has been approved for the. Toxicology Program by. Rajesh Agarwal, Chair. Alvin M. Malkinson, Advisor. David Ross. Robert J. Mason. David W.H. Riches.
ALVEOLAR MACROPHAGES ACCELERATE LUNG TUMORIGENESIS THROUGH INSULIN-LIKE GROWTH FACTOR-1 AND CYTOKINE PRODUCTION by JASON MICHAEL FRITZ
PR
EV
IE
W
B.S., University of Denver, 2005
A thesis submitted to the
Faculty of the Graduate School of the University of Colorado in partial fulfillment of the requirements for the degree of Doctor of Philosophy Toxicology Program 2011
UMI Number: 3467213
All rights reserved INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted.
IE W
In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion.
UMI
EV
Dissertation Publishing
PR
UMI 3467213 Copyright 2011 by ProQuest LLC. All rights reserved. This edition of the work is protected against unauthorized copying under Title 17, United States Code.
ProQuest LLC 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 48106-1346
This thesis for the Doctor of Philosophy degree by Jason Michael Fritz has been approved for the Toxicology Program by
W
Rajesh Agarwal, Chair Alvin M. Malkinson, Advisor
IE
David Ross
Robert J. Mason
Date M | l ^ | \\
PR
EV
David W.H. Riches
11
Fritz, Jason Michael (Ph.D., Toxicology) Alveolar macrophages accelerate lung tumorigenesis through insuhn-like growth factor-1 and cytokine production Thesis directed by Professor Alvin. M. Malkinson
Chronic inflammatory lung disease or exposure to noxious agents such as cigarette smoke greatly increases lung cancer risk, highlighting the importance of chronic inflammation in the progression of this disease. Macrophage numbers increase with
IE W
tumor progression in chemically-induced mouse models of non-small cell lung cancer (NSCLC), and high pulmonary macrophage content correlates with poor prognosis in human NSCLC. Despite this positive association, the specific role of alveolar
PR EV
macrophages in lung tumorigenesis is unclear.
Media conditioned by alveolar macrophages (M0CM) stimulates neoplastic lung proliferation in vitro, and the responsible factors are primarily 40%), regardless of smoking status (39; 75). Reports of
IE
mutation incidence vary widely as a function of patient ethnicity, classification of NSCLC subtype, and study year. One recent meta-analysis of 700 lung ACs found that
PR
EV
-30% bore KRAS mutations, and suffered 2.4x higher risk of death within 2 yrs. of
Breast Prostate Colon Pancreas Lung (Smoker) Lung (NS) Other
Figure 1.1 Annual cancer deaths by tissue site. U.S. yearly cancer death incidence, expressed as percentage originating in lung, pancreas, colon, prostate and breast, as indicated. Death from lung cancer further delineated into smoking (smoker) and neversmoker (NS) populations. (6; 102)
2
diagnosis vs. other mutations (80; 98). Another study reported 18-23% of NSCLC patients bearing KRAS mutations; while KRAS mutation incidence was not significantly different between smokers and never-smokers, the spectrum of KRAS mutations was dramatically different between the two populations (69; 198). Only -10% of Western NSCLC patients bear EGFR mutant tumors, but EGFR mutations are much more frequent in Asian populations, and KRAS mutations relatively infrequent (80; 183-185). Mutations are important prognostic indicators for lung cancer treatment, as patients with EGFR mutations respond favorably to EGFR-specific tyrosine kinase inhibition therapy,
W
which is contraindicated for patients bearing KRAS tumors (41; 183).
IE
Chemically-induced mouse models of human non-small cell lung cancer. Murine lung tumors share numerous morphological, histologic and molecular characteristics with
EV
human AC, making mouse primary lung tumorigenesis models analogous to human disease, more so than tumor xenograft models (111; 139; 219). Mouse lung tumors arise
PR
spontaneously, and can also be induced with chemicals, transgenic oncogenes, and radiation (140). "Field carcinogenesis" is the idea that an entire tissue or organ is made more susceptible to carcinogenesis, e.g. cigarette smoke induces a field carcinogenic effect on exposed lung airways. Chemically-induced mouse lung carcinogenesis more closely recapitulates the field carcinogenesis that is induced in human lungs than transgenic mouse lung tumor models which employ targeted expression of a single mutant gene (63; 103). In mouse models of human NSCLC, differences in the frequency and nature of mutations depend on genetic background, carcinogen, and method of exposure (140; 141). For instance, transplacental carcinogenesis with N-ethylnitrosourea at different gestational stages shifts Kras mutations from codon 12 to 61, while ethyl
3
carbamate (urethane) induces different frequencies of codon 61 Kras mutations in adenomas (AD) vs. AC in adult mice (79; 176). Metabolites of 3-methylcholanthrene (MCA), a polycyclic aromatic hydrocarbon formed by carbon pyrolysis and a component of cigarette smoke, mutate Kras at codon 12 (11; 259). MCA treatment alone induces 0-5 lung tumors/mouse, in a strain and dosedependent manner (15; 47). When MCA administration is followed by chronic exposure to the non-carcinogenic pneuomotoxin, butylated hydroxytoluene (BHT), lung tumor multiplicity is increased in strains sensitive to BHT-induced inflammation (15). Mice
EW
metabolize BHT to BHT-OH via cytochrome P450 2B-mediated hydroxylation of a tertbutyl group, which is further oxidized to the quinone methide BHT-OH-QM (49; 145;
EV I
233). The BHT-OH metabolite can recapitulate all the pneumotoxic effects of parent BHT compound when administered at one-fourth the dose (120). Metabohsm of BHT to BHT-OH is species-specific: rats do not metabolize BHT -> BHT-OH, and do not
PR
experience BHT-mediated pneumotoxicity (145). Tumor promoting effects of BHT appear tightly linked to chronic lung inflammation and macrophage infiltration, induced by BHT in sensitive mouse strains such as A/J and BALB/cBy (cBy) (15). MCA initiated tumors progress slowly from 6-20 wks.; small hyperplastic lesions become microscopic adenomas, and -25% of these progress to macroscopic ADs (179). AC incidence increases between 20-42 wks. with no significant change in tumor multiplicity, indicating that MCA-induced ADs are precursor lesions for mouse lung AC (140). Urethane was previously used as a dental anesthetic since high acute doses induce gentle anesthesia in both rodents and humans, but dental use ceased in circa 1945 after urethane was discovered to increase mouse lung tumor multiplicity by 15-fold. Despite
4
this, urethane was used as a co-solvent for hydrophobic analgesics in Japan into the early 1970's, until found to be absorbed trans-placentally (18; 171; 175). Urethane is currently used commercially in the production of pesticides, fumigants and cosmetics. Urethane is now known to induce a wide spectrum of mouse tumors, with cancer type and organ location based upon strain and dose (160; 174; 228). Diethyl pyrocarbonate (DEPC), an anti-microbial agent previously added to wine, reacts with ammonia at physiological pH to form urethane (132). This prompted intensive investigation into the presence of urethane in the food chain, resulting in soft drink safety limits set at 10 ppb, with higher
W
limits of 100-350 ppb allowed in alcoholic beverages (18). Urethane is abundant in
IE
cigarette smoke, foods involving fermentation such as yogurt, cheese and bread, and alcoholic beverages such as whiskey, rum and wine (18; 266). The chemical
EV
fermentation process does not create urethane; however, other substrates present during fermentation such as heat, fungal carbamyl phosphate, urea and/or citrulline
PR
spontaneously react, producing urethane (100). For example, toasting bread doubles urethane content (266). The average daily human urethane intake (ADI) is estimated at 10-20 ng/kg body weight/day, or 1.4 ug /day for a 70 kg person, coming mainly from bread products (266). Smoking one pack of cigarettes doubles daily urethane intake, as will drinking a single glass of wine. Urethane levels roughly correlate to alcohol content in liquors, although exceptions such as stone-fruit brandy and saki add 50-100 ug urethane to ADI per serving! (266). The cumulative risk of developing cancer for a packa-day smoker and heavy drinker (3-5 servings of whiskey or 1 serving of brandy/day) is 0.02% based upon urethane-content alone; this comes to ~1 urethane-induced cancer per 5,000 members in this sub-population (266). Despite nearly 60 years of epidemiological
5
and experimental evidence and its ubiquitous presence in the food chain, urethane was only recently classified by the National Toxicology Program as "reasonably anticipated to be a human carcinogen" in 2000 (93). Urethane is absorbed and metabolized rapidly, regardless of administration route; 90-95% is exhaled as CO2, ethanol and ammonia within 2-8 hrs., with 3-4% excreted in the urine as either parent compound or metabolites (18; 93). While urethane could be metabolized through either cytochrome P450 (CYP) or esterase pathways, carcinogenic metabolites result from two oxidation steps catalyzed by CYP2E1; ethyl carbamate ->
EW
vinyl carbamate -> vinyl carbamate epoxide (74; 122). Vinyl carbamate is considered a proximal carcinogen to urethane, since vinyl carbamate induces similar lung tumor
EV I
multiplicity, and these progress to AC more rapidly than urethane at l/20th the dose (5; 125; 138). However, vinyl carbamate does not directly bind DNA and requires metabolic activation to become carcinogenic: only the final epoxide product adducts to DNA, with
PR
mutations resulting from mis-match repair (18). Thorough investigation has determined that only the ethyl carbons adduct to DNA in vitro or in vivo, and replacing the ethylchain with methyl-, propyl- or butyl- groups renders the carbamate non-carcinogenic (18). CYP2E1 is responsible for 96% of the urethane that is metabolized to CO2, and is required for vinyl carbamate epoxide formation. Other CYPs contribute 3% to overall urethane metabolism, while unidentified esterases convert 0.5% in vivo (74; 93). Interestingly, urethane-induced anesthesia results from the unmetabolized parent compound, which explains the rapid recoveries of experimental animals (unpublished observations), as serum urethane levels drop precipitously within 30-60min. following administration (18; 93). While CYP2E1 metabolizes urethane in humans and rodents
6
alike, rodents metabohze urethane 12-fold more rapidly, with serum clearance estimated at 50 ug/mL blood/hr vs. 4 ug/mL blood/hr in humans (18). Both bronchiolar Clara cells and alveolar type II pneumocytes metabohze urethane, and lung tumors bearing Kras codon 61 mutations may arise from either cell type (155; 172; 230). Urethane induces -40 lung tumors in A/J mice in the absence of subsequent chemical promotion, thus removing a possibly confounding variable from tumor growth analysis (176; 229). These lung tumors progress from hyperplastic foci -^ microadenoma -> AD -> AC over a
carcinogen injection (64).
EW
period of-50 wks.; 50-70% of observed AC arise within adenomas by -35 wks. after
A similar tumor progression and time-frame is also observed in lung tumors
EV I
induced by 4-(methylnitrosamino)-l-(3-pyridyl)-l-butanone (NNK), another carcinogen found in cigarette smoke (17). NNK is metabolized by CYP2B1 in mice (or human CYP2A5), and adducts are detected in both Clara and type II lung cells, similar to
PR
urethane. While lesions were only found along alveolar spaces, a few type II cells from ADs and ACs contained morphological characteristics of Clara cells, suggesting that neoplastic lung cells adopt those cellular properties which confer the greatest growth advantage (17). Eighty percent of NNK-initiated tumors bear Kras codon 12 mutations similar to those induced by MCA, while the remaining 20% have Kras mutations at codon 61 like those induced by urethane (17; 176). Transgenic mouse models of human non-small cell lung cancer. Chemical agents have been used to study tumor induction and progression in animal models for over 100 years, since Yamagiwa's first observations of coal tar-induced rabbit ear sarcomas in 1905 (255; 256). Ten years ago, two transgenic mouse lung cancer models were
7
