Metabolic reprograming and increased inflammation by cadmium exposure following early-life respiratory syncytial virus infection: the involvement of protein S-palmitoylation

Abstract

Early-life respiratory syncytial virus (RSV) infection (eRSV) is one of the leading causes of serious pulmonary disease in children. eRSV is associated with higher risk of developing asthma and compromised lung function later in life. Cadmium (Cd) is a toxic metal, widely present in the environment and in food. We recently showed that eRSV reprograms metabolism and potentiates Cd toxicity in the lung, and our transcriptome-metabolome-wide study showed strong associations between S-palmitoyl transferase expression and Cd-stimulated lung inflammation and fibrosis signaling. Limited information is available on the mechanism by which eRSV reprograms metabolism and potentiates Cd toxicity in the lung. In the current study, we used a mouse model to examine the role of protein S-palmitoylation (Pr-S-Pal) in low dose Cd-elevated lung metabolic disruption and inflammation following eRSV. Mice exposed to eRSV were later treated with Cd (3.3 mg CdCl₂/l) in drinking water for 6 weeks (RSV + Cd). The role of Pr-S-Pal was studied using a palmitoyl transferase inhibitor, 2-bromopalmitate (BP, 10 µM). Inflammatory marker analysis showed that cytokines, chemokines, and inflammatory cells were highest in the RSV + Cd group, and BP decreased inflammatory markers. Lung metabolomics analysis showed that pathways including phenylalanine, tyrosine and tryptophan, phosphatidylinositol and sphingolipid were altered across treatments. The BP antagonized metabolic disruption of sphingolipid and glycosaminoglycan metabolism by RSV + Cd, consistent with BP effect on inflammatory markers. This study shows that Cd exposure following eRSV has a significant impact on subsequent inflammatory response and lung metabolism, which is mediated by Pr-S-Pal, and warrants future research for a therapeutic target.

Respiratory syncytial virus (RSV) causes severe lung diseases, bronchiolitis, and pneumonia, in infants, young children, and the elderly. Moreover, RSV infection in infants increases risk of subsequent development of asthma later in life (Falsey et al., 2005; John et al., 2003; Piedimonte, 2013; Westerly and Peebles, 2010). The RSV is a significant global health care burden, causing 3–4 million yearly hospitalizations worldwide, and this burden became apparent in the recent outbreak of RSV (Mosscrop et al., 2022). In addition to severe disease following RSV infection, increasing evidence shows that the immune system is altered following the early-life infection, which may impact lung function later in life (Malinczak et al., 2020). Moreover, previous studies show amplified severity of RSV bronchiolitis by environmental cadmium (Cd) exposures such as passive smoking (Bradley et al., 2005; Semple et al., 2011).

Cd is a naturally occurring toxic metal. While a large body of research shows that high dose Cd exposure, for example, occupational exposure and cigarette smoking, causes acute and chronic lung toxicities (ATSDR, 2012), less is known about the impacts of low-level Cd exposure at levels found in nonsmokers, most of which come from dietary intake. Accumulation of Cd in the general population is usually assessed by urinary concentration, which reflects a global change of Cd content in the body rather than the local change in specific tissues, such as in the lung. Cd accumulates in human organs progressively with age (Ruiz et al., 2010) because humans do not have effective mechanisms for Cd elimination (Satarug and Moore, 2004; Suwazono et al., 2009). Importantly, such cumulative toxic impacts of dietary Cd on lung health have been recently shown by animal and epidemiological studies (Chandler et al., 2016b, 2019; Hu et al., 2019a,b; Jarrell et al., 2022; Park et al., 2020). Specifically, our previous study in a mouse model showed that the levels of inflammatory marker cytokines, chemokines and inflammatory cells were substantially increased with Cd exposure following early-life RSV infection (eRSV), implying that cumulative low dose Cd exposure following eRSV has a significant impact on subsequent inflammation and lung metabolism (Jarrell et al., 2022). This previous study provided fundamental findings for the current study.

Despite the critical impacts of low-level Cd on lung health, the mechanistic relationship between Cd exposure following early-life lung injury and lung dysfunction in later life has not been fully elucidated. Interestingly, our recent findings in a transcriptome-metabolome-wide association study (TMWAS) showed that Cd-caused metabolic disruption was strongly associated with zdhhc11, a member of the dhhc palmitoyltransferase family (Hu et al., 2018a). This provided a profound resource for developing the hypothesis for the current study. Consequentially, the major goal of this study is to determine the role of protein S-palmitoylation (Pr-S-Pal) in lung inflammation by Cd exposure following eRSV. This was achieved using a mouse model and a chemical inhibitor of protein palmitoyl transferases, 2-bromopalmitate (BP).

To test the hypothesis that Pr-S-Pal regulates Cd-amplified lung inflammation and metabolic disruption after eRSV, lung tissues and plasma were analyzed with our established workflow applying statistics and bioinformatics to high-resolution metabolomics (HRM) and inflammation markers (Chandler et al., 2016a,b, 2019; Go et al., 2015a; Hu et al., 2018a, 2019a,b; Jarrell et al., 2022). To examine the mechanistic relationship between RSV infection at a young age and its risk for impaired lung function later in life under environmental Cd exposure, we used mice that were exposed to RSV as infants, mimicking early ages in humans. Mice were recovered, and then evaluated for the effects of subsequent low level Cd exposure and Pr-S-Pal inhibition on lung inflammation and injury as mature adults. This study elucidates a critical role for Pr-S-Pal in Cd-potentiated pathogenesis of RSV infection.

Materials and methods

Animals, RSV infection, Cd exposure, and BP treatment

Experimental protocols for animal studies were approved by Georgia State University (A21004) and Emory University (201700691) Institutional Animal Care and Use Committees, and experiments were performed in accordance with the guidelines and regulations. C57BL/6J mice purchased from Jackson Laboratory (Bar Harbor, Maine) were housed and bred in clean facilities and fed standard mouse diet (Laboratory Rodent Diet 5001, LabDiet, St. Louis, Missouri) for breeding colonies. Male mice were selected in this study because a recent study showed exacerbation of allergic reaction in male mice relative to female mice post RSV infection (Malinczak et al., 2019). In addition, this allowed us to compare the results with our previous findings and to align with the baseline conditions with our previous studies (Chandler et al., 2019; Hu et al., 2018a, 2019a,b; Jarrell et al., 2022). Experimental treatments for 6 groups are illustrated schematically (Supplementary Figure 1) and details are as follows: Infant mice (14 days old) were intranasally exposed to saline (vehicle control [VC]) or RSV [RSV line 19 (19F)] at a low dose (4 × 10⁴ PFU per g of body weight) as previously described (Cormier et al., 2010; Han et al., 2011; Matsuse et al., 2000). The RSV 19F strain was previously described (Moore et al., 2009; Stokes et al., 2011) and kindly provided by Dr Martin Moore (Emory University). Starting 21 days after RSV exposure and at 5 weeks old, mice were treated with either 3.3 mg CdCl₂/l (RSV ± Cd group) or untreated (RSV only group) drinking water for 6 weeks. The infection scheme for RSV was refined and the exposure period for Cd was shortened from those of our previous investigation. This refined exposure paradigm provided sufficient combined effects from eRSV and Cd for the purposes of this investigation, ensured recovery from RSV prior to Cd exposure, and minimized effects due to Cd in older age. To test for Pr-S-Pal, a subset of the RSV + Cd group was treated with BP (Sigma-Aldrich, Cat. no. 238422) in drinking water along with Cd treatment (RSV ± Cd ± BP group, 3.3 mg CdCl₂/l and 10 µM BP). Subsets of the control group were treated with BP (BP only group, 10 µM BP) or Cd in drinking water (Cd only group, 3.3 mg CdCl₂/l) starting 5 weeks old for 6 weeks. An additional VC group received no treatments. All Cd and BP treatments in drinking water were refreshed weekly throughout the 6 weeks exposure. Experimentation sample size (n = 8 per group) was determined for sufficient statistical power and sufficient total amounts of tissues required for all assays performed in this study. Mice were given sterile-filtered drinking water without or with 3.3 mg/l CdCl₂ (Sigma-Aldrich, St. Louis, Missouri) throughout the experimental period. Cd contents in standard mouse diet and drinking water were negligible (65.5–68.8 ng/g food and 50 ng/l water) compared with 3.3 mg/l CdCl₂ in drinking water.

Quantification of Cd in lung by inductively coupled plasma mass spectrometry

Lung tissue ¹¹⁴Cd was quantified and normalized to tissue mass as previously described (Chandler et al., 2016b). Briefly, 50 mg of the lung tissue was subjected to wet acid digestion using nitric acid and hydrogen peroxide. Indium was added to samples as an internal standard prior to digestion. Samples were digested in a MARS 6 microwave digestion system (CEM, Matthews, North Carolina). After digestion, each sample was diluted to 10 ml and analyzed in triplicate in a random order using inductively coupled plasma mass spectrometry (ICP-MS) (Thermo Scientific iCAP Q ICP-MS, Bremen, Germany), in collision cell mode using kinetic energy discrimination with helium. A linear standard curve was established on a range of 0.25–64 ppb in the same run as samples. Samples were analyzed alongside 2 reference urine samples (NIST SRM 2668 levels 1 and 2).

Histopathology

Intact lungs were harvested at the time of completing Cd exposure for 6 weeks. For histology, left lung tissues were fixed with 10% neutral buffered formalin and embedded in paraffin blocks as described previously (Hwang et al., 2014, 2016). At least 8 sections of lung tissues from each group (VC, RSV, Cd, RSV + Cd, RSV + Cd + BP, BP) were stained with hematoxylin and eosin (H&E) to evaluate lung inflammation. Individual lung tissues were scored by blind examination for histopathology analysis to ensure an unbiased assessment of the samples. Inflammation and focal aggregates of infiltrating epithelial alveolar cells in the airways, blood vessel, and interstitial pneumonitis were measured using a severity score system defined as 0 (no lesion; normal), 1 (mild inflammation; hypertrophy of bronchiolar cells, <20% of lung affected), 2 (moderate inflammation; normal thickness of a single cuboidal cell, 20%–40% of lung affected), 3 (marked inflammation; slight expansion and distension of the airway, 40%–60% lung affected), and 4 (severe inflammation, thickening that occludes the airway lumen resulting in a very narrow lumen, >60% lung affected with tissue necrosis or damage) (Derscheid et al., 2013; Klopfleisch, 2013). Images were acquired by using an Axiovert 100 (Zeiss, Oberkochen, Germany) at 50× magnification.

Cytokine and chemokine measurements

To measure cytokines and chemokines in lung, we used lung tissue lysates (Chandler et al., 2016a; Hu et al., 2019b). Briefly, the lung tissue was homogenized using the plunger of a syringe and a sterile cell strainer (mesh size 70 µm) in 1.5 ml of Roswell Park Memorial Institute (RPMI) 1640 medium (Corning, San Diego, Arizona) containing penicillin and streptomycin. Later, lung lysates were harvested by centrifugation (2000 rpm, 20 min). Cytokines including tumor necrosis factor (TNF)-α, interleukin (IL)-4 and IL-13, and chemokines including RANTES (regulated upon activation, normal T cells expressed and secreted) and KC (kerotinocyte cytokine) were measured in lung lysates by cytokine (eBioscience, San Diego, California) or chemokine ELISA using Ready-Set-Go kits following the manufacturer’s procedures (R&D Systems, Minneapolis, Minnesota) (Lee et al., 2015).

High-resolution metabolomics

Lung tissues (20–30 mg) were used to extract metabolites in acetonitrile:water (2:1) containing internal standards (Go et al., 2015a) following the procedures as described previously (Chandler et al., 2016a,b; Hu et al., 2018b). Each sample was analyzed with a Q Exactive HF Orbitrap mass spectrometer (85–1275 m/z) (Thermo Fisher Scientific, Waltham, Massachusetts); each analysis was performed with 3 technical replicates. Chromatographic separation was achieved with a Waters XBridge BEH Amide XP HILIC (50 × 2.1 mm, 2.6 µm particle size) column combined with positive electrospray ionization (ESI; HILIC(+)) and with an endcapped C18 column (Higgins Targa C18 50 × 2.1 mm², 3 μm particle size) with negative ESI (C18(−)). Mass spectral data were extracted with apLCMS v6.6.8 (Yu et al., 2009) and xMSanalyzer v2.0.6.1 (Uppal et al., 2013), recovering metabolic features with high resolution mass to charge (m/z) paired with retention time (RT). Data were prefiltered to retain only features with nonzero values in >70% in all samples and >80% in each group, and data from triplicate analyses were median-summarized prior to statistical and bioinformatic analyses.

Metabolomics data analysis

Metabolomics data were quantile normalized and log₂ transformed prior to analysis. One-way ANOVA using limma v3.46.0 was performed to select features that differed between the 6 experimental groups. An additional analysis comparing RSV + Cd and RSV + Cd + BP was performed using T-testing in limma. Hierarchical clustering analysis and Manhattan plots as a function of m/z were used for untargeted metabolome comparison of the significant features differentiating treatment groups (raw p < .05 by limma). Selected features (raw p < .05) were further studied by pathway enrichment analyses using mummichog v2 (Li et al., 2013) and annotated with xMSannotator v1.3.2 (Uppal et al., 2017) with the use of HMDB [Human Metabolome Database (http://www.hmdb.ca/)] which provides level 3 determination of tentative identity according to Metabolomics Standards Initiative (Wang et al., 2018). This approach protects against type 2 statistical error by including all features at p < .05 and protects against type 1 statistical error by permutation testing in pathway enrichment analysis (Uppal et al., 2016).

Western blot analysis

Tissues were lysed in buffer containing CelLytic M Cell Lysis Reagent (Sigma-Aldrich) and 2× Laemmli Sample Buffer (Bio-Rad) at a 1:1 ratio, 10 mM sodium glycerophosphate, 10 mM sodium pyrophosphate, 1 mM NaF, 1 mM Na₃VO₄, and protease inhibitor cocktails (Cai et al., 2022). After sonication, samples were resolved on SDS-PAGE and transferred to nitrocellulose membranes (Bio-Rad). Membranes were probed with an antibody specific for LC3A (Cell signaling Technology) followed by a fluorophore-conjugated secondary antibody. The signals were detected and quantified using the Odyssey Infrared Imaging System (LI-COR).

Statistics

Measures were analyzed using Prism v5 software (GraphPad Software, Inc., La Jolla, California) and R v3.6. Results are presented as mean ± standard error (SE). Unless otherwise noted, comparisons of all 6 experimental conditions were performed using 1-way ANOVA with post hoc testing using Tukey’s honest significant difference (HSD) for pairwise comparison of all 6 groups. To maintain clarity, only significant post hoc comparisons between control and RSV + Cd groups, between Cd and RSV + Cd groups, and between RSV + Cd and RSV + Cd + BP groups are displayed in figures. For analysis of ordinal inflammation score results, all 6 conditions were analyzed using the Kruskal-Wallis test with post hoc testing using a Wilcoxon test with Benjamini-Hochberg correction. For inflammation scoring, significant post hoc comparison of control and Cd and comparison of control and BP groups are also reported in figures. p-Values <.05 were considered statistically significant. Significantly different subsets of treatments are indicated by asterisks in all figures.

Results

Pr-S-Pal plays a role in increased cd deposit in lung in mice exposed to cd following eRSV

As described above, the main goal of this study was to determine the role of Pr-S-Pal in Cd-enhanced lung inflammation and pathological phenotypes in mice exposed to RSV at infant age. Our recent 2 important findings are foundations for this study, (1) low-dose Cd potentiated metabolic reprograming following eRSV (Jarrell et al., 2022) and (2) zddhc11 palmitoyl transferase was the main hub associated with Cd-altered metabolome cluster (Hu et al., 2018a). While relying on the key findings of these prior studies, to achieve the main goal for the study, we used BP to inhibit Pr-S-Pal mechanism by blocking palmitoyltransferases and focused on comparison between RSV + Cd and RSV + Cd + BP groups.

We first examined BP effects on Cd lung accumulation and body weight gain (Supplementary Figure 2). The Cd concentrations in lung tissue measured by ICP-MS show that BP co-treatment with Cd significantly decreased the lung Cd levels that were elevated by RSV infection (VC, RSV, BP: ≤0.0063, Cd: 0.011 ± 0.001, RSV±Cd: 0.013 ± 0.001, RSV±Cd±BP: 0.009 ± 0.002, Mean±SE ng/mg tissue, n = 7–9, *p < .05, ***p < .001, Supplementary Figure 2A). To test whether inhibition of Pr-S-Pal by BP had any effect on body weight gain, mice were weighed daily for the 5 days after RSV infection and then every 7 days during Cd and BP treatment until the end of the study. No significant effects on weight gain were observed, with body weight gain of 32.3% ± 0.9% for controls (VC), 26.4% ± 2.3% for the RSV + Cd group, and 31.1% ± 1.7% for the RSV + Cd + BP group (Supplementary Figure 2B).

Inhibition of Pr-S-Pal by BP significantly lowered lung proinflammatory markers induced by low-level cd exposure following eRSV

Our previous study showed that Cd potentiated inflammation following RSV infection at infant age in mice (Jarrell et al., 2022). No information is available, however, on the underlying mechanism of Cd-amplified toxicity after eRSV (eRSV + Cd). Given that BP blocked the effect of eRSV + Cd on weight, we examined the effects of BP on inflammation stimulated by eRSV and subsequent low-dose Cd exposure for 6 weeks. Proinflammatory cytokines, TNFα, IL-4, and IL-13, and chemokines, RANTES and KC, were analyzed by ELISA (Figure 1). While RSV alone and Cd alone appeared to have subtle effects on expression of cytokines and chemokines, supporting that a minimal inflammation occurs in the lung with either dietary Cd treatment for 6 weeks or early-life RSV infection, exposure to low-dose Cd for 6 weeks following eRSV caused substantially higher production of all the proinflammatory markers than control (RSV + Cd vs. VC, *p < .05, n = 8–10, Figure 1), consistent with our previous study (Jarrell et al., 2022). The levels of all these cytokines and chemokines increased in the lungs of the RSV + Cd group were significantly less by BP treatment (RSV + Cd + BP, *p < .05, n = 8–10), indicating that aggravated Cd toxicity with proinflammatory response in later life following eRSV could involve a Pr-S-Pal regulatory mechanism. The results from additional histological examination of lung sections by hematoxylin and eosin (H&E) staining are shown (Figure 2A). In this, the inhibitory effect of BP measured by infiltrating inflammatory cells to the lung (lung airway and blood vessel, with a similar trend in interstitial space) is shown (Figure 2B), consistent with the results of inflammatory cytokine and chemokine measurements shown above. Moreover, we measured the stress hormone cortisol and its metabolites in plasma by targeted metabolomics as an indication of inflammation (Figure 2C). Cortisol was increased in the RSV + Cd group (*p < .05, RSV + Cd vs. VC), and hydroxycortisol exhibited a similar trend. No effects were observed with dehydrocorticosterone, and BP did not significantly decrease levels of cortisol and its metabolites.

High-resolution metabolomics-identified inflammatory and immunologic metabolic pathways are associated with regulation of Pr-S-Pal

Our recent study of pathway enrichment analysis on the lung metabolites showed that metabolic pathways of amino acids and vitamin D were altered by Cd exposure for 16 weeks followed by RSV infection at infant age (Jarrell et al., 2022). To examine the link between Pr-S-Pal regulatory metabolism and the metabolic response and potentiation of inflammation by RSV + Cd exposure, we applied HRM of lung tissues (Figure 3; Supplementary Figure 3) and plasma (Supplementary Table 1) in all 6 groups (see Figure 1, VC, RSV alone, Cd alone, RSV + Cd, RSV + Cd + BP, BP, n = 8). Mass spectral data preprocessing from HILIC(+) yielded 13 681 metabolic features in lungs from 6 conditions. ANOVA showed 1048 lung metabolic features were changed by RSV, Cd, RSV + Cd, RSV + Cd + BP, and BP treatment at p < .05 (without adjustment for multiple comparisons). Of these, 7 were changed at FDR <0.05 (Figure 3A). Pathway enrichment analysis on the 1048 lung metabolites using mummichog software (Li et al., 2013) is shown by a bubble plot and indicates that metabolic pathways of Phe, Tyr, and tryptophan (Trp) (p = .007, P1), Tyr metabolism (p = .02, P2), mannose type O-glycan synthesis (p = .025, P3), sphingolipid metabolism (p = .04, P4), and aminoacyl-tRNA synthesis (p = .048, P5) were altered by RSV, Cd, RSV + Cd, RSV + Cd + BP, and BP (Figure 3B). In a complementary analysis, mass spectral data preprocessing from C18(−) yielded total 5875 metabolic features in lungs from 6 conditions. ANOVA showed 512 and 34 of 5875 lung metabolites were changed by RSV, Cd, RSV + Cd, RSV + Cd + BP, and BP treatment at p < .05 (without adjustment for multiple comparisons) and at FDR <.05, respectively (Supplementary Figure 3A). Pathway enrichment analysis on the 512 lung metabolites is shown by a bubble plot. Altered metabolic pathway indicates that pathways of Phe, Tyr, and Trp (p = .005, P1), phosphatidylinositol signaling (p = .02, P2), glycosylphosphatidylinositol synthesis (p = .03, P3), aminosugar metabolism (p = .05, P4), and Phe metabolism (p = .05, P5) were altered by RSV, Cd, RSV + Cd, RSV + Cd + BP, and BP (Supplementary Figure 3B). In addition to lung HRM, we measured systemic alterations following the forementioned challenges (Figure 1) by examining plasma HRM (Supplementary Table 1). Notably, HRM results of HILIC(+) and C18(−) on lung and plasma largely overlapped and suggested the potential target pathways, Phe, Tyr, Trp, sphingolipid, and phosphatidylinositol were altered by RSV, Cd, BP, and coexposure of virus and chemicals.

BP protected metabolic alterations by eRSV and subsequent cd exposure

For further analysis of metabolic alterations, we measured abundance of representative metabolites of above significantly disturbed metabolic pathways comparing the 6 groups, and the results are presented by whisker plots (Figure 4). The metabolites of phenylalanine (Phe) and tyrosine (Tyr) pathway, including Phe, Tyr, and phenylpyruvate, were substantially higher in the RSV + Cd group than the Cd group (Cd vs. RSV + Cd, *p < .05, **p < .01, n = 8/group); however, only Tyr was increased in the RSV + Cd group relative to control. BP treatment did not significantly reduce these levels to resemble controls (Figs. 4A–C). On the other hand, the metabolites of sphingolipid and phosphatidylinositol signaling pathway, including phosphatidylethanolamine (PE) and inositol triphosphate (IP3), were substantially lower in RSV + Cd group than control group (VC vs. RSV + Cd, *p < .05, **p < .01, n = 8/group). In contrast, BP treatment prevented from lowering abundance of sphingolipid pathway metabolites (RSV + Cd + BP vs. RSV + Cd, *p < .05, **p < .01, n = 8/group, Figs. 4D and 4E). Fructose 6-phosphate from the aminosugar pathway was also substantially higher in RSV + Cd group than control group (VC vs. RSV + Cd, *p < .05, n = 8/group), and conversely, BP prevented RSV + Cd-induced elevation (RSV + Cd + BP vs. RSV + Cd, *p < .05, n = 8/group, Figure 4F). Together, the results of elevated Tyr and Phe metabolites, and decreased glycosylsphinglipid and phosphatidylinositol signaling metabolites in RSV + Cd group, and reversed response by BP in the latter, are consistent with the finding above showing protective effect of BP on RSV + Cd-enhanced proinflammatory markers.

Pr-S-Pal regulates sphingolipid and glycosaminoglycan metabolism

As shown above, the protective role of BP in RSV + Cd-caused metabolic alterations is evident. Therefore, we further assessed BP impacts directly on RSV + Cd by comparing between the RSV + Cd and RSV + Cd + BP groups using HRM obtained from HILIC(+) methodology. ANOVA showed that abundances of 1700 and 336 metabolic features were different between the 2 groups at p < .05 (without adjustment for multiple comparisons) and at FDR <0.2, respectively (Figs. 5A–C, n = 8/group). Of 1700 features, 879 were higher and 821 were lower in the BP treated group relative to the non-BP treated RSV + Cd group (Figure 5C). Pathway enrichment analysis on the 1700 metabolites indicated that sphingolipid (p = .007) and glycosaminglycan (p = .047) metabolic pathways are 2 major target pathways of BP (Figure 5D). Representative metabolites including dehydrosphinganine, PE, and n-acetylgalacosamine (GalNAc) differ between 2 groups (Figure 5E). This result is consistent with the above findings and reinforce the important role of Pr-S-Pal in sphingolipid and aminosugar metabolism altered by RSV + Cd exposure.

BP reversed suppression of autophagy induced by cd exposure following eRSV

In our recent studies, we found that Cd stimulated mTORC1 (mechanistic target of rapamycin complex 1) signaling and Cd amplified mTORC1 activity following RSV infection (Jarrell et al., 2022). Moreover, it is well known that mTORC1 regulates the homeostasis of lipid metabolism (Go et al., 2020; Laplante and Sabatini, 2012) and inhibits autophagy. Autophagy regulates inflammation by removing pathogens and damaged organelles that act as inflammatory triggers, for example, RSV, Cd, or RSV infection following Cd exposure (RSV + Cd). Because of improvements in sphingolipids and phosphatidylinositol signaling, and inflammatory pathway by BP observed, we examined the link between Pr-S-Pal and autophagy. As a representative marker of autophagy activation, we analyzed and compared the changes of microtubule-associated protein 1A/1B-light chain 3 (LC3). LC3 is distributed ubiquitously in mammalian tissues and cells. A cytosolic form of LC3-I is conjugated to PE to form LC3-PE conjugate (LC3-II), which is recruited to autophagosomal membranes. LC3-II/LC3-I ratio, a commonly used as a marker for autophagy activity, was measured in the lungs from the 6 groups (Supplementary Figure 4). The results showed that autophagy activity measured by LC3-II/LC3-I was significantly lower in the RSV + Cd group compared with controls, and BP improved its activity (Supplementary Figure 4), suggesting that Pr-S-Pal is involved in regulation of autophagic activity. These data are consistent with PE response to eRSV + Cd and eRSV + Cd + BP as shown in Figure 4D.

Discussion

Accumulating research on the exposome emphasizes the critical importance of cumulative lifelong exposures on subsequent health outcomes yet relatively little understanding exists for effects of early-life infections on subsequent toxicologic exposures. The present research was undertaken to gain knowledge of mechanisms underlying an adverse effect of early-life RSV on subsequent toxicity due to Cd. Cd is number 7 on Agency for Toxic Substances and Disease Registry (ATSDR) Substance Priority List, and toxic effects from occupational exposures and cigarette smoking are well established. Toxicities from lower levels, principally from food, are difficult to evaluate in humans, and we have developed mouse models to study toxicities of lung Cd at levels found in nonsmoker’s lung without occupational exposure. These studies show that even at levels found in nonsmoking adults, Cd caused changes in the mouse lung metabolome (Chandler et al., 2016a,b; Hu et al., 2018a, 2019a), transcriptome (Chandler et al., 2016b; Hu et al., 2017, 2018a), and redox proteome (Go et al., 2013, 2014; Hu et al., 2019a) with effects on airway reactivity (Chandler et al., 2016b), glycolysis and lipid metabolism (Go et al., 2015b), inflammation (Chandler et al., 2019), and profibrotic signaling (Hu et al., 2017). This research further showed that Cd potentiated adverse reaction to H1N1 influenza virus infection (Chandler et al., 2019) and RSV infection (Hu et al., 2019b).

Of considerable toxicologic concern, we recently found that early-life RSV infection, from which mice appeared to fully recover, had increased lung Cd induced inflammation and toxicity (Jarrell et al., 2022). The present research shows that inhibition of Pr-S-Pal improved Cd-stimulated inflammation and metabolic alterations by eRSV. Dynamic palmitoylation is an important posttranslational modification for regulating protein localization, trafficking, and signaling activities. The homeostasis of Pr-S-Pal is maintained by both palmitoyltransferases (which add fatty acid to the protein targets) and thiolase (which removes fatty acid from the proteins) (Anderson and Ragan, 2016). In our previous integrative omics study (transcriptome-metabolome-wide study) (Hu et al., 2018a) to examine metabolome-transcriptome interactions under the influence of Cd and their effects on lung, we observed, for the first time, that zdhhc11 was the main hub associated with Cd-altered metabolome cluster. Zinc finger DHHC domain-containing palmitoyltransferase (zDHHC11) is one of the major mammalian palmitoyl transferases; however, the function of palmitoyl transferases in environmental stressor, for example, influenza, RSV, SARS-CoV2, heavy metal Cd-induced signaling and subsequent lung disease remains elusive. Therefore, in this study we used a nonspecific chemical inhibitor, BP, to block activity of palmitoyltransferases irreversibly rather than targeting a specific palmitoyltransferase.

As shown in the results, overall, BP treatment led to a protective response to lung inflammation caused by Cd exposure following eRSV. Co-treatment with BP led to a return of proinflammatory markers to levels resembling controls. General trends in proinflammatory markers regarding eRSV and Cd resembled those observed in our previous study (Jarrell et al., 2022). Basal levels and magnitude of changes varied somewhat, especially in the cases of TNFα and IL-13, but this may be explained by differences in number of RSV exposures and age of mice at termination of the experiment. The results further suggest potential target proteins critical for improving cell signaling and metabolism that were dysregulated by RSV + Cd. For example, based on observation with the decreased Cd deposit in the lung of RSV + Cd + BP group, divalent metal transporters, ZIP8 (SLC39A8), and ZIP14 (SLC39A14) could be target proteins altered by RSV + Cd because these metal transporters have critical roles in lung cell physiology (Besecker et al., 2008; Liang et al., 2022) and also known to have predictive cysteine sites that might be undergoing palmitoylation and depalmitoylation (swisspalm.org/proteins).

Secondly, on the basis of our data with the BP-improved metabolic response to RSV + Cd-caused disruption in sphingolipid and phosphatidylinositol pathway measured by lung and plasma HRM, potential target proteins of Pr-S-Pal could be the ones associated with these lipid signaling pathway. Decreased IP3 in the RSV + Cd group and improvement of IP3 in the RSV + Cd + BP group (Figure 4E) provide important evidence to suggest IP3 receptor and phospholipases as target proteins. IP3 is an inositol phosphate signaling molecule made by hydrolysis of phosphatidylinositol biphosphate (PIP2) by phospholipase C (PLC). IP3 is a second messenger molecule used in signal transduction in cells and Pr-S-Pal of these proteins mediate essential signaling pathways for cell growth and proliferation.

Identification of the sphingolipid pathway as the major pathway, yielding decreased levels of PE by RSV + Cd treatment and regained abundance of PE by RSV + Cd + BP (Figure 4D) supports the anti-inflammatory role for BP. PE, the second most abundant glycerophospholipid in eukaryotic cells, is an endogenous, bioactive phospholipid linked to the phosphatidylcholine (PC). Available data indicate that PE has diverse cellular functions, including membrane topology and fusion, mitochondrial biogenesis and autophagy, and interferes with leukocyte reactions and thus decreases the inflammatory activation (Calzada et al., 2016; Eros et al., 2009; Go et al., 2015a). Tryptophan metabolism, and altered Phe, Tyr, and Phe/Tyr ratio emerged as a central hub for metabolic control of inflammation and immunological processes (Felger et al., 2013; He et al., 2018). Intracellular levels of these amino acids are shown to be maintained by autophagy (Kwak et al., 2011; Sorgdrager et al., 2019), a cellular recycling system for proteins and organelles under stress conditions. Autophagy is induced to compensate for cellular demands and to restore the amino acid pool, and it is closely associated with mTORC1 signaling (Kwak et al., 2011). mTORC1 activity is negatively associated with autophagy activation, and in a prior study, we found that RSV + Cd activated mTORC1 signaling (Jarrell et al., 2022). Thus, the present finding that BP improved autophagy activity which was decreased by RSV + Cd (Supplementary Figure 4) identifies potential target proteins of Pr-S-Pal as critical regulators for mTORC1 activity and autophagy. This suggests that in the event of RSV + Cd-stimulated mTORC1, inhibited autophagy activity, and escalated lung inflammation might be mitigated by blocking palmitoylation.

A proposed scheme (Figure 6) summarizes the role of BP in antagonizing pathophysiological lung response to RSV + Cd by blocking Pr-S-Pal effects. In this scheme, BP (1) decreased accumulation of Cd in the lung via inhibiting Pr-S-Pal in metal transporters (ZIP14, ZIP8), (2) regained abundance of IP3 via IP3 receptor and PLC, (3) improved amount of PE via PC N-methyltransferase, PS decarboxylase, and CDP (cytidine diphosphate) ethanolamine pathway, and (4) stimulated autophagy and inhibited mTORC1 signaling further decreased inflammatory markers. Taken together, this research provides new mechanistic insight into the protective role of BP in lung inflammation in association with metabolic dysregulation stimulated by low dose Cd exposure following eRSV, and provides a foundation for future studies to identify palmitoylation target proteins, biomarkers for adverse effects of eRSV, and strategies to reduce risks from eRSV.

Supplementary data

Supplementary data are available at Toxicological Sciences online.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Acknowledgments

The authors gratefully acknowledge the technical help of ViLinh Tran on sample analysis by mass spectrometer. The authors want to acknowledge the Oklahoma University Health Science Center (OUHSC) Vision Research Facilities for the Cellular Imaging Core services. The cores are supported by NIH/NEI grant P30EY027125 to Dr. Michelle C. Callegan and an unrestricted grant from Research to Prevent Blindness to the Dean McGee Eye Institute.

Funding

This work was supported by National Institute of Environmental Health Science Grants R01 ES031980 (Y.M.G.), R21 ES031824 (D.P.J. and Y.M.G.), P30 ES019776 (D.P.J.), R01 ES032189 (D.P.J. and Y.M.G.) and F32 ES033908 (Z.J.), National Eye Institute grant R01 EY026999 (Y.C.), and National Institute of Allergy and Infectious Disease Grants R01 AI54656 (S.K.) and R21 AI147042 (S.K.).

References

Author notes