Assay of lapatinib in murine models of cigarette smoke carcinogenesis

Abstract

Lapatinib, a dual tyrosine kinase inhibitor targeting the epidermal growth factor receptor (EGFR) and human epidermal growth factor receptor 2 (HER-2), is prescribed for the treatment of patients with metastatic breast cancer overexpressing HER-2. Involvement of this drug in pulmonary carcinogenesis has been poorly investigated. We used murine models suitable to evaluate cigarette smoke-related molecular and histopathological alterations. A total of 481 Swiss H mice were used. The mice were exposed to mainstream cigarette smoke (MCS) during the first four months of life. After 10 weeks, MCS caused an elevation of bulky DNA adducts, oxidative DNA damage and an extensive downregulation of microRNAs in lung. After four months, an increase in micronucleus frequency was observed in peripheral blood erythrocytes. After 7.5 months, histopathological alterations were detected in the lung, also including benign tumors and malignant tumors, and in the urinary tract. A subchronic toxicity study assessed the non-toxic doses of lapatinib, administered daily with the diet after weaning. After 10 weeks, lapatinib significantly attenuated the MCS-related nucleotide changes and upregulated several low-intensity microRNAs in lung. The drug poorly affected the MCS systemic genotoxicity and had modest protective effects on MCS-induced preneoplastic lesions in lung and kidney, when administered under conditions that temporarily mimicked interventions either in current smokers or ex-smokers. On the other hand, it caused some toxicity to the liver. Thus, on the whole, lapatinib appears to have a low impact in the smoke-related lung carcinogenesis models used, especially in terms of tumorigenic response.

Introduction

Lapatinib, or N-[3-chloro-4-[(3-fluorophenyl)methoxy]phenyl]-6-[5-[(2methylsulfonylethylamino)methyl]-2-furyl]quinazolin-4-amine, is a dual tyrosine kinase inhibitor targeting both the epidermal growth factor receptor (EGFR) and human epidermal growth factor receptor 2 (HER-2, c-erbB2, c-neu). Lapatinib is indicated for the treatment of patients with advanced breast cancer or metastatic breast cancer whose tumors overexpress HER-2 (1). The US Food and Drug Administration approved this quinazoline derivative in 2007, and the European Commission granted a conditional marketing authorization in 2008, which has been renewed every year by the European Medicines Agency.

Clinical trials have suggested that this drug can suppress the progression of atypical ductal hyperplasia and ductal carcinoma in situ to invasive breast cancer (2), an effect that was previously observed in preclinical studies and in particular in a mouse model of HER-2-overexpressing estrogen receptor (ER)-negative mammary cancer, apparently through an antiproliferative effect (3). These findings provide a compelling rationale for testing HER-targeting drugs not only for the therapy but also for prevention of breast cancer in women at moderate-to-high risk (4).

Since a variety of malignancies are associated with the mutation or increased expression of members of the EGFR or ErbB family, several drugs have been developed to target ErbB-related pathways (5). Focusing on pulmonary carcinogenesis, lapatinib was found to reduce cell proliferation, DNA synthesis and colony formation capacity in A549 bronchoalveolar carcinoma cells in vitro, and A549 tumor-bearing mice treated with lapatinib had significantly less tumors (6). These findings are in line with the established role of both EGFR (7) and HER-2 (8) in smoke-related lung carcinogenesis. However, lapatinib monotherapy did not induce a significant number of tumor regressions in non-small cell lung cancer (9).

We developed suitable models for evaluating the efficacy and tolerability of chemopreventive agents in mice exposed to MCS in terms either of molecular biomarkers (10–15) and/or of lung tumors, both benign and malignant, and other histopathological alterations in multiple organs (16–22).

In the present study, we evaluated lapatinib for the ability to modulate both biomarkers and histopathological alterations either in MCS-free mice or in mice exposed to MCS since birth. In particular, early biomarkers included assessment of nucleotide modifications in the lung, such as bulky DNA adducts and 8-hydroxy-2′-deoxyguanosine (8-oxo-dGuo), and the analysis of the expression of the majority of mouse microRNAs (miRNAs) known thus far. The systemic genotoxicity was evaluated by measuring the frequency of micronucleated (MN) circulating erythrocytes. In the medium term (7.5 months), tumors and other histopathological alterations were evaluated in the lung, liver, kidney and urinary bladder of mice exposed to MCS and receiving lapatinib in the diet under conditions mimicking interventions either in current smokers or in ex-smokers.

The results obtained provide evidence that, in these models, lapatinib is able to attenuate the MCS-related genotoxic and epigenetic alterations but has modest protective effects on induction of lung cancer.

Materials and methods

Experimental design

The experimental design involved three separate studies, including (a) a subchronic toxicity study aimed at selecting the dose of dietary lapatinib, (b) a study assessing early molecular biomarkers in the lung of mice and (c) a chemoprevention study evaluating modulation by lapatinib of systemic genotoxic damage, lung tumors and other histopathological alterations in various organs of MCS-exposed mice. The first two studies were performed in the Genoa laboratory, whereas the third study was performed in the Sofia laboratory.

Mice

A total of 481 strain H mice, 381 newborns and 100 post-weanling at the start of each study, were obtained from the Animal Laboratory of the National Center of Oncology (Sofia, Bulgaria). Neonatal mice of this strain, which are originated from Swiss albino mice, are sensitive to the induction of lung tumors and other histopathological lesions by MCS (15–22).

The mice were housed in Makrolon™ cages on sawdust bedding and maintained on standard rodent chow (Teklad 2018, Harlan Laboratories in Genoa and Kostinbrod in Sofia) and drinking water ad libitum. The animal room temperature was 23±2°C and the relative humidity was 55%, with a 12h day–night cycle. Housing, breeding and treatment of mice were in accordance with NIH and European (86/609/EEC Directive) guidelines. The Institution’s Animal Welfare Assurance was approved by the NIH Office of Laboratory Animal Welfare. The study was authorized by the Italian Ministry of Health.

Exposure to cigarette smoke

A whole-body exposure of mice to MCS was achieved by burning either 3R4F Kentucky reference cigarettes (University of Kentucky, Lexington, KY) in the Genoa laboratory or commercially available cigarettes (Melnik King Size, Bulgartabac) in the Sofia laboratory, as described previously (16–21). The two brands of cigarettes have similar tar and nicotine contents. In fact, 3R4F cigarettes have a declared content of 9.4mg tar and 0.7mg nicotine and deliver 12mg CO each, whereas Melnik cigarettes have a declared content of 9.0mg tar and 0.8mg nicotine and deliver 10mg CO each. MCS was generated by drawing 15 consecutive puffs, each of 60ml and lasting 6 s, by using a syringe connected with the exposure chamber. Each daily session of treatment with MCS involved 6 consecutive exposures, lasting 10min each, with 1 min intervals during which a total air change was made. The average concentrations of total particulate matter in the exposure chambers used in the two laboratories were 684 and 547mg/m³, respectively.

Subchronic toxicity study

Lapatinib, supplied by NCI via MRIGlobal (Kansas City, MO), was incorporated in the diet at 4 dose levels (250, 500, 1000 and 2000mg/kg diet), which were selected based on literature data in mouse studies and taking into account the therapeutic doses used in humans, especially in children (23). Each lapatinib dose and the lapatinib-free control were administered to 20 post-weanling mice (10 males and 10 females). The mice were inspected daily for general appearance and behavior and were weighed at weekly intervals for six weeks.

Treatment of mice for evaluating intermediate biomarkers in the lung

Molecular biomarkers were evaluated in four groups of neonatal mice, each composed of 10 mice (5 males and 5 females). Group A, mice kept in filtered air for 10 weeks (sham-exposed mice); Group B, mice exposed to MCS for 10 weeks, starting within 12h after birth (MCS-exposed mice); Group C, mice receiving lapatinib (1600mg/kg diet) for 6 weeks, starting after weaning (4 weeks); Group D, MCS-exposed mice receiving lapatinib after weaning until the end of the experiment. The mice were inspected daily and weighed at weekly intervals. At 10 weeks of age all mice were killed by CO₂ asphyxiation and their lungs were collected. The left lung, to be used for DNA analyses, was frozen at −80°C. The right lung, to be used for miRNA analyses, was immersed in RNAlater® solution for 24h and then transferred at −80°C.

Evaluation of nucleotide modifications in lung

DNA was extracted individually from the lung of all 40 mice used in the present study and purified by using a commercially available kit (GenElute™ Mammalian Genomic DNA Miniprep kit, Sigma, St Louis, MO). Bulky DNA adducts were enriched by butanol extraction and measured by ³²P post-labeling as described previously (24). A blank and a positive control (benzo(a)pyrene-deoxyguanosine) were also tested. The same lung DNA preparations were evaluated for oxidative DNA damage by measuring 8-oxo-dGuo by ³²P post-labeling as described previously (24).

Evaluation of miRNA expression in lung

The 20 lung specimens collected from either sham-exposed mice or MCS-exposed mice, in the absence of chemopreventive agent, were processed individually in order to evaluate the interindividual variability, whereas the specimens from the lapatinib-treated mice, either MCS-free or MCS-exposed, were processed as two pools from five mice each, one for males and one for females. RNA was extracted by using Triazol and column chromatography. Quantification of RNA and evaluation of its integrity were performed as described previously (25). The expression of miRNAs was evaluated by microarray analysis and validated by real-time quantitative polymerase reaction (qPCR).

MiRNA microarray analyses were performed by using the 7th generation miRCURY LNA™ microRNA Array (Exiqon, Woburn, MA), which contains 3100 capture probes covering human, mouse and rat miRNAs. In particular, this array covers 1135 mouse miRNAs, which represent the 88.6% of the mouse miRNAs listed in miRBase 19. The miRNA microarray data are available at GEO database (http://www.ncbi.nlm.nih.gov/GEO/, GEO number requested).

In addition, the expression of two miRNAs (miR-322 and miR-326) was validated by qPCR as described previously by using the miRCURY LNA™ Universal RT microRNA PCR reaction kit (Exiqon) (24) and FAM/HEX fluorescent molecular beacons. Fluorescent qCPR amplification products were quantified by taking into account the number of the first positive amplification cycle and referring to the amount of 5S rRNA used as an internal reference standard. For each sample, qPCR was repeated four times in independent experiments and the results were expressed as means ± SE of normalized Fluorescent Units (FU).

Evaluation of systemic cytogenetic damage, lung tumors and other h istopathological alterations

A total of 341 neonatal mice (175 males and 166 females) was available for this study. The mice were divided into 4 groups, as follows. Group A, 94 mice (45 males and 49 females) kept in filtered air for 7.5 months (sham-exposed mice); Group B, 109 mice (55 males and 54 females) exposed to MCS for 4 months, starting within 12h after birth, and then kept in filtered air for an additional 3.5 months (MCS-exposed mice); Group C, 66 MCS-exposed mice (36 males and 32 females) receiving dietary lapatinib (1600mg/kg diet), starting after weaning and continuing daily until the end of the experiment; Group D, 70 MCS-exposed mice (39 males and 31 females) receiving dietary lapatinib (1600mg/kg diet), starting after discontinuation of exposure to MCS (4 months) and continuing daily until the end of the experiment.

At 4 months, peripheral blood was collected from the tail lateral vein from 60 mice, including 20 mice (10 males and 10 females) per each one of Groups A, B and C. The slides were smeared onto slides (two slides/mouse) and stained with May–Grünwald–Giemsa. The frequency of MN normochromatic erythrocytes (NCE) was scored by analyzing microscopically 50 000 NCE in each mouse.

All mice surviving at 7.5 months of age were killed by CO₂ asphyxiation. Complete necropsies of both mice that died prematurely and mice killed after 7.5 months were performed. Lungs, liver, kidney, urinary bladder and all organs with suspected macroscopic lesions were fixed in 10% formalin, cut in standardized sections, stained with hematoxylin and eosin, and subjected to standard histopathological analysis. In particular, the accessory, middle and caudal lobes of the right lung were cut into two pieces each, whereas the cranial lobe was left uncut. The left lung was cut into three pieces. This accounted for a total of 10 lungs sections per mouse to be subjected to standard histopathological analyses. Three sections were analyzed per each kidney, and four sections per liver.

Photographs showing examples of morphological appearance of MCS-induced lesions are shown in our previous papers (16,19).

The results were expressed as incidence, indicating the number and percent of all histopathological lesions and, in addition, as multiplicity for those lesions, such as microadenomas, adenomas and malignant tumors, that are characterized by multiple foci.

Data processing and statistical analysis

Comparisons between groups regarding survival of mice and incidence of histopathological lesions were made by χ ² analysis. Body weights, frequency of MN NCE and multiplicity of microadenomas, adenomas and malignant tumors were expressed as means ± SE of the mice composing each experimental group, and comparisons between groups were made by Student’s t-test for unpaired data. Microarray data were log transformed, normalized and analyzed by GeneSpring® software version 7.2 (Silicon Genetics, Redwood City, CA) after local background subtraction. Expression data were median centered by using the GeneSpring normalization option. Comparisons between experimental groups were done by evaluating the fold variations of quadruplicate data generated for each miRNA. In addition, the statistical significance of the differences was evaluated by means of the GeneSpring ANOVA applied by using Bonferroni multiple testing corrections. As inferred from volcano-plot analysis, differences with P < 0.05 and >2-fold variations between experimental groups were taken as significant.

Results

Subchronic toxicity study and choice of the lapatinib dose

All 20 post-weanling mice used as untreated controls and the 80 mice receiving lapatinib with the diet, at 4 dose levels (250, 500, 1000 and 2000mg/kg diet), survived throughout this 6-week experiment. At any dose, no sufferance or behavioral alterations of mice was observed at daily inspection. The body weight of untreated mice (combined genders) was 16.0±0.47g (mean ± SE) at the beginning of the experiment, and 23.0±0.47, 25.2±0.52, 27.0±0.69, 27.8±0.86, 28.7±0.88 and 29.2±0.95g after 1, 2, 3, 4, 5 and 6 weeks, respectively. At any time and dose and irrespective of gender, administration of lapatinib did not affect the body weight by more than 10%, as compared with controls (data not shown). Hence, we decided to use lapatinib, in all subsequent experiments, at 1600mg/kg diet, which is the 80% of the highest dose tested.

Body weights of the mice used for evaluating molecular biomarkers

Exposure of mice to MCS for 10 weeks, starting at birth, resulted in a significant decrease of body weight after 7–10 weeks. The daily administration of lapatinib did not significantly affect the body weights (data not shown).

Bulky DNA adducts and oxidative DNA damage in lung

Table I summarizes the results relative to measurement of bulky DNA adducts and 8-oxo-dGuo in the lungs of mice, as related either to exposure to MCS for 10 weeks, starting at birth, and/or to treatment with lapatinib for 6 weeks, starting after weaning. Compared with Sham, exposure to MCS resulted in a significant increase of both bulky DNA adducts (14.5-fold in combined genders) and 8-oxo-dGuo (2.9-fold). Administration of lapatinib to MCS-free mice did not affect the baseline levels of these DNA lesions, whereas its administration to MCS-exposed mice resulted in a significant attenuation of both biomarkers.

Table I.

Bulky DNA adducts and 8-oxo-dGuo evaluated by ³²P post-labeling in mouse lung

Treatment	Gender	Adducts/108 nucleotides	8-oxo-dGuo/105 nucleotides
Sham	M	1.3±0.20	1.7±0.25
	F	1.2±0.17	2.1±0.20
	M + F	1.3±0.12	1.9±0.19
Lapatinib	M	1.9±0.35	2.2±0.26
	F	1.1±0.30	1.9±0.12
	M + F	1.5±0.26	2.1±0.14
MCS	M	17.9±1.77**	5.0±0.08**
	F	20.0±3.30**	6.0±0.44**
	M + F	18.9±1.59**	5.5±0.26**
MCS + lapatinib	M	11.0±1.98**,***	4.1±0.30**,***
	F	12.5±1.99**	4.6±0.28*,***
	M + F	11.8±1.35**,****	4.4±0.21**,***

The data are means ± SE of the results obtained in 5 mice/treatment/gender.

*P < 0.01, **P < 0.001, compared with sham-exposed mice of the same gender;

***P < 0.05, ****P < 0.01, compared with MCS-exposed mice.

Expression of pulmonary miRNAs by microarray

Figure 1A shows a scatter-plot relating the expression of 1135 pulmonary miRNAs, as evaluated by microarray analysis, in mice receiving dietary lapatinib with that of sham-exposed mice. The symbols falling above the diagonal belt refer to miRNAs that were upregulated more than 2-fold by lapatinib, and those falling below the diagonal belt refer to miRNAs that were downregulated more than 2-fold. At a glance, it appears that most dysregulated miRNAs were upregulated by lapatinib, but all of them were expressed at rather low levels of intensity. In fact, bidimensional PCA showed that the allocation of symbols representing the average expression of miRNAs in sham-exposed mice and lapatinib-treated mice is not very close but falls in the same quadrant (Figure 2). By restricting the analysis to those miRNAs that were dysregulated to a statistically significant extent, as compared with sham-exposed mice, volcano-plot analysis (not shown) indicated that one of them (miR-3109) was downregulated 2.1-fold (P < 0.05). Other 19 miRNAs (1.7%) were significantly upregulated in lapatinib-treated mice. Their list is shown in Table II, along with the indications of the main functions of lapatinib-upregulated miRNAs, which were inferred from literature data and from the Targetscan database (www.targetscan.org). The functions associated with lapatinib-upregulated miRNAs include DICER (miR-92a), metabolism (miR-223), estrogen receptors (miR-27a, miR-139), stress response (miR-27a), with particular reference to NFκB-mediated stress response (miR-181c, miR-181d), protein repair (miR-27a, miR-322), oncogene suppression (miR-489, miR-511), tumor suppressor genes (miR-27a, miR-34b, miR-885), apoptosis (miR-19b), cell proliferation (miR-19b, miR-20a, miR-292, miR-322), cell cycle arrest (miR-362), cell differentiation (miR-19b, miR-216a, miR-326, miR-341), stem cell differentiation (miR-702), and angiogenesis (miR-19b, miR-292).

Fig. 1.

Scatter-plots relating the expression of 1135 pulmonary miRNAs, as evaluated by microarray analysis, in mice receiving dietary lapatinib with that of sham-exposed mice (A), in MCS-exposed mice compared with sham-exposed mice (B), and in MCS-exposed mice treated with lapatinib with MCS-exposed mice in the absence of the drug (C). Each dot represents a miRNA, whose expression intensity can be inferred from the position on the x and y axes. The central diagonal lines indicate equivalence in the intensity of miRNA expression in the mice treated as indicated in the x and y axes. The outer diagonal lines indicate 2-fold differences in miRNA expression between the indicated treatments. The dots falling in the upper left areas refer to miRNAs whose expression was >2-fold higher in the mice treated as indicated in the y axis, whereas the dots falling in the bottom right areas refer to miRNAs whose expression was >2-fold higher in the mice treated as indicated in the x axis.

Fig. 2.

Bidimensional principal component analysis showing the allocation of sham-exposed mice, lapatinib-treated mice, MCS-exposed mice, and MCS-exposed mice treated with lapatinib according to the overall expression of 1135 pulmonary miRNAs.

Table II.

MiRNAs modulated by MCS and/or lapatinib in mouse lung

MiRNA	MiRNA expression (microarray fluorescence units)
	Sham	Lapatinib	MCS	MCS + lapatinib	Main regulated functions
miR-19b	0.40	1.34*	0.52	1.64*****	Cell proliferation, differentiation, apoptosis, angiogenesis
miR-20a	1.14	1.84*	0.50*	1.96***	Cyclin D1
miR-27a	0.77	1.56*	0.83	1.38***	Tumor suppressor genes, cell proliferation, stress response, protein repair, EGFR, HER-2
miR-34b	0.55	1.03*	0.29*	1.18****	P53
miR-92a	0.59	1.26*	0.70	1.41***	DICER
miR-139	0.33	0.72*	0.41	1.35***	HER-2
miR-181c	0.56	1.39*	0.43	1.12***	NFκB stress response
miR-181d	0.60	1.84**	0.74	1.79***	NFκB stress response
miR-216a	0.20	0.62*	0.34	0.84****	Cell differentiation
miR-223	0.52	0.94*	0.31	0.98***	CYP3A4
miR-292	0.54	1.24**	0.88	2.03***	Cell proliferation, angiogenesis
miR-322	0.21	0.60**	0.33	0.89****	Protein repair, cell proliferation
miR-326	0.30	1.22**	0.38	1.37***	Cell differentiation
miR-341	0.42	1.30*	0.40	1.24***	Cell differentiation
miR-362	0.19	0.75**	0.46	1.15***	Cell cycle arrest
miR-489	0.42	0.89*	0.65	1.34***	Oncogene (PTPN 11) suppression
miR-511	0.20	0.76**	0.34	1.17****	Oncogene (TRIB2) suppression
miR-702	0.69	1.40*	0.59	1.31***	Stem cell differentiation
miR-885	0.37	0.76**	0.47	0.84***	P53

*P < 0.05, **P < 0.01, compared with Sham; ***P < 0.05, ****P < 0.01, *****P < 0.001, compared with MCS-exposed mice in the absence of chemopreventive agent.

Exposure of mice to MCS resulted in an extensive downregulation of miRNAs. Scatter-plot analyses (Figure 1B) showed that the downregulating effect of MCS on miRNA expression, as compared with Sham, mainly affected miRNAs that were expressed at high levels of intensity (upper right part of the expression cloud, red color dots). Indeed, the majority of these miRNA alterations were statistically significant (P < 0.05). In particular, volcano-plot analyses (not shown) revealed that 62 miRNAs out of the 1135 tested (5.5%) were significantly downregulated by MCS. The massive effect of MCS on miRNA expression is confirmed by the fact that MCS and Sham fell far away and in two opposite quadrants at PCA (Figure 2).

The interindividual variability was low, the SE being the 4.8% of the mean in sham-exposed mice and the 7.5% in MCS-exposed mice.

Administration of lapatinib resulted in a partial restoration of MCS-downregulated miRNAs. As shown in Figure 1C, a number of miRNAs were either above or below the 2-fold variation belts when comparing their expression in MCS-exposed mice treated with lapatinib with that in MCS-exposed mice in the absence of the drug. Administration of lapatinib to MCS-exposed mice significantly upregulated the expression of all those miRNAs that had been upregulated by lapatinib in smoke-free mice (Table II). Two of them (miR-20a and miR-34b) had been downregulated by MCS (Table II). When analyzing the data by PCA, MCS + Lapatinib tended to slightly depart from MCS but still fell in the same quadrant (Figure 2).

Expression of pulmonary miRNAs by qPCR

The expression intensity of miR-322, measured by qPCR, was 0.79 ± 0.15 FU (mean ± SE) in the lung of sham-exposed mice, 3.25±0 .61 in the lung of mice treated with lapatinib (P < 0.05 versus Sham), 0.50±0.03 in the lung of MCS-exposed mice, and 4.29±0.15 in the lung of MCS-exposed mice treated with lapatinib (P < 0.01 versus both Sham and MCS).

For miR-326, the expression intensity measured by qPCR was 0.65±0.08 FU (mean ± SE) in the lung of sham-exposed mice, 4.96±1.29 in the lung of mice treated with lapatinib (P < 0.05 versus Sham), 1.38±0.21 in the lung of MCS-exposed mice, 6.12±1.07 in the lung of MCS-exposed mice treated with lapatinib (P < 0.05 versus both Sham and MCS).

Thus, the results obtained by using qPCR are in line with those obtained by using microarray, but the differences related to treatment with lapatinib were more pronounced when using qPCR.

Survival and body weights of the mice used for evaluating cl astogenicity and histopathological alterations

A total of 341 mice (175 males and 166 females) were available for this study. Exposure to MCS did not affect survival. However, compared with sham-exposed mice, the body weight was significantly decreased in MCS-exposed mice since the 1st month until the 4th month, when exposure to MCS was discontinued. Thereafter, MCS-exposed mice tended to recover the body weight gain, and at 7 months of life their body weights were no longer different from those of sham-exposed mice. Administration of lapatinib to MCS-exposed mice did not affect the MCS-related loss of body weight (data not shown).

Clastogenicity in peripheral blood erythrocytes

At 4 months of life, when discontinuing exposure to MCS, a total of 3 million NCE (50 000 NCE in 60 mice) were scored for the presence of MN NCE. Exposure of mice to MCS resulted in a moderate but statistically significant (P < 0.01) increase in the frequency of MN NCE, as compared with sham-exposed mice, in both males (1.9±0.14 versus 1.3±0.01) and females (1.2±0.06 versus 0.9±0.08). Administration of lapatinib to MCS-exposed mice did not significantly affect the clastogenic damage (2.1±0.14 in males and 1.0±0.12 in females). However, the value recorded in MCS-exposed, lapatinib-treated females was no longer significantly different from that recorded in sham-exposed females.

Lung tumors and other histopathological alterations

Table III summarizes the results of histopathological analyses in the lung, liver, kidney and urinary bladder of 341 mice. A number of significant changes were observed in MCS-exposed mice, as compared with sham-exposed mice. In particular, MCS caused a significant increase in the incidence of lesions in the lung, including emphysema, alveolar epithelial hyperplasia, blood vessel proliferation and hemangiomas, microadenomas, adenomas and malignant tumors. The multiplicities of microadenomas, adenomas and malignant tumors were also significantly increased by MCS. In addition, MCS damaged the urinary tract by causing significant increases in the incidences of both tubular epithelial hyperplasia in kidney (both genders) and papillary epithelial hyperplasia in the urinary bladder (males only).

Table III.

Incidence and multiplicity of tumors and other histopathological alterations in the lung, liver and urinary tract of Swiss H mice

Organ: histopathological alteration	Gender	Sham	MCS	MCS + lapatinib (A)	MCS + lapatinib (B)
		(45M + 49F)	(55M + 54F)	(36M + 32F)	(39M + 31F)
Lung
Emphysema: incidence (%)	M	1 (2.2%)	6 (10.9%)	3 (8.3%)	3 (7.7%)
	F	0	9 (16.7%)**	8 (25.0%)***	3 (9.7%)*
	M+F	1 (1.1%)	15 (13.4%)	11 (16.2%)**	6 (8.6%)*
Alveolar epithelial hyperplasia: incidence (%)	M	2 (4.4%)	17 (30.9%)***	13 (36.1%)***	12 (30.8%)***
	F	2 (4.1%)	14 (25.9%)**	11 (34.4%)***	8 (25.8%)***
	M+F	4 (4.3%)	31 (28.4%)***	24 (35.3%)***	20 (28.6%)***
Bronchial epithelial hyperplasia: incidence (%)	M	1 (2.2%)	4 (7.3%)	2 (5.6%)	5 (12.8%)
	F	0	3 (5.6%)	4 (12.5%)*	4 (12.9%)
	M+F	1 (1.1%)	7 (6.4%)	6 (8.8%)*	9 (12.9%)
Blood vessel proliferation and hemangiomas: incidence (%)	M	2 (4.4%)	5 (9.1%)	3 (8.3%)	2 (5.1%)
	F	1 (2.0%)	6 (11.1%)*	1 (3.1%)	2 (6.5%)
	M+F	3 (3.2%)	13 (11.9%)*	3 (4.4%)	4 (5.7%)
Microadenomas
Incidence (%)	M	0	32 (58.2%)***	20 (55.6%)***	17 (43.6%)***
	F	0	29 (52.7%)***	16 (50.0%)***	14 (45.2%)***
	M+F	0	61 (55.9%)***	36 (52.9%)***	29 (41.4%)***
Multiplicity (mean ± SE)	M	0	7.7±1.32***	5.4±1.37***	4.6±1.23***
	F	0	10.2±1.57***	4.1±1.19**	5.3±1.31***
	M+F	0	8.9±1.03***	4.8±0.91***,*****	4.9±0.90***,*****
Adenomas
Incidence (%)	M	1 (2.2%)	14 (25.5%)**	11 (30.6%)***	8 (20.5%)***
	F	2 (4.1%)	13 (24.1%)**	9 (28.1%)**	6 (19.4%)***
	M+F	3 (3.2%)	27 (24.8%)***	20 (29.4%)***	14 (20.0%)***
Multiplicity (mean ± SE)	M	0.04±0.04	4.1±1.13***	3.9±1.30***	1.7±0.80*
	F	0.06±0.05	2.5±1.03***	4.3±1.43***	2.6±1.20*
	M+F	0.05±0.03	3.3±0.76***	4.1±0.95***	2.1±0.69**
Malignant tumors
Incidence (%)	M	0	7 (12.7%)*	3 (8.3%)*	4 (10.3%)*
	F	0	6 (11.1%)*	4 (12.5%)*	3 (9.7%)*
	M+F	0	13 (11.9%)***	7 (10.3%)**	7 (10.0%)**
Multiplicity (mean ± SE)	M	0	0.4±0.16*	0.1±0.08	0.3±0.20
	F	0	0.3±0.15*	0.3±0.23	0.1±0.07
	M+F	0	0.3±0.11**	0.2±0.12	0.2±0.12
Liver
Parenchymal degeneration: incidence (%)	M	1 (2.2%)	3 (5.5%)	1 (2.8%)	2 (5.2%)
	F	1 (2.0%)	3 (5.6%)	6 (18.8%)*,****	3 (9.7%)
	M+F	2 (2.1%)	6 (5.5%)	7 (10.3%)*	5 (7.1%)
Kidney
Tubular epithelial hyperplasia: incidence (%)	M	1 (2.2%)	16 (29.1%)***	3 (8.3%)****	2 (5.1%)****
	F	0	13 (24.1%)***	6 (18.8%)**	3 (9.7%)*
	M+F	1 (1.1%)	29 (26.6%)***	9 (13.2%)**,****	5 (7.1%)*,****
Urinary bladder
Papillary epithelial hyperplasia: incidence (%)	M	1 (2.2%)	7 (12.7%)*	3 (8.3%)	4 (10.3%)
	F	0	0	1 (3.1%)	1 (3.2%)
	M+F	1 (1.1%)	7 (6.4%)*	4 (5.9%)	5 (7.1%)*

A = The oral administration of lapatinib started after weaning and continued until the end of the

experiment (7.5 months).

B = The oral administration of lapatinib started after discontinuation of exposure to MCS (4 months) and continued until the end of the experiment (7.5 months).

*P < 0.05, **P < 0.01, ***P < 0.001, as compared with Sham of the same gender; ****P < 0.05 and *****P < 0.01, as compared with MCS-exposed mice of the same gender, in the absence of chemopreventive agent.

Lapatinib had poor effects on MCS-induced histopathological alterations. Both when given under conditions simulating an intervention in current smokers (column A in Table III) and in ex-smokers (column B), this drug only lowered the multiplicity of microadenomas, an effect that became statistically significant when combining the two genders. In addition, under both conditions lapatinib inhibited the MCS-induced tubular epithelial hyperplasia of kidney, especially in male mice. On the other hand, lapatinib was hepatotoxic to female mice, as shown by an increase in the incidence of liver parenchymal degeneration, which was significant when the drug was administered concurrently with exposure to MCS.

Of the 66 foci of malignant lung tumors detected in MCS-exposed mice, irrespective of treatment with lapatinib, 34 (51.5%) were adenosquamous carcinomas, 14 (21.2%) were squamous cellular carcinomas, 10 (15.2%) were bronchioloalveolar carcinomas, 6 (9.1%) were squamous cell carcinomas in situ and 2 (3.0%) were adenocarcinomas.

Discussion

The present study evaluated for the first time the effects of the antitumor drug lapatinib in a carcinogenesis model using mice exposed to MCS. This complex mixture caused a variety of significant alterations, both genotoxic and epigenetic, such as increases in bulky DNA adduct levels, oxidative DNA damage and extensive downregulation of miRNA expression in lung, systemic clastogenic damage as well as histopathological alterations. These included pulmonary lesions, such as emphysema, alveolar epithelial hyperplasia, blood vessel proliferation and hemangiomas, microadenomas, adenomas, and malignant tumors and damage to the urinary tract, such as tubular epithelial hyperplasia in kidney and papillary epithelial hyperplasia in the urinary bladder.

These findings, which support the interpretation that MCS is a strong promoting agent, are in line with our previous studies in mice exposed to MCS early in life, both in terms of molecular biomarkers (10–15) and of tumors and other histopathological alterations (16–22). Clearly, all these studies have the limitation that the molecular analyses are done on whole tissues. Therefore, they do not evaluate the contributions of different types of cells and cannot discriminate the role of stem cells from that of non-stem cells. It is noteworthy that exposure of mice to MCS resulted in a body weight loss, implying a negative effect either on cell proliferation or survival. In contrast, in target tissues MCS had a mitogenic effect characterized by hyperplasias, preneoplastic and neoplastic lesions, which suggests a stimulation of stem cells. A previous study demonstrated upregulation of stem cell antigen 1 in the lung of mice exposed to smoke since birth (26).

Both MCS and lapatinib were necessarily used at doses that are hardly comparable with those to which humans are realistically exposed. In particular, MCS doses cannot be compared with those to which active smokers are exposed, especially because MCS is inhaled by humans as an undiluted complex mixture, while mice are exposed whole-body. The dose of lapatinib used in the present study (1600mg/kg diet) corresponds to approximately 250mg/kg body weight by the end of our experiment. A phase I trial of lapatinib in children having a median age of 9.3 years indicated that a daily dose of 1800mg/m² is well tolerated (27), which would correspond to 60mg/kg body weight. This is not far away from the dose administered in the present study to mice from puberty to early adulthood. In general, unless using huge numbers of animals, the need for using high doses of test agents is a common drawback in experimental studies. Nevertheless, these studies have the advantage of being performed under well controlled experimental conditions, in the absence of confounding factors and provide mechanistic insights.

Lapatinib had marginal effects on the systemic MCS clastogenicity but protected the lung from MCS-induced nucleotide alterations and, in part, from miRNA expression dysregulation. The attenuation of DNA adduct levels by lapatinib may be ascribed to the fact that this tyrosine kinase inhibitor is an inactivator of CYP3A4, most probably via the formation and further oxidation of its O-dealkylated metabolite to a quinoneimine that covalently modifies the CYP3A4 apoprotein and/or heme moiety (28). Interestingly, we found that, in both smoke-free and MCS-exposed mice, lapatinib upregulated miR-223, which is a known inhibitor of CYP3A4 (29).

On the other hand, the observed attenuation of oxidative DNA damage by lapatinib correlates with upregulation of miRNAs involved in the NFκB-mediated stress response (miR-181c and miR-181d). In fact, the tumor promoters that activate NFκB increase oxidative stress within the cell (30), and exposure of lung epithelial cells to the oxidative stress induced by cigarette smoke stimulated aberrant EGFR phosphorylation/activation (7). One of the miRNAs involved in the NFκB-mediated stress response (miR-181a) is correlated with EGFR expression (31). Similarly, we found that a miRNA involved in P53-mediated stress response (miR-34b) was upregulated by lapatinib, which also counteracted its downregulation by MCS. A miRNA of the same family (miR-34a) exhibited a synergistic interaction with the EGFR inhibitor erlotinib (32).

Interestingly, a bioinformatic study identified 6 clusters of miRNAs that target EGFR in lung cancer (33). One of them included miR-27a, which was also upregulated in the present study by lapatinib in mouse lung. Therefore, it appears that this miRNA, which is involved in the regulation of tumor suppressing genes, cell proliferation, cell response, and protein repair, not only is a predictor of EGFR-related lung cancer in humans but is also targeted by lapatinib in mice. As inferred from the Targetscan database (www.targetscan.org), miR-27a targets both EGFR and HER-2 with a context score of 0.39.

HER-2 is the other tyrosine kinase targeted by lapatinib. Our results provide evidence that miRNAs play a role in lapatinib-related modulation of this receptor. In fact, the oral administration of this drug to mice resulted in the significant upregulation of miR-139 in the lung of smoke-free mice and, even more strikingly, in the lung of MCS-exposed mice. There is an inverse correlation between miR-139 expression in gastric cancer cells and HER-2 levels (34).

Likewise, the observed upregulation by lapatinib of miRNAs involved in cell proliferation, apoptosis, and angiogenesis in mouse lung is consistent with the finding that treatment with this drug was associated in vitro with reduced cell proliferation and apoptotic cell death in A549 bronchoalveolar carcinoma cells as well as with a dramatic reduction in angiogenesis in xenographs of these cells in nude mice treated with lapatinib (6). It is interesting that miRNAs associated with tumor suppression, apoptosis, cell cycle arrest, cell differentiation and angiogenesis are those cellular phenotypes linked to connexin expression and gap junction functions (35). It should be noted that, in mammary tumors induced by methylnitrosourea in rats, lapatinib treatment was associated not only with inhibition of EGFR and HER-2 but also with changes in a number of other signaling molecules, including IGF-1R, Akt, and downstream targets such as GSK3, P27, P53 and cyclin D1 presumably leading to impaired proliferation, apoptosis, or cellcycle arrest (36).

In addition to the analysis of intermediate molecular biomarkers, we evaluated the ability of lapatinib to modulate the histopathological alterations in mice exposed to MCS during the first 4 months of life and killed at the age of 7.5 months. These experiments were carried out by administering lapatinib either after weaning, thus mimicking an intervention in current smokers, or after discontinuation of exposure to MCS, thus mimicking an intervention in ex-smokers. Under both conditions, the only protective effect in the lung consisted in a moderate but statistically significant decrease in the multiplicity of microadenomas. These preneoplastic lesions are larger than hyperplastic foci but, at variance with adenomas, are only detectable at the microscopic level and tend to regress spontaneously (37). Therefore, microadenomas are probably related to MCS-related chronic inflammation, which is a key mechanism in CS carcinogenesis (38,39), and do not necessarily contribute to the development of malignant lung tumors.

This substantially negative finding is not surprising. In fact, the cigarette smoke-activated EGFR was not inhibited by the tyrosine kinase inhibitors AG1478, erlotinib, and gefitinib in lung epithelial cells (7), in spite of the fact that EGFR and its downstream signaling are increased in the bronchial epithelium of smokers (40,41) as well as in the lungs of smoke-exposed rats (42). In addition, lapatinib did not induce a significant number of tumor regressions in NSLC in a phase II trial (9).

In the urinary tract, lapatinib inhibited the induction of kidney tubular epithelial hyperplasia by MCS, especially in male mice. Again, it should be noted that these lesions do not necessarily contribute to the development of renal cancer. Although the potential implication of the EGF/EGFR pathway in renal cell carcinoma has been suspected, the further development of drugs such as lapatinib in clinical trials cannot be supported (43). On the other hand, lapatinib did not affect the incidence of papillary epithelial hyperplasia of urinary bladder, which was selectively increased by MCS in males, consistently with the known ability of estrogens to reduce chronic bladder inflammation (44).

The murine model used was able to point out adverse effects of lapatinib, and specifically an increase in the incidence of liver parenchymal degeneration, which was statistically significant in females treated with the drug concurrently with exposure to MCS. It is noteworthy that hepatobiliary adverse effects have been observed in women suffering from metastatic breast cancer treated with lapatinib (45) and that this drug was hepatotoxic in an experimental rat model (46). Lapatinib is sufficiently potent in inducing hepatic injury as to require a boxed label warning (47). The mechanism of lapatinib hepatotoxicity appears to be based on the fact that this drug is an inactivator of CYP3A4 (28). In fact, dexamethasone, a CYP3A4 inducer, increased the formation of hepatotoxic lapatinib metabolites (48). As previously mentioned, we found that, in the lung of both smoke-free and MCS-exposed mice, lapatinib upregulated the CYP3A4 inhibitor miR-223.

In conclusion, the results of the present study indicate that lapatinib is able to inhibit the MCS-related increase of bulky DNA adducts and oxidative DNA damage in mouse lung. These nucleotide alterations can be restored via DNA repair mechanisms and therefore are not necessarily predictive of the evolution towards fixation of DNA damage and cancer initiation. Our experimental data provide evidence that early events of DNA damage may not be predictive of cancer development. There is need to evaluate post-genomic domains for a better predictivity (49). For this reason, we evaluated modulation of miRNA expression. Lapatinib was able to upregulate several miRNAs involved in a variety of important cellular functions and carcinogenesis mechanisms, including modulation of both EGFR and HER-2. However, the affected miRNAs were generally expressed at low levels of intensity. The only two miRNAs that were significantly downregulated by MCS and upregulated by lapatinib were miR-20a, an important cyclin D regulator (50), and miR-34b, an established P53 effector (51). The conclusion that lapatinib has a low impact in the smoke-related lung carcinogenesis models used is supported by the findings that, in the medium term, this drug poorly affected the systemic genotoxicity of MCS and had just moderate protective effects towards preneoplastic lesions in lung and kidney. On the other hand, it exhibited some toxicity in the liver of MCS-exposed mice.

Funding

National Cancer Institute (contract HHSN-261200433000C); Bulgarian Ministry of Education, Youth and Science.

Conflict of Interest Statement: None declared.

Glossary

Abbreviations:

EGFR

epidermal growth factor receptor

HER-2

human epidermal growth factor receptor 2

MCS

mainstream cigarette smoke

ER

estrogen receptor

MN

micronucleated

qPCR

real-time quantitative polymerase reaction

NCE

normochromatic erythrocytes.