Differential Pulmonary Toxicity and Autoantibody Formation in Genetically Distinct Mouse Strains Following Combined Exposure to Silica and Diesel Exhaust Particles

Background Inhalation of airborne particulate matter, such as silica and diesel exhaust particles, poses serious long-term respiratory health risks. Silica exposure can lead to silicosis and systemic autoimmune diseases, while DEP exposure is linked to asthma and cancer. Combined exposure to silica and DEP, common in mining, may have more severe effects. This study investigates the separate and combined effects of silica and DEP on lung injury, inflammation, and autoantibody formation in two genetically distinct mouse strains, thereby aiming at understanding the interplay between genetic susceptibility, particulate exposure, and disease outcomes. Silica and diesel exhaust particles were administered to mice via oropharyngeal aspiration. Assessments of lung injury and host response included in vivo lung micro-computed tomography, lung function tests, bronchoalveolar lavage fluid analysis including inflammatory cytokines and antinuclear antibodies, and histopathology with particle colocalization. Results Silica exposure elicited a well-established inflammatory response marked by inflammatory infiltrates, release of cytokines, and chemokines, alongside limited fibrosis, indicated by collagen deposition in the lungs of both C57BL/6J and NOD/ShilLtJ mice. Notably, these strains exhibited divergent responses in terms of respiratory function and lung volumes, as assessed through micro-computed tomography. Additionally, silica exposure induced airway hyperreactivity and elevated antinuclear antibody levels in bronchoalveolar lavage fluid, particularly prominent in NOD/ShiLtJ mice. Lung tissue analysis revealed DEP loaded macrophages and co-localization of silica and DEP particles. Conclusion Mouse strain variations exerted a substantial influence on the development of silica induced lung alterations. Furthermore, the additional impact of diesel exhaust particles on these silica-induced effects was minimal.


Background
Airborne particulate matter (PM) inhalation poses a signi cant threat to long-term respiratory health, with adverse effects such as interstitial lung diseases, increased susceptibility to infections, and chronic obstructive pulmonary disease.Certain occupations present a heightened risk to workers due to exposure to various airborne particulates.One well-known airborne particulate is crystalline silica, a naturally occurring mineral commonly found in rocks, sand, and soil, presenting a major occupational inhalation hazard for workers in various industries, including construction and mining (1).Epidemiological data from regulatory agencies including the Occupational Safety and Health Administration (OSHA) in the U.S.
(2) and EU-OSHA in Europe, have estimated that over 2 million workers in the U.S. and approximately 5.3 million workers in Europe are potentially exposed to hazardous levels of silica dust.
Silicosis, a chronic lung disease characterized by in ammation and nodular brosis, is a well-known health issue stemming from silica dust inhalation.Although silicosis is not a new disease, recent outbreaks occurred in young workers involved in jeans sandblasting and in workers handling arti cial granite or engineered stone (3,4), showing that silica dust exposure and silicosis remain relevant to this day.Beyond silicosis, inhalation of silica has also been linked to systemic autoimmune diseases (SAD) such as systemic lupus erythematosus (SLE), systemic sclerosis (SSc), and rheumatoid arthritis (RA) (5,6).This association underscores the connection between inhalation exposures and systemic effects, raising further concerns about the broader health implications for exposed workers.
In the context of silicosis and SAD associated with inhalation of silica particles, the question arises whether there is an association between both disease pathways.Cases of autoimmune diseases associated with prior silica exposure have been documented independently of a silicosis diagnosis (7).
Additionally, silicosis cases have been observed where speci c autoantibodies are signi cantly present (1).Doll et al. highlighted that silicosis patients exhibited an increased prevalence of particular autoantibodies (8).However, because the presence of these autoantibodies has not been correlated with pulmonary alterations in silicosis, the role of these autoantibodies in the pathophysiology of silicosis remains unclear.Additionally, research conducted by Mayeux et al. (9) in a murine model exposed to silica, demonstrated a close association between silicosis, markers of lung in ammation and brosis, lung biomarkers, and autoantibodies against extractable nuclear antigens.Given the shared in ammatory pathways in the initial stages of silicosis development and the presumed pathogenesis of silica-associated autoimmunity, an intricate interplay between these disease states is not unthinkable.Nevertheless, investigating this intricate relationship is a complex task, and the development of an animal model that more accurately resembles human silicosis holds promise for yielding novel insights.
Another airborne particulate common in mining and other dusty trades is diesel exhaust particles (DEP).DEP are present in diesel engine emission, which is a highly complex mixture of chemical substances in either gas or particle form.Exposure has been associated with enhanced allergic sensitization, development and aggravation of asthma, chronic bronchitis, decreased lung function, airway in ammation, decreased vascular function and development of cancers, as reported in epidemiological studies (10)(11)(12)(13)(14)(15), adding another layer of complexity to the risks faced by workers in mining.
While extensive research exists on the individual health effects of DEP and silica exposure, little is known about the impact of their combined exposure.Combined exposure to silica and DEP is common during mining operations, including hydraulic fracturing for oil or gas, as well as above-and underground mining operations (16)(17)(18)(19).Studies suggest that combined exposure to different types of particles or other environmental factors, such as viruses, may induce more pronounced effects compared to those caused by the individual compounds (20) (21).
Research has shown the substantial in uence of genetic susceptibility on the extent of silicosis or pulmonary in ammation elicited by speci c triggers (22)(23)(24).Furthermore, genetic predisposition assumes even greater signi cance in the context of systemic autoimmune diseases (25).As genetic susceptibility plays a pivotal role not only in the development of pulmonary in ammation but also in the broader spectrum of autoimmunity, our investigation incorporated two murine strains.To provide comprehensive insights, we selected the extensively characterized C57BL/6J strain, well-studied in both silicosis and autoimmunity, and the NOD/ShiLtJ strain, distinguished by its chronic in ammatory phenotype and heightened proclivity for autoimmune responses.Speci cally, we investigated how exposure to DEP, silica particles, and their combination impacts lung in ammation, lung function, airway hyperreactivity and local and systemic antinuclear antibodies in two mouse strains with differences in sensitivity.

Results
C57BL/6J and NOD/ShiLtJ mice display differences in lung volumes and baseline lung function in response to silica and DEP exposure.
In vivo micro-computed tomography (micro-CT, µCT) scans were performed to evaluate aerated (ALV) and non-aerated lung volumes (NALV) (ml), total lung volumes (ml) (TLV), mean total lung density (Houns eld units [HU]), mean aerated lung density (HU), and mean non-aerated lung density (HU).Scans were performed at two different time points, 8 and 12 weeks after the start of the experiment (see "Methods" Fig. 9 for experimental design).Visual inspection of transverse sections from the micro-CT images revealed a visibly higher number of dense areas in silica and silica + DEP exposed, but not DEP exposed mice compared to vehicle mice (Additional Fig. 1).When determining the aerated and non-aerated lung volumes (based on delineated area of interest and a cut-off in density), non-aerated lung volumes (NALV), which directly quanti es in ammatory and brotic disease burden (26), were higher in both silica exposed C57BL/6J (Fig. 1a) and NOD/ShiLtJ (Fig. 1b) mice compared to vehicle and DEP exposed mice, both in week 8 and week 12. Responses were in a similar extent, as fold changes over vehicle were not signi cantly different between strains (Additional File 1).These results were re ected in the mean density of the scans, as it was observed that silica exposed mice (both strains) demonstrated signi cantly higher mean lung densities than DEP and vehicle mice (Additional Fig. 3), primarily due to higher mean aerated lung densities in both strains (Fig. 1c&d).Moreover, DEP exposed NOD/ShiLtJ mice exhibited a higher mean aerated lung density compared to vehicle mice, but only at 12 weeks post-exposure (Fig. 1d).The density of aerated and non-aerated lung volumes re ects the composition of the alveoli and the surrounding tissues, including the epithelial layer, capillaries, extracellular matrix, and small airways, respectively.Higher density is typically associated with lung edema and the accumulation of in ammatory cells.
Remarkably, also total lung volumes (TLV) were signi cantly higher in silica exposed mice compared to vehicle exposed mice, in both mouse strains (Fig. 1a&b), and silica exposed C57BL/6J mice also exhibited higher aerated lung volumes (ALV) compared to vehicle and DEP exposed mice (Fig. 1a).These results can be attributed to a compensatory mechanism known to happen in mice during brosis or in ammation, but not in humans.However, in NOD/ShiLtJ mice (Fig. 1b), no differences in ALV could be observed between experimental groups.The differences in response between the strains for ALV is con rmed by the fold change comparisons, which were signi cantly different for silica and silica + DEP (Additional File 1).Additionally, TLV and NALV, but not ALV, were signi cantly higher in vehicle exposed NOD/ShiLtJ mice compared to C57BL/6J mice (Additional Fig. 2).These ndings suggest that NOD/ShiLtJ mice have a higher baseline in ammatory state compared to C57BL/6J mice.DEP exposed mice did not show any signi cant differences in their aerated, non-aerated, or total lung volumes compared to vehicle exposed mice, and none of the effects induced by silica were signi cantly enhanced by DEP co-exposure.
These results were further re ected in the baseline lung function tests.In silica and silica + DEP exposed C57BL/6J mice, inspiratory capacity (IC) (Additional Fig. 4) and forced expiratory volume in the rst 0.1 second (FEV 0.1 ) (Fig. 1e) were signi cantly higher compared to vehicle mice (Fig. 1c).In contrast, these lung function biomarkers did not show signi cant increases upon silica and/or DEP exposure in NOD/ShiLtJ mice (Fig. 1f).On the contrary, FVC was observed to be lower in silica exposed NOD/ShiLtJ mice compared to vehicle mice.Additionally, tissue damping (G) (Fig. 1f) and tissue elastance (H) (Additional Fig. 4), but not tissue hysteresivity (G/H) (Additional Fig. 4), were signi cantly lower in silica and silica + DEP exposed C57BL/5J mice, but not in NOD/ShiLtJ mice.DEP exposed mice did not show signi cant differences from vehicle mice for the included parameters measured by FlexiVent, and DEP coexposure did not signi cantly enhance the effects induced by silica exposure.No signi cant differences were observed between groups for Newtonian airway resistance (Rn) (Fig. 1e,f) and peak expiratory ow (PEF) (Additional Fig. 4).Differences in responses between the strains were also statistically con rmed by fold change comparisons, as outlined in Additional File 1.
Silica and DEP exposure elicit differential airway hyperreactivity responses in C57BL/6J and NOD/ShiLtJ mice.FEV 0.1 and airway resistance (Rn), both represented as % of baseline, were measured at baseline and after methacholine challenge (0, 1.25, 2.5, 5, 10, 20 and 40 mg/ml) to assess airway hyperreactivity (AHR) (Fig. 2).None of the experimental groups of C57BL/6J mice reached cut-off values for hyperreactivity as assessed by %FEV 0.1 (Fig. 2c) and %Rn (Fig. 2a).DEP, silica and silica + DEP exposed NOD/ShiLtJ mice, on the other hand, showed a signi cantly enhanced decrease in %FEV 0.1 upon methacholine challenge (Fig. 2d), reaching a 20% decrease with 20-40 mg/ml methacholine (Fig. 2f).%Rn of baseline did not show signi cant differences between experimental groups for NOD/ShiLtJ mice (Fig. 2b), but all the experimental group means reached the cut-off value of 200%.In conclusion, NOD/ShiLtJ mice seem to display hyperreactivity, which is more pronounced with exposure to DEP and/or silica, while this response is lacking in C57BL/6J mice.NOD/ShiLtJ mice show higher extent of lung in ammatory response upon silica exposure, based on lung histology and bronchoalveolar lavage uid cell counts.
The micro-CT and lung function analyses were complemented by qualitative histological examination of lung tissue, and analysis of in ammatory markers in bronchoalveolar lavage (BAL) uid (Additional File 2).H&E-stained lung sections showed mononuclear in ammatory in ltrates around bronchioles and vasculature as well as interstitially (Fig. 3a) after silica and silica + DEP exposure in both strains.In silicaonly exposed C57BL/6J mice, in ltrates were more apparent than in the silica + DEP exposed C57BL/6J mice.In NOD/ShiLtJ mice, similar presentations of in ammatory in ltrates were observed in silica and silica + DEP exposed mice.In addition, also vehicle and DEP exposed NOD/ShiLtJ mice presented with several in ammatory in ltrates, but in a lesser extent compared to the silica and silica + DEP group.Furthermore, silica and silica + DEP exposed mice also demonstrated presence of bi-and multinucleated cells, indicative of presence of giant cells, a feature of a chronic in ammatory state in the lungs.
Notably, NOD/ShiLtJ mice showed more abundant in ammatory in ltrates upon silica exposure compared to C57BL/6J mice.The in ammatory in ltrates in the vehicle groups of NOD/ShiLtJ mice, together with results from the exposed groups, con rms a predisposition to an in ammatory phenotype.This observation aligns with the higher non-aerated lung volume observed in vehicle exposed NOD/ShiLtJ mice compared to C57BL/6J mice (Additional Fig. 2).
Micro-CT scans and lung function measures alone do not allow for the differentiation between brosis and lung in ammation.Therefore, we conducted an additional speci c evaluation of lung brosis using H&E and Sirius red-stained lung tissue.Sirius Red stained lung tissue showed areas with collagen deposition in silica and silica + DEP exposed mice in both strains (Additional Fig. 5).In addition, a standardized grading scale was used to quantify the degree of pulmonary brosis in the H&E-stained lung sections.Individual and average values of pulmonary brosis scores for each experimental group are shown in Fig. 4b.Silica and silica + DEP exposed C57BL/6J mice were scored signi cantly higher than vehicle and DEP exposed mice.Silica, but not silica + DEP exposed NOD/ShiLtJ mice were scored signi cantly higher than vehicle and DEP exposed mice.However, although mild brosis was observed as shown by the brosis scores and the collagen deposition, none of the sections showed overt brosis, as none of the scores were higher than four on a scale of eight.
Silica and silica + DEP exposed mice (both strains) showed a relative increase of neutrophils, resulting in a relative decrease in macrophages (Fig. 4).Fold change comparisons showed how this increase in neutrophils was signi cantly more apparent in NOD/ShiLtJ mice compared to C57BL/6J mice (Additional File 1).When looking at the absolute cell numbers (total counts) (Additional Fig. 6), it was evident that the number of macrophages increased upon silica and silica + DEP exposure (C57BL/6J) or remained consistent (NOD/ShilLtJ) across all experimental groups.Eosinophil numbers did not increase upon silica and/or DEP exposure (Additional Fig. 6, absolute counts), while lymphocyte numbers showed a mild increase in numbers, signi cant in NOD/ShiLtJ mice, but not in C57BL/6J mice.
BAL uid in ammatory cytokines show similar responses upon silica and DEP exposure in C57BL/6J and NOD/ShiLtJ mice.
Hierarchical co-clustering of in ammatory cytokine levels showed how vehicle mice clustered with DEP exposed mice, and how silica mice clustered with silica + DEP exposed mice (Fig. 5).This was further supported by comparing the groups for each cytokine using Two-way ANOVA, showing how almost all the included cytokines and chemokines were upregulated in silica and silica + DEP exposed mice (Additional Fig. 7).The most robust responses in both strains were observed for the macrophage and neutrophil-attracting chemokines MCP-1, MIP-1a, MIP-2, KC/GRO and IP-10 in silica and silica + DEP exposed mice.In addition, pro-in ammatory cytokines related to a Th1 response were upregulated in silica exposed mice in both strains, including IFN-γ and IL-6 (more upregulated in C57BL/6J mice) and IL-15 (more upregulated in NOD/ShiLtJ mice).Additionally, both strains showed an upregulation of IL-1β, which is indicative of the activation of the NLRP3 in ammasome.Moreover, both strains showed an upregulation of IL-33 and IL-9, related to a Th2 response.IL-9 is associated with airway remodeling in the context of asthma, rather than lung in ammation or brosis.Finally, both strains showed an upregulation of IL-17A/F, related to a Th17 response, important in autoimmunity.
Interestingly, within NOD/ShiLtJ mice but not C57BL/6J mice, an additional upregulation of TNF-α, IL-10, IL-12p70 and IL-27p28/IL-30 was detected.Notably, these cytokines all display a regulatory function.IL-10 being an anti-in ammatory cytokine, whereas IL-12p70 forms a link between the innate and adaptive immune system.Nonetheless, it is important to emphasize that the responses were not uniform across all individual mice within a strain.BAL uid antinuclear antibody levels increase upon silica exposure, with stronger responses in NOD/ShiLtJ mice compared to C57BL/6J mice.
To evaluate the development of a local and systemic autoimmune response, antinuclear antibodies (ANA) presence was investigated in BAL uid and serum.ANA scores in BAL uid were signi cantly higher in silica and silica + DEP exposed NOD/ShiLtJ and silica exposed C57BL/6J mice, compared to vehicle and DEP exposed (Fig. 6) mice, with a high variation in responses between individual mice.Antinuclear antibody scores in serum were not signi cantly different between experimental groups.Additionally, vehicle NOD/ShiLtJ mice had higher ANA scores both in serum and BAL uid compared to vehicle C57BL/6J mice.Again, it is evident that the responses are not uniform across all individual mice within a strain.BALF ANA upon silica exposure correlates with extent of lung in ammation in NOD/ShiLTJ mice.
To investigate whether we could nd correlations between the endpoints within the silica and silica + DEP groups, which were taken together as not signi cantly different for none of the included endpoints, correlation matrices for C57BL/6J and NOD/ShiLtJ mice were established (Additional Fig. 8).As expected, lung function parameters and micro-CT biomarkers display a strong correlation in both strains.Interestingly, ANA scores in BALF of NOD/ShiLtJ mice correlated negatively with FVC (Pearson R = -0.5968,p = 0.0188) and IC (Pearson R = -0.6662,p = 0.0067), indicating that a stronger lung in ammatory response is correlated with a higher extent of ANA in BAL uid.In C57BL/6J mice, ANA values did not show correlation with lung in ammation biomarkers.In NOD/ShiLtJ, the majority of cytokines and chemokines correlated with each other.However, correlations were less obvious in C57BL/6J mice.DEP loaded macrophages and particle co-localization.
To evaluate particle localization within the lung tissue, lung histological slides were examined using light microscopy for DEP and Raman microscopy for both DEP and silica.The localization of DEP in macrophages was more distinct in Sirius Red-stained lung tissue sections compared to H&E-stained sections.DEP loaded macrophages were observed 10 weeks post last dose in both DEP exposed and DEP + silica exposed C57BL/6J and NOD/ShiLtJ mice.The occurrence of DEP loaded macrophages was notably more pronounced in NOD/ShiLtJ mice exposed to silica + DEP when compared to C57BL/6J mice exposed to DEP and silica + DEP (Fig. 7b).Additionally, in silica + DEP exposed mice, DEP was present in the lung tissue itself compared to almost exclusively in macrophages in DEP-only exposed mice, as observed by qualitative examination.In addition to analyzing the localization of DEP particles through histological examination, we utilized Raman spectroscopy to visualize the co-localization of silica and DEP particles and their uptake by macrophages.In unstained depara nized tissue sections, we observed the co-localization of silica and DEP particles within macrophages of silica + DEP exposed C57BL/6J mice (Fig. 7a).These ndings highlight the enhanced visualization of DEP localization in macrophages through Sirius Red staining and provide insight into the co-localization of silica and DEP particles within macrophages using Raman spectroscopy on unstained tissue sections.

Discussion
In this study, we investigated the relationship between lung in ammation, airway hyperreactivity, and antinuclear antibody (ANA) presence in the lungs and systemically.We exposed two immunophenotypically distinct mouse strains to silica alone and in combination with DEP.Our aim was to discern variations in pulmonary in ammatory responses at both cellular and cytokine levels, while also assessing clinical implications using lung function measurements and micro-CT.This approach enabled us to establish potential correlations between mouse strain immunophenotypes and their local and systemic autoantibody responses, shedding light on the complex interplay of these factors in pulmonary health.The study design is primarily exploratory, investigating relatively underexplored domains such as combined exposure effects in two immunophenotypically distinct mouse strains.The study's strength lies in its comprehensive examination of a wide array of outcomes.This extensive analysis offers valuable new insights into these uncharted territories.
Inhalation of particulate matter, such as silica and DEP, triggers intricate respiratory responses (28).Silica particles, deposited in the alveoli and alveolar ducts, activate alveolar macrophages, initiating an in ammatory cascade characterized by pro-in ammatory cytokine release, ultimately leading to chronic in ammation and brotic changes, as seen in silicosis (29).Crystalline silica's well-documented toxicity results from its crystalline structure and the introduction of surface charge or silanol-containing groups during processing (30,31).Reactive oxygen/nitrogen species (ROS/RNS) further sustain lung in ammation.DEP, on the other hand, due to their ultra ne nature, penetrate deep into the lungs, where alveolar macrophages engulf them and release ROS and in ammatory mediators, causing oxidative stress, airway damage, and exacerbating pre-existing respiratory conditions (32).Unlike silica, DEP are rather linked to chronic obstructive pulmonary disorder (COPD), emphysema, and cancer rather than brosis, highlighting the divergent outcomes of DEP exposure compared to silica (33).Given the wellestablished impact of these particulates on lung function, our study incorporated comprehensive lung function measurements.Furthermore, the diagnostic approach for silicosis in humans often employs micro-CT scans to visualize the disease's progression.Correspondingly, we adopted micro-CT scans for our murine model, aligning our diagnostic methodology with the clinical standards used in human cases, thereby facilitating a comprehensive evaluation of silicosis development in our experimental context.
Our data showed how NOD/ShiLtJ and C57BL/6J mice respond differently in terms of lung function measurements upon silica and/or DEP exposure, and that these ndings were in line with the ndings from micro-CT scanning.However, the compensatory mechanism that is more obvious in the C57BL/6J mice, consistent with ndings from a study by Dekoster et al. (34) in male C57BL/6J mice, makes it di cult to evaluate which strain develops the worst lung injury and lung function decline in response to silica.The fact that the NOD/ShiLtJ mice did not exhibit an increase in aerated lung volumes following silica exposure, but rather a decrease, is a notable departure from the C57BL/6J model, and more closely resembling what is observed in exposed human subjects.To the best of our knowledge, no studies have been published yet on lung function and lung in ammation after oropharyngeal silica exposure in the NOD/ShiLtJ mouse.Our data suggest that the NOD/ShiLtJ mouse might be a more realistic model of silicosis or silica-induced in ammation compared to the C57BL/6J mouse, especially when aiming to include lung function assessment, as the C57BL/6J mice appear to be resilient for the loss in lung function that comes with the development of silicosis.
As DEP are known to induce airway hyperreactivity (AHR), we included an AHR test using methacholine (35,36).The exacerbation of AHR in NOD/ShiLtJ mice by silica exposure suggests the potential of silica exposure on promoting hyperreactivity in the respiratory system, which is con rmed in only few other studies with mice (37).Silica nanoparticles, however, have been examined more extensively and have been shown to induce AHR (38,39).Reports of hyperreactivity in silicosis patients or exposed human subjects, seem to be lacking.One pilot study including 12 silicosis patients demonstrated a normal prevalence of AHR of around 11% (40).The lack of hyperreactivity reported with silicosis suggests that the AHR upon silica exposure is an effect that only occurs in mouse strains with genetic susceptibility for AHR, such as is observed here to be case with the NOD/ShiLtJ mouse.NOD/ShiLtJ mice and C57BL/6J mice are known to exhibit signi cantly distinct baseline immunophenotypic characteristics.C57BL/6J mice were included as a well-studied strain for silicosis and lung in ammation, not spontaneously developing autoimmunity nor developing autoimmune disease upon silica exposure (41).NOD/ShiLtJ mice, on the other hand, have a chronic in ammatory state, represented by high serum immunoglobulin levels compared to C57BL/6J mice (42).Reported immune parameters in literature appeared to be worse in female mice, which supports the choice for female mice in the current study (42).Our study further con rmed this in ammatory state, as mono-and binucleated in ltrates were found in lung tissue sections of vehicle exposed NOD/ShiLtJ mice, which were absent in C57BL/6J mice.
Silica inhalation induces local lung damage and the release of damage-associated molecular patterns (DAMPs), which activate the innate immune system through the toll like receptors (43,44).This activation leads to the release of several in ammatory mediators, subsequently recruiting macrophages, neutrophils and lymphocytes to the site of injury.Macrophages play a central role in engul ng silica particles, while neutrophils and lymphocytes contribute to the immune response and tissue repair (45).This response was also observed in our study, in both mouse strains, represented by an increase in mainly neutrophils and macrophages in C57BL/6J mice, and predominantly neutrophils and some lymphocytes in NOD/ShiLtJ mice.It is also remarkable that DEP loaded macrophages (%) were more present in silica + DEP exposed NOD/ShiLtJ mice compared to DEP and silica + DEP exposed C57BL/6J mice, while total cell counts show how C57BL/6J mice had more macrophages in their BAL uid upon silica exposure compared to NOD/ShiLtJ mice.As there are less macrophages present in the NOD/ShiLtJ mice, the relative DEP load per macrophage will be higher, which could explain the observed results.Furthermore, a wide array of in ammatory cytokines and chemokines was upregulated in BAL uid of silica exposed mice, with responses being similar between C57BL/6J and NOD/ShiLtJ mice.Although some cytokines were upregulated more in one strain than the other, heatmapping revealed no clear consensus or clusters of cytokines that differed in response between the strains.
Overall, our study reveals strong discrepancies between C57BL/6J mice and NOD/ShiLtJ mice in terms of lung function and micro-CT biomarkers.The ndings suggest that C57BL/6J mice exhibit greater resilience to silica exposure compared to NOD/ShiLtJ mice, likely due to their compensatory increase in aerated lung volume.However, intriguingly, both strains exhibit strikingly similar immune responses at the cellular and cytokine levels when exposed to a high dose of silica.These ndings indicate that, despite their contrasting baseline immune pro les, both mouse strains mount a robust and consistent immune reaction to silica exposure.This suggests that additional factors, possibly related to lung morphology or other aspects, may account for the observed differences between NOD/ShiLtJ and C57BL/6J mice (46, 47).NOD/ShiLtJ mice are commonly used as a type 1 diabetes model, as, depending on the conditions, approximately 50-80% of female NOD/ShiLtJ mice spontaneously develop type 1 diabetes.Moreover, these mice also display a propensity for polyautoimmunity, including a low incidence of autoimmune thyroiditis and Sjögren's syndrome (48).This was also con rmed in our study, as vehicle exposed NOD/ShiLtJ mice displayed low ANA positivity, while C57BL/6J mice did not exhibit ANA positivity under vehicle exposure.Furthermore, when exposed to heat killed Mycobacterium bovis, NOD/ShiLtJ mice exhibit phenotypic features reminiscent of lupus-like autoimmunity (49).The NOD/ShiLtJ mouse's autoimmune phenotype lies in the MHC region, speci cally in the context of the H-2g7 haplotype.In contrast, C57BL/6J mice exhibit an MHC haplotype, H-2b, which is less or not associated with autoimmunity.The H-2g7 haplotype carried by NOD/ShiLtJ mice is notable for its association with a defect in central tolerance mechanisms, leading to improper negative selection and is essential for the development of type 1 diabetes in these mice (50,51).Next to the autoimmune-associated MHC haplotype, the NOD/ShiLtJ mice also bear some other genetic variants impacting immune tolerance, and they exhibit multiple aberrant immunophenotypes including defective antigen presenting cell immunoregulatory functions, defects in the regulation of the T lymphocyte repertoire, defective NK cell function, defective cytokine production from macrophages (52) and impaired wound healing.Therefore, these two strains represent a non-in ammatory, non-autoimmune prone versus an in ammatory phenotype (including polyautoimmunity) concept through which we could take into account the possible in uence of genetic background on lung in ammatory and autoimmune features observed.
While the precise mechanisms linking lung in ammation, silicosis, and autoimmunity remain unclear, recent research has shed light on some key pathways.Chronic exposure to particulates like silica in the lungs can lead to cellular toxicity, tissue damage, in ammation, brosis, and the recruitment of autoreactive T and B cells, ultimately culminating in autoimmunity.Notably, silica-induced lung in ammation has been associated with the formation of ectopic lymphoid structures (ELS) within lung tissue, which may contribute to local autoantibody production.However, it's important to note that this phenomenon appears to be in uenced by speci c genetic backgrounds.While it is well-documented that silica exposure can induce ANA in lupus-prone strains (41) like NZBWF1/J mice (53), MRL mice (54), BXSB mice (54), and a subset of diversity outbred mice ( 9), our study represents the rst documented case of an exacerbation of the ANA response in NOD/ShiLtJ mice following silica exposure.Furthermore, it is of interest whether worse lung in ammation and lung function are correlated with a higher extent of ANA formation in the lung.Here, it was established that BAL uid ANA were signi cantly inversely correlated with FVC and IC in the NOD/ShiLtJ mouse, but not the C57BL/6J mouse, indicating that the processes that determine the intensity of lung function decline are also involved in the processes of local ANA production.
While the combined exposure to silica and DEP did not elicit effects distinct from those induced by silica alone, low dose DEP exposure independently did elicit subtle yet noteworthy outcomes.Speci cally, our low dose of DEP induced airway hyperreactivity in NOD/ShiLtJ mice, as evidenced by a decline in FEV 0.1 during methacholine challenge, even at 10 weeks after the last dose.Furthermore, DEP-induced lung in ammation in our study did provoke a discernible in ammatory response in NOD/ShiLtJ mice, detectable by micro-CT.However, with the exception of a slightly higher average macrophage count in DEP exposed mice compared to vehicle exposed mice, most other outcomes, including cytokine levels, did not exhibit differences between the two groups in both strains.This may stem from variations in the sensitivity of the different endpoints employed, with micro-CT proving to be one of the more sensitive and robust measures.Of importance to note is that the used dose of DEP (4 x 10 µg) was chosen to re ect a realistic low exposure dose encountered in daily life, signi cantly lower than the doses employed in other studies of lung in ammation, with the lowest doses being 25 µg x 3 (total of 75 µg) (55).Consequently, our lower dose (total of 40 µg) might not have been su cient to induce additional pronounced effects observed in studies using higher DEP doses.Nevertheless, that makes our ndings even more relevant, considering the majority of endpoints were assessed approximately 10 weeks after the last dose.
Moreover, DEP particles were still visibly present and detectable using both visual examination of cytospins from BAL uid and Raman spectroscopy on lung tissue slides, also 10 weeks after the last dose.
In the endpoints where DEP did induce signi cant effects compared to vehicle exposed mice, DEP did not enhance the effects of silica-alone exposure.An aspect to consider is that both exposures involve particulate exposure, which may trigger similar pathways and thereby fail to induce synergistic or signi cantly exacerbated effects.Moreover, the effects induced by the established dose of silica (4 mg) are large and might dilute out the effects induced by DEP.Further investigations are warranted to elucidate the underlying mechanisms and fully comprehend the observed interactions between DEP and silica in the context of lung effects and brosis.

Conclusions
Our ndings strongly support the notion that genetic background, and therefore strain variations, exert a substantial in uence on the development of silica-induced lung injury.This underscores the potential value of formal genetic analyses, employing a wider range of strains or recombinant inbred strains derived from these mice.For instance, exploring the Collaborative Cross recombinant inbred strains could prove instrumental in identifying potential loci associated with susceptibility to silica-induced in ammation.

Preparation of particle suspensions
Crystalline silica was baked at 200°C for 1 hour to remove endotoxin contamination prior to use.After baking, silica particles were suspended in sterile 0.9% saline + 0.05% Tween at a concentration of 20 mg/ml for silica only exposure (1 mg per dose of 50 µl) and 40 mg/ml for combined (silica + DEPs) exposure (1 mg per dose of 25 µl).DEP were suspended in sterile 0.9% saline + 0.05% Tween at a concentration of 0.2 mg/ml for diesel only exposure (10 µg per dose of 50 µl) and 0.4 mg/ml for combined exposure (10 µg per dose of 25 µl).Fresh suspensions were made for every group and sonicated for 10 minutes in a bath sonicator to ensure uniform dispersion.The suspension was vortexed immediately before use to obtain a homogeneous suspension.

Animals
Eight-week-old female NOD/ShiLtJ (n = 36) and C57BL/6J (n = 36) mice were purchased from Charles River Laboratories (Belgium) and housed 4-5 mice/cage.Mice were housed in a conventional animal facility with 12h dark/light cycles in individually ventilated cages and were given free access to drinking water and food.Mice were given two weeks of acclimatization before the start of experiments.All experimental procedures were approved by the animal ethics committee of KU Leuven (P111/2021) in compliance with national and European regulations.Background information about the study design and mouse strains can be found in Additional File 3.

Experimental protocol
For each strain, four experimental groups were included (n = 9/group): Vehicle (V): control animals were exposed to vehicle only (0.9% saline + 0.05% Tween); Mice received four doses over the course of two weeks, with two administrations per week, using oropharyngeal aspiration under iso urane anesthesia, as shown in Fig. 8.

In vivo Lung Micro-Computed Tomography (µCT)
In brief, mice were anesthetized by inhalation of 1.5-2% iso urane in oxygen and scanned in supine position using an in vivo µCT scanner (Skyscan 1278, Bruker µCT, Kontich, Belgium) (34).Scanning parameters and details about the procedure are described in Additional File 4.

Lung function parameters and airway hyperreactivity
Lung function was assessed using the FlexiVent FX system (SCIREQ, EMKA Technologies, Montreal, Canada), and mice were subsequently euthanized.Measurements were performed as described by Devos et al. (57,58).Brie y, the system was designed to measure both forced oscillations (QP3 perturbation) and forced expiration parameters and was operated with FlexiWare™ 7.6 software.The system was equipped with a FX1 module, a negative pressure forced expiration (NPFE) extension for mice, and a small particle size Aeroneb® Lab nebulizer (2.5-4 µm; Aerogen, Galway, Ireland).Mice were anesthetized with pentobarbital (IP, 120 mg/kg body weight, Dolethal®) and once su ciently anesthetized, a tracheotomy was performed to insert an 18-gauge metal cannula.Mice were quasi-sinusoidally ventilated with a tidal volume of 10 mL/kg and a frequency of 150 breaths/min to mimic spontaneous breathing.At the start of the experiment, two successive deep in ations were applied to maximally in ate the lungs to a pressure of 30 cmH 2 O to open the lungs, and lungs were allowed to equilibrate at that pressure over a period of 3s.The gas compression-corrected volume was read as inspiratory capacity (IC, ml).Airway resistance (Newtonian) (Rn), tissue damping (G) and tissue elastance (H) were assessed using Quick Prime 3. Tissue hysteresivity (G/H) was calculated based on tissue damping and tissue elastance.Forced expiratory volume in 0.1 second (FEV 0.1 ), forced vital capacity (FVC) and peak expiratory ow (PEF) were assessed using the NPFE.Tiffeneau-index was calculated based on FEV 0.1 and FVC (FEV 0.1 /FVC).After performing all perturbations at a baseline level, airway hyperreactivity (AHR) to increasing methacholine concentrations (0, 1.25, 2.5, 5, 10, 20 and 40 mg/ml) (59) was assessed using the forced oscillation technique (QP3 perturbation and NPFE with the same system.

BAL uid differential cell counts
Lungs (right and left lobes) were lavaged with 0.7 ml sterile saline (0.9% NaCl) three times in situ.Collected bronchoalveolar lavage uid ( rst lavage and the pooled second and third lavage) was centrifuged at 1000 g for 10 min and respective supernatant was stored at − 80°C.The rst lavage supernatant was used for cytokine and anti-nuclear autoantibody (ANA) analyses.Cell pellet was resuspended in 1 ml saline, and 250 µl of the resuspended cells were spun at 300 g for 6 min (Cytospin, 3, Shandon, TechGen, Zellik, Belgium) onto microscope slides, air-dried, and stained (Diff-Quick ® Method, Thermo-Fisher Scienti c, Massachusetts, US).A total of 100 cells/animal were manually counted using a light microscope to obtain the ratio of macrophages, eosinophils, neutrophils, and lymphocytes.
To evaluate DEP uptake by alveolar macrophages, DEP loaded macrophages were counted using BAL uid cytospin slides.The percentage of loaded macrophages was determined by manually counting a total of 100 macrophages using a light microscope (Additional Fig. 9).

BAL uid anti-nuclear using Indirect Immuno uorescence (IIF)
Antinuclear antibody (ANA) presence was evaluated in the supernatant of 1:10 diluted BAL uid samples using NOVA Lite® HEp-2 ANA slides (Inova Diagnostics).The experimental procedure involved applying the diluted samples (1:10) onto HEp-2 cell-containing slides and incubating them for 1 hour at 21°C.Subsequently, slides were washed to eliminate unbound antibodies and immersed in phosphate-buffered saline (PBS) for 5 minutes.Detection of bound antibodies was accomplished by incubating slides with goat anti-mouse IgG Alexa Fluor 488 (Southern Biotech, 1030-30) diluted in 0.05% PBS-Tween (1:400) for 1 hour at 21°C.Following another wash in 0.05% PBS-Tween for 5 minutes, slides were covered with a coverslip.For the semi-quantitative assessment of uorescence intensity, two representative images per mouse were acquired.The evaluation was conducted by three independent scorers (LJ, FL, NH) in a blinded manner.A scoring system, as described by the manufacturer, was employed to evaluate the intensity of the uorescence, as follows: (4+) Brilliant apple green uorescence, exhibiting a brighter intensity compared to the positive control.
Each scorer independently assigned a score to the observed uorescence intensity, ensuring consistency and minimizing bias.The nal intensity score for each sample was determined by averaging the scores assigned by the three scorers.

Lung histopathology and particle colocalization
After lavage, the left lung lobe was lled with 4% formaldehyde and tied off, removed from the body and immersed in 4% formaldehyde for xing (at least 48h), whereafter formaldehyde was replaced with 70% ethanol.Para n embedded tissue sections (5 µm) were stained with Hematoxylin and Eosin (H&E) for general cellular and tissue morphology and Sirius Red for the presence of collagen bers.Adjacent sections were used for the two different staining.Sections were blindly examined by a professional pathologist (AV) using light microscopy.In addition, a standardized grading scale, the modi ed Ashcroft scale, was used to grade pulmonary brosis in H&E-stained sections by two separate scorers (FL, MG).Brie y, ve elds of H&E-stained lung tissue were inspected using a 20-fold objective.Each eld received a grade from 0 to 8, based on short descriptions of alveolar septa and lung structure and mostly based on reference images (27).Grades were added up and divided by the number of elds to obtain a brotic index (FI) for each mouse per group (n = 4 mice/group).Grades from scorers were averaged to obtain a nal grade ± SD for each experimental group.
To detect and qualitatively examine colocalization of silica and DEP particle deposition in alveolar macrophages, particles inside the tissue were visualized in an unstained depara nized section using Dxr3xi Raman imaging microscope (Thermo Fisher Scienti c; Scan setting-Laser Power 2 mW, exposure time 0.01 sec (100 Hz), number of scans = 8, image pixel size 0.2 µm).One section for each experimental group was scanned of the C57BL/6J mice.

Statistical and data visualization
Scatter plots and stacked were created in GraphPad/Prism (Graphpad Software version 9.3.1,La Jolla, CA, https://www.graphpad.com/)representing mean ± SD, unless mentioned otherwise.
Experimental groups were compared using One-way ANOVA within strains, or Two-way ANOVA with input regarding experimental execution and result interpretation.All authors provided feedback on the nal manuscript.

Figure 7 Representative
Figure 7