Woolfenden AE, Lawrence S, Babar I, Vogel S et al. (2005). Cell 121: 823–835.

  20. Lagasse E, Weissman IL . (1996). J Immunol Methods 197: 139–150.

  21. Linnoila RI, Szabo E, DeMayo F, Witschi H, Sabourin C, Malkinson A . (2000). Ann NY Acad Sci 923: 249–267.

  22. Lukacs NW, Kunkel SL, Allen R, Evanoff HL, Shaklee CL, Sherman JS et al (1995). Am J Physiol 268: L856–861.

  23. Meuwissen R, Linn SC, van der Valk M, Mooi WJ, Berns A . (2001). Oncogene 20: 6551–6558.

  24. Okada F, Rak JW, Croix BS, Lieubeau B, Kaya M, Roncar L et al. (1998). Proc Natl Acad Sci 95: 3609–3614.

  25. Ono M, Torisu H, Fukushi J, Nisjie A, Kuwano M . (1999). Cancer Chemother Pharmacol 43: 69–71.

  26. Perl AKT, Wert SE, Nagy A, Lobe CG, Whitsett JA . (2002). Proc Natl Acad Sci 99: 10482–10487.

  27. Poznansky MC, Olszak IT, Evans RH, Wang Z, Foxall RB, Olson DP . et al (2002). J Clin Invest 109: 1101–1110.

  28. Rodenhuis S, Slebos RJ . (1990). Am Rev Respir Dis 142: S27–30.

  29. Sparmann A, Bar-Sagi D . (2004). Cancer Cell 6: 447–458.

  30. Strieter RM, Polverini PJ, Kunkel SL, Arenberg DA, Burdick MD, Kasper J et al. (1995). J Biol Chem 270: 27348–27357.

  31. Szabo E, Goheer A, Witschi H, Linnoila RI . (1998). Cell Growth Differ 9: 475–485.

  32. Vojtek AB, Der CJ . (1998). J Biol Chem 273: 19925–19928.

  33. Walz A, Burgener R, Car B, Baggiolini M, Kunkel SL, Strieter RM . (1991). J Exp Med 174: 1355–1362.

  34. Watnick RS, Cheng YN, Rangarahan A, Ince TA, Weinberg RA . (2003). Cancer Cell 6: 447–458.

Download references

Acknowledgements

We are grateful to Dr Matthew L Meyerson for sharing his unpublished K-ras mutational data set. KKW is supported by NIH Grant K08AG 2400401, the Sidney Kimmel Foundation for Cancer Research and the Joan Scarangello Foundation to Conquer Lung Cancer. This work was also supported by NIH Grant R01 HL70321 (SDS).

Author information

Correspondence to K-K Wong.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Ji, H., Houghton, A., Mariani, T. et al. K-ras activation generates an inflammatory response in lung tumors. Oncogene 25, 2105–2112 (2006) doi:10.1038/sj.onc.1209237

Download citation

Keywords

  • K-ras
  • inflammation
  • lung cancer
  • macrophages
  • neutrophils

Further reading