Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Considerable attention has recently been focused on the adverse health effects of particulate air pollution (1). Exposures to these pollutants induce decrements in lung function and enhance respiratory illness (e.g. chronic cough, bronchitis, pneumonia), and have also been attributed to a number of significant acute mortality episodes (1, 2). Although the epidemiologic evidence strongly supports a relationship between adverse health effects and particulate exposure, the precise toxicologic or physiologic mechanisms have not been elucidated.

It is postulated that some subpopulations are at increased risk to the effects of particle exposure. These populations include children, the elderly, and patients with pre-existing chronic heart diseases, chronic obstructive pulmonary diseases, and compromised immune systems (3). Animal studies have found cardiotoxicity in healthy dogs exposed to concentrated urban air (8), and cardiac arrhythmia induction in rats exposed to fugitive residual oil fly ash (ROFA) (9). Pulmonary hypertension induced by monocrotaline exacerbated the cardiac arrhythmia effects of ROFA (9) and concentrated air particles (10), lending support for the role of pre-existing disease as an important risk factor in particulate-induced morbidity and mortality.

Genetic background is another potentially important risk factor that may influence an individual response to particle challenge. The present study was designed to identify murine chromosomal loci that determine responsivity to particles and, by comparative mapping, suggest human candidate susceptibility genes. We have developed a murine model to investigate the exposure effects of sulfate (SO₄²)-associated carbon black particles on lung function (11, 12). Combination of carbon black with sulfur dioxide (SO₂) in the presence of high humidity creates an acid-coated carbon particle (ACP) species (12) that mimics one component of acid aerosols found in high concentrations in many large metropolitan areas, particularly in the eastern United States (15). The chosen phenotype for these studies was Fc receptor-mediated phagocytic function of alveolar macrophages (AMs) because these cells represent an important "first line" in immune host defense. The integrated activity of the phagocytic and immune system provides pulmonary defense against bacteria and other environmental pathogens. Phagocytosis of the invading pathogen is one of the events coordinated with intracellular killing (oxidative burst) to maintain sterility of the respiratory tract. Phagocytic function has been used as an indicator of pulmonary toxicity and, particularly, AM integrity in a number of models (e.g., 16, 17). A disruption of any of the steps involved in intracellular killing, including phagocytosis, compromises host defense. Increased risk of lower airways infection after particle exposure has been reported in several epidemiologic studies (3, 4, 7), and is presumably mediated through effects on lung defense and AM function (17).

Four-hour exposure to ACP induces reversible impairment of Fc receptor-mediated phagocytic function in sensitive strains of mice (18). Previous studies have indicated that neither SO₄² alone nor carbon particles alone impairs AM function (12). Further, the ACP challenge enhances mortality in mice infected with the influenza A/PR8/34 virus (19). The genetic linkage studies described below used inbred C3H/HeJ (C3) and C57BL/6J (B6) mice. These strains of mice display two easily distinguishable AM function phenotypes after ACP challenge, which we termed resistant and responsive (susceptible), respectively (18). The phenotypic difference exhibited by the two strains forms the basis of the present investigation.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals

Inbred (B6, C3) and hybrid (B6C3F₁/J) mice (6 to 8 wk of age) were purchased from the Jackson Laboratories (Bar Harbor, ME). Animals were housed in microisolation cages within an antigen- and virus-free room. Water and mouse chow (Agway Pro-Lab RMH 1000) were provided ad libitum. Sentinel animals were examined periodically (titers and necropsy) to ensure that the animals remained free of infection. Mice were handled in accordance with the standards established by the United States Animal Welfare Acts set forth in National Institutes of Health guidelines and the Johns Hopkins University School of Hygiene and Public Health Animal Care and Use Committee.

Breeding Studies

Breeding studies between the B6 and C3 mice were conducted in our animal facilities. The following crosses were generated: male B6C3F₁/J were bred with B6 and C3 females to produce B6 backcross (B6:BX) and C3 backcross (C3:BX) progeny, respectively; B6C3F₁/J were intercrossed to produce B6C3F₂ progeny. Progeny were weaned at 3 to 4 wk of age, separated according to sex and housed in microisolation cages until they reached the appropriate age for experimentation (6 to 8 wk).

ACP Generation and Exposures

Carbon black Regal 660 with a specific surface area of 90 m²/g (empirical formula C₉₁₀H₃₄O₁₀; composition: 96.90% carbon, 0.30% hydrogen, and 1.42% oxygen; 25 nm parent size; a generous gift of Cabot Corp., Billerica, MA) was used for these studies. The chamber characteristics, aerosol and SO₂ generation and monitoring, particle size analysis, and particle-associated SO₄² analysis have been detailed previously (11). Briefly, a flow-past nose-only inhalation chamber was used for the exposure of mice to a mixture of carbon black aerosol and SO₂ at 85% relative humidity (RH). The carbon black aerosol was generated with a Wright dust feed (BGI, Inc., Waltham, MA) and monitored with a real-time monitor (RAM-1; M.I.E. Technologies, Inc., Bedford, MA). Time-weighted carbon black aerosol concentrations were measured using 25-mm, 0.2-µm-pore-size membrane filters (HT200; Gelman Sciences, Ann Arbor, MI) held in an electrically conductive, 25-mm-diameter, open-faced filter cassette (#01-038-1; Fisher Scientific Co., Pittsburgh PA). Aerodynamic aerosol size distributions were measured using a 10-stage Sierra cascade impactor (Anderson, Inc., Atlanta, GA). Metered SO₂ from a pressurized tank containing 1.5% SO₂ in air (Matheson Gas Co., East Rutherford, NJ) was mixed with diluent before entering the exposure chamber. The concentration of SO₂ was monitored continuously with a pulsed fluorescence SO_x monitor (Thermo Environmental Instruments, Inc., Franklin, MA). Particle-associated SO₄² analysis was performed by a modification of ASTM Method No. 4500 (14). Mice were exposed for 4 h at RH of 85% to a target concentration of 10 mg carbon black/m³ (measured gravimetrically) and 10 ppm SO₂. Average concentration of particle-associated sulfates was 285.4 ± 30.6 µg/m³. Mass median aerodynamic diameter (MMAD) was 0.29 µm with a geometric standard deviation (GSD) of 2.7. It should be noted that not all SO₂ is converted to SO₄² in the mixing chamber. Therefore, mice were exposed to a mixture of SO₂ and particle-associated SO₄². However, we have demonstrated previously that SO₂ alone (10 ppm) has no effect on AM phagocytic function in this model (12). Therefore, macrophage effects reported in this study are attributed to particle-associate SO₄².

Bronchoalveolar Lavage and Cell Preparation

AM retrieval and assessment of pulmonary inflammation were done using bronchoalveolar lavage (BAL). Mice were killed by cervical dislocation and lungs were lavaged four times (0.35 ml/ kg) in situ with sterile phosphate-buffered saline (PBS) supplemented with 0.01% ethylenediaminetetraacetic acid (EDTA). The PBS contained (in g/liter): 5.43 NaCl; 0.50 Na₂EDTA; 0.57 Na₄EDTA; 4.73 NaH₂PO₄; and 0.40 KH₂PO₄. Recovered BAL fluid (BALF) from each mouse was pooled and immediately cooled to 4°C. Lavage returns were then centrifuged (500 × g, 4°C) and supernatants decanted. Cells were resuspended in 0.8 ml RPMI 1640 (GIBCO Co., Grand Island, NY), supplemented with 10% newborn calf serum, and counted with a hemocytometer. Cell viability was determined by the method of trypan blue stain exclusion.

Fc Receptor-Mediated Phagocytosis Assay

AM phagocytosis was determined as previously described (20). Three 200-µl aliquots of BALF cell suspensions were allowed to adhere to 22-mm² albumin-coated glass coverslips in 35 × 10 mm plastic petri dishes (Becton-Dickinson, Franklin, NJ) for 45 min (37°C, 5% CO₂, 95% RH). After monolayering, the fluid was removed and immediately replaced with 1.5 ml of 0.5% sensitized sheep red blood cells (RBC) (Becton-Dickinson) in RPMI medium and the resulting suspension was incubated at 37°C for 45 min. After removal of the RBC by aspiration and washing of the monolayers with RPMI medium, noningested RBC were hypotonically lysed for 10 s, followed by several rinses with culture medium. Monolayers were then dried, fixed with methanol, and stained with Wright-Giemsa. Stained monolayers were read microscopically at ×1,000 to quantify the percentage of AMs containing RBC and the number of RBC ingested per phagocytic macrophage. A total of 100 AMs was scored on each slip monolayer, with three monolayers counted per animal. Values (phagocytic index) were expressed as percent of the air-exposed control mean.

Experimental Protocol

Initially, the kinetics of AM phagocytosis dysfunction was characterized in the B6 and C3 progenitors. Mice from each strain were divided randomly into two groups per strain and were placed in single-animal cylindrical restrainers. One group from each strain was exposed to ACP for 4 h. Age-matched mice from each strain were exposed simultaneously to filtered air to serve as controls. Exposure conditions and recovery period (0, 1, 3, 7, and 14 d) were determined from previous studies (18). For segregation and linkage analyses F₁, F₂, and BX mice were exposed to ACP as described previously, and AM phagocytic function was assessed after 3 d. This time point was chosen to phenotype the progeny because the kinetics study identified the greatest difference in phagocytic function between progenitor B6 and C3 mice at this time (see RESULTS).

Segregation Analysis

The following formula from Wright (21) was used to estimate the number of genes that segregate with ACP-induced dysfunction of AM Fc receptor-mediated phagocytosis: n = (P2  F₁)²/4(²_F2  ²_F1), where n is an estimate of the number of independent loci; F₁ and P2 are the mean phagocytic index responses to ACP exposure in B6C3F₁/J and B6 mice, respectively; and ²_F2 and ²_F1 are computed variances of the F₂ and B6C3F₁/J cohorts, respectively. This formula is based on the assumptions that there is semidominance at all loci and that all loci make equal contributions to the trait.

The data analysis program, Statistical Analysis for Genetic Epidemiology (S.A.G.E.), was used to perform segregation analyses (22). Two sets of models were tested, homoscedastic and heteroscedastic. Both sets were used to find the best fit of the data sets, whether the raw data were used (heteroscedastic) or the data were forced to normalization by power transformation (homoscedastic). (For a complete description of the S.A.G.E. program, see Reference 22.) The CLUSTR subprogram of S.A.G.E. was used to estimate group means and variances, and to identify the power transformation that best normalized the data in the genetically homogeneous groups (i.e., B6, C3, B6C3F₁/J). This transformation was applied to the entire data set (B6, C3, F₁, both BX, and F₂) in the subsequent segregation analyses. The goodness of fit of inheritance of homoscedastic models, including one-locus, two-locus, mixed loci, and polygenic, were evaluated using the subprogram BCROSS. One-locus and polygenic models of inheritance were also evaluated in heteroscedastic models with BCROSS. Heteroscedastic models do not assume normality and data are not transformed to force normality. Comparisons were made between each inheritance model and the unrestricted model to determine whether the restrictions placed on a model significantly decreased its likelihood. In the segregation analysis program, parameters that could be restricted included any of the group means, common variances, littermate variances, and the recombination fraction between two loci. The likelihood of each model was compared with the unrestricted model using chi square (²) analysis and Akaike's Information Criterion (AIC), which is defined as: AIC = 2 (log likelihood  number of restrictions). Any restriction placed on a model lowers the maximum likelihood of the model, regardless of the goodness of fit. AIC was used to correct for this decrease in maximum likelihood across models in which different restrictions were applied. Of all the models tested, that with the smallest AIC number is considered the most likely to explain the data. Degrees of freedom were calculated as the difference between the number of restrictions in the model tested and those in the general model.

DNA Extraction and Genotyping

DNA was extracted from a kidney of each phenotyped animal and prepared for polymerase chain reaction (PCR). PCR reactions were run in 96-well plates with 12.5 µl total volume: 1 µl DNA (80 ng), 1.25 µl 10× reaction buffer (500 mM KCl, 100 mM Tris [pH 8.3], and 15 mM MgCl₂), 0.25 µl 10 mM deoxynucleotide triphosphates (equal mixture of deoxyadenosine, deoxycytidine, deoxyguanosine, and deoxythymidine triphosphates), 0.5 µl unlabeled primer (10 µM), 0.4 U Taq polymerase, and 0.5 µl [³²P]adenosine triphosphate 5'end-labeled primer (ratio of labeled to unlabeled primer is 0.15), and brought to volume with double distilled (dd)H₂O. Primers for simple sequence-length polymorphisms (SSLPs) that differed between B6 and C3 progenitors (i.e., informative for linkage mapping) were purchased from Research Genetics (Huntsville, AL). Amplification was performed for 30 cycles (94°C for 30 s, 55°C for 30 s, 72°C for 30 s), preceded by a denaturation step of 10 min at 94°C and followed by 7 min at 72°C. PCR products were resolved on 6% acrylamide gels.

Linkage Analyses

Linkage analyses were initiated by scanning the entire genome for associations between SSLPs and the ACP phenotype in 25 selected high-responder (n = 13) and nonresponder (n = 12) mice from the F₂ cohort (i.e., selective genotyping) (23, 24). Interval analyses were performed by fitting a regression equation for the effect of a hypothetical quantitative trait locus (QTL) at the position of each SSLP marker and at 1-centimorgan (cM) intervals between SSLPs. The dominance properties of each putative QTL were evaluated by performing interval analyses using free, additive, recessive, and dominant regression models. The regressions and significance of each association (likelihood ratio ² statistic) were calculated by the Map Manager QTb27 program, which is distributed electronically and available at http://mcbio.med.buffalo.edu/mmQT.html (25). Putative QTLs were further analyzed by including the entire cohort and additional markers within the chromosomal interval identified by selective genotyping. Permutation tests were performed on the phenotype and genotype data to establish empirically the significance thresholds of all QTL mapping results (Map Manager QTb27 and following the methods of Churchill and Doerge [26]). For the genome scan, 10,000 permutations were performed to establish significant and suggestive linkage threshold values. These values correspond to the genome-wide probabilities proposed by Lander and Kruglyak (27).

Statistics

Statistical analysis of ACP-induced AM dysfunction in B6 and C3 mice was done by three-factor analysis of variance. The factors were strain (B6 and C3), exposure (ACP, air), and time (0, 1, 3, 7, and 14 d). The Student-Newman-Keuls a posteriori multiple-range procedure was used to compare means (28). Significance was accepted at P < 0.05.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Kinetics of ACP-Induced AM Dysfunction

Relative to air-exposed controls, the AM phagocytic function of B6 mice was significantly (P < 0.05) depressed 1, 3, and 7 d after exposure, and the AM responses to ACP were attenuated by 14 d after exposure (Figure 1). There were no significant (P > 0.05) ACP effects on AM phagocytic function in C3 mice at any point after exposure. As demonstrated previously (18), there were no statistically significant (P > 0.05) increases in lavageable inflammatory cells or total BALF protein (a marker of epithelial hyperpermeability) in either strain at any time point following exposure (data not shown).

View larger version (16K): [in this window] [in a new window]   Figure 1.   Time course of Fc receptor-mediated phagocytosis of AMs from B6 and C3 mice 0, 1, 3, 7, and 14 d after exposure to acid-coated particles or filtered air. Mice were exposed for 4 h at 85% relative humidity to a target concentration of 10 mg/m3 and 10 ppm SO2. Average concentration of particle-associated sulfates was 285.4 ± 30.6 µg/m3. MMAD was 0.29 µm with a GSD of 2.7. Age-matched B6 and C3 mice were exposed to filtered air to serve as controls. Phagocytic index values are expressed as percent of the air-exposed control mean at each respective time point. Means ± standard error of the mean (SEM) are presented. *P < 0.05 for air versus particle exposure. +P < 0.05 for B6 versus C3.

Segregation Analyses

To understand further the role of genetic background in susceptibility to ACP effects on AM phagocytic function, segregant (BX and intercross) and nonsegregant (F₁ hybrid) populations derived from B6 and C3 progenitors were phenotyped. The frequency-response distributions of phagocytic indices after ACP challenge for each of these cohorts are shown in Figure 2. Distributions of the AM responses in F₁ and C3:BX populations were similar to that of the C3 progenitor, and the means did not differ significantly (P > 0.05) among the three groups (Table 1). In the segregant B6:BX and F₂ populations, the ranges of the AM phagocytic responses to ACP challenge overlapped those of B6 and C3 progenitors, and the means were significantly different from those of both (i.e., intermediate) (Table 1).

View larger version (30K): [in this window] [in a new window]   Figure 2.   Frequency distribution of the Fc receptor-mediated phagocytosis index (percent of control) from B6 and C3 mice and their progeny 3 d after 4-h exposure to ACPs. Male C3 mice were bred with B6 females to produce B6C3F1/J hybrids. Male B6C3F1/J mice were bred with B6 and C3 females to produce B6:BX and C3:BX progeny, respectively; B6C3F1/J were intercrossed to produce B6C3F2 progeny.

The number of loci estimated to segregate with the ACP susceptibility phenotype in these cohorts was estimated by the formula of Wright (21). The minimum number of loci as determined by this formula was 1.27. We also used S.A.G.E. (22) to estimate the number of segregating ACP susceptibility genes in this model. Among the homoscedastic models tested to explain the segregation data, the highest likelihood was for the two unlinked loci general hypothesis, as shown in Table 2. However, all of the homoscedastic models tested were significantly different from the general unrestricted model. The AIC was consistent with the likelihood and ² analyses.

Linkage Analyses

The animals selected for the initial genome screen (n = 25, or 50 meioses) were genotyped at 147 SSLP markers spaced to provide complete coverage of the mouse genome with > 95% confidence (24). To conform to assumptions of normality and homoscedasticity (homogeneity of variances) for the analyses, the phagocytic index (phenotype) data from B6, C3, and F₂ animals were transformed with the power transformation (0.7858) calculated by the CLUSTR subprogram of S.A.G.E. (22).

The free regression model, which fits separate regression coefficients for additive and dominance components, was the best predictor of the trait value as a function of the QTL genotype. Interval analyses identified a significant QTL on chromosome 17, and four suggestive QTLs on chromosomes 10, 11, 14, and 18 (Figure 3).

View larger version (39K): [in this window] [in a new window]   Figure 3.   A genome-wide search for QTLs by selective genotyping of the F2 cohort, i.e., the most extreme high and low responders to ACP challenge. The x-axis for each plot is the length of the chromosome in centimorgans. The y-axis is the likelihood ratio 2 statistic. The upper and lower horizontal lines in each plot represent significant and suggestive linkage thresholds, respectively. Putative QTLs on chromosomes 10, 11, 14, 17, and 18 that were identified by the genome scan were further analyzed with entire F2 cohort (see MATERIALS AND METHODS).

Each of the putative QTLs was analyzed further by including the entire F₂ cohort (n = 105 animals, 210 meioses) and additional SSLPs. Only the chromosome 17 QTL exceeded the threshold value for statistically significant linkage as determined empirically by permutation test (Figure 4). The amount of the total trait variance explained by the QTL at SSLP D17Mit125, the marker with the highest likelihood ratio statistic in the QTL, was approximately 30%. The regression coefficient values for the additive effect in this model largely followed the likelihood ² statistics for linkage.

View larger version (21K): [in this window] [in a new window]   Figure 4.   Plot of the likelihood ratio 2 statistic (solid line), additive regression coefficient (dotted line), and dominance regression coefficient (dashed line) for the association of ACP-induced AM dysfunction phenotype with polymorphic SSLP markers (open circles) on chromosome 17. Interval mapping was done by simple linear regression (see MATERIALS AND METHODS). SSLP markers are positioned on the linkage map according to the Mouse Genome Database. The upper and lower horizontal lines represent significant and suggestive linkage thresholds, respectively, as determined by permutation test (Map Manager QTb27) (26).

The QTL on chromosome 11, but not chromosomes 10, 14, and 18, exceeded the threshold for suggestive linkage after the remaining F₂ animals were considered in the analysis. The amount of total trait variance explained by this QTL (at D11Mit29) was approximately 13%. Composite interval mapping was done to determine the potential influence of suggestive QTLs identified by selective genotyping on linkage of the ACP phenotype to chromosome 17. There were no detectable effects on the chromosome 17 QTL when controlling for the suggestive QTLs on chromosomes 10, 11, 14, and 18. Results of the linkage analyses are therefore in agreement with the segregation analyses and suggest that a major QTL on chromosome 17 and a minor QTL on chromosome 11 account for a significant portion of the genetic variance in susceptibility to immune dysfunction induced by ACP exposure.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Epidemiologic studies have associated exposures to high particulate and SO₂ concentrations with increased morbidity and mortality in urban cities throughout the United States and other industrialized countries. The biologic basis for interindividual variability in response to these exposures has not been determined. There is evidence to suggest that subpopulations, such as the elderly and individuals with pre-existing pulmonary and cardiovascular diseases, are at increased risk to the effects of exposure to airborne particulate matter. It is likely that additional host factors are important in determining individual susceptibility to particulate-induced morbidity and mortality. It is well established that genetic background influences susceptibility to many inhaled environmental agents, including ozone (29), nitrogen dioxide (30), and silica (31). We demonstrated previously in inbred mice that the interstrain variance in the phagocytic response to ACP challenge was greater than the intrastrain variance, and that susceptibility was likely inherited as a recessive trait (18). The present study was designed to further define the mode of inheritance of responsivity to particle challenge, and to identify quantitative trait loci and potential candidate susceptibility genes.

As described previously (18), acute exposure to ACP significantly and reversibly attenuated AM phagocytosis in B6 mice but had no effect on AM function in C3 mice. The AM dysfunction is likely a direct effect of exposure to ACP because there was no detectable inflammation in either strain at any time after particle exposure. The differential effect on AM function and immune defense among inbred strains of mice may have important implications for understanding the epidemiologic relationship between particle exposure and lower respiratory illness (e.g., 3, 4, 7). That is, it was demonstrated previously that ACP challenge exacerbates influenza-induced mortality in particle-responsive mice (19). Further, Zelikoff and colleagues (17) have shown that repeated exposure of rabbits to H₂SO₄ aerosol reduces uptake and intracellular killing of pathogenic bacteria by AMs. Therefore, it may be hypothesized that individuals with genetic predisposition to particle-induced AM dysfunction are more susceptible to viral and bacterial infection of airways that may lead to hospital respiratory admission.

The segregation analysis models imply that there are two major genes responsible for the differential responsiveness to AM phagocytic dysfunction after ACP exposure. The segregation data were best explained by S.A.G.E. with the two unlinked loci hypothesis. This homoscedastic model is also in agreement with the estimate of the minimal number of susceptibility loci (1.27), as determined by the formula described by Wright (21). Among the heteroscedastic models (in which assumptions of equal variances and normal distributions were not met), the one locus general and one locus, C3 dominant models best described the segregation data. There was little statistical discrimination between the two models based on log likelihood calculations and AIC. Only the one locus, C3 dominant model was not significantly different from the general model. The segregation analyses therefore indicated that one or two loci most likely account for a statistically significant portion of the genetic variance in this model, and that the dominant trait is carried by the C3 strain.

To identify the chromosomal location of the ACP susceptibility loci, a genome-wide scan was performed with an intercross cohort derived from B6 and C3 mice. The greatest linkage for ACP susceptibility was to a segment on chromosome 17. The susceptibility QTL is located in the interval between D17Mit147 (18.1 cM) and D17Mit64 (20.6 cM) and spans approximately 3 cM. The QTL was examined for candidate genes by comparative mapping between murine and human genomes, and a number of candidate genes were identified (Figure 5). The candidates include heat shock proteins Hsp70-1 and Hsp70-3 that have been suggested to have an important protective role in cell and tissue stress response to oxidants and other environmental challenges (32). Tumor necrosis factor- (Tnf ) is also located in the QTL, and critical roles in host defense and apoptosis as well as susceptibility to oxidant exposure have been described for this locus (29, 36). Additionally, a number of histocompatibility loci (e.g., H2-Pa, H2-Pb, H2-Oa, H2-Ma, H2-Ab1, H2-Aa, H2-Eb2) are found within the interval bounded by the QTL, and a potential role may be hypothesized in ACP-induced immune dysfunction. Another pair of candidate genes located slightly proximal to the peak of the QTL are mast cell proteases 6 (Mcpt6) and 7 (Mcpt7). Mast cells have been demonstrated to modulate pulmonary epithelial injury and immune responses to nonspecific stimuli such as ozone (39), and tryptases have been proposed to have an important role in these processes (40, 41). It is conceivable that particle exposures stimulate mast cells to release tryptases that alter AM microenvironment. These changes may, in turn, affect macrophage phagocytic function.

View larger version (46K): [in this window] [in a new window]   Figure 5.   Schematic representations of candidate genes for the chromosome 17 QTL for susceptibility to ACP-induced AM phagocytic dysfunction. Approximate locations of each locus in the mouse and human genome are shown. Note that TPS1 and TPS2 have only chromosomal assignment.

A suggestive QTL was located on chromosome 11. Candidate genes within the QTL include inducible nitric oxide synthase (Nos2), as well as a number of small inducible CC-chemokines (monocyte chemotactic proteins 1 and 3 [Scya2, Scya7]; macrophage inflammatory proteins 1 and  [Scya3, Scya4]; regulated on activation, normal T cells expressed and secreted [Scya5]; and eotaxin [Scya11]). A functional role for each has been described in the pathogenesis of asthma, fibrosis, and other lung diseases (42- 45). All of the murine candidate genes have human homologues on chromosomes 6 (chromosome 17 QTL) and 17 (chromosome 11 QTL) (Figure 5). Synteny of the candidate genes within the QTLs is conserved between the mouse and human genomes.

It is especially important to note that the major and minor QTLs for ACP susceptibility overlap directly with QTLs described for responsiveness to ozone (O₃) (29). In segregant B6:BX and B6C3F₂ cohorts derived from B6 and C3 mice, significant and suggestive linkages to chromosomes 17 and 11, respectively, were identified for susceptibility to lung inflammation induced by 48-h exposure to 0.3 ppm O₃. Further, the chromosome 11 QTL is also similar to one of the QTLs described for susceptibility to acute lung injury (Ali1) (46).

These observations may have important implications in understanding the determinants of susceptibility to air pollutants and the epidemiologic associations of pollutants with respiratory morbidity and mortality. Particulates usually occur as mixtures with other pollutants, including sulfates and oxidants such as O₃. Indeed, a number of epidemiologic studies that monitor mixtures have found respiratory symptoms (coughing, wheezing, lung function decrements) and other respiratory health outcomes (acute asthma exacerbation, lower respiratory tract infection) associated with multiple pollutants (47). It is not clear whether the health effects are due to additive or synergistic properties of the pollutants. Results of our studies suggest that there may be specific genes within the major and minor QTLs that confer susceptibility to distinct response phenotypes induced by two prevalent air pollutants, sulfate-associated particles and O₃. Fine-mapping studies are currently underway to narrow the chromosomal length of the susceptibility QTLs to less than 1 cM. Subsequent physical mapping approaches will ultimately lead to the identification of the susceptibility genes and strain-specific polymorphisms.

This is the first demonstration of genetic loci that are important determinants of responsiveness to particle-induced morbidity. Identification of these susceptibility genes, and understanding their regulation, will lead to a better mechanistic understanding of the causative effects of the pollutants and pollutant mixtures on respiratory health. Gene characterization will also provide a means for identifying susceptible individuals and developing strategies for intervention to prevent lung injury induced by exposure to environmental pollutants.