https://orcid.org/0000-0001-9090-027X
Figures
Abstract
By applying dual proteome profiling to Salmonella enterica serovar Typhimurium (S. Typhimurium) encounters with its epithelial host (here, S. Typhimurium infected human HeLa cells), a detailed interdependent and holistic proteomic perspective on host-pathogen interactions over the time course of infection was obtained. Data-independent acquisition (DIA)-based proteomics was found to outperform data-dependent acquisition (DDA) workflows, especially in identifying the downregulated bacterial proteome response during infection progression by permitting quantification of low abundant bacterial proteins at early times of infection when bacterial infection load is low. S. Typhimurium invasion and replication specific proteomic signatures in epithelial cells revealed interdependent host/pathogen specific responses besides pointing to putative novel infection markers and signalling responses, including regulated host proteins associated with Salmonella-modified membranes.
Author summary
As causative agents of infectious diseases, intracellular bacterial pathogens have evolved diverse immune escape and survival strategies while thriving inside their hosts. Whereas the residence of the pathogen inside the host can impact the functioning of the latter drastically, the bacterial macromolecular content is typically vastly outstripped by that of the host. This sizeable difference in macromolecular content has thus far hindered proteome profiling of bacteria and its infected host in an integrated and unbiased manner. Here, the power of dual proteome profiling–a universal mass spectrometry-based proteomic strategy for unravelling the intricate interplay between pathogen and host—is presented for the exploration of host as well as pathogen proteomic responses over the time course of the infection, thereby advancing our understanding of infectious diseases.
Citation: Fels U, Willems P, De Meyer M, Gevaert K, Van Damme P (2023) Shift in vacuolar to cytosolic regime of infecting Salmonella from a dual proteome perspective. PLoS Pathog 19(8): e1011183. https://doi.org/10.1371/journal.ppat.1011183
Editor: Andreas J. Baumler, University of California Davis School of Medicine, UNITED STATES
Received: February 6, 2023; Accepted: June 19, 2023; Published: August 3, 2023
Copyright: © 2023 Fels et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE [81] partner repository with the dataset identifier PXD018610 for DDA data and with the dataset identifier PXD019863 for DIA data. The generated spectral libraries, FASTA databases, and search results are organized in the Open Science Framework (OSF) project page [82] and are available at https://osf.io/u96bk/.
Funding: This work was supported by the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation program (PROPHECY grant agreement No 803972) to PVD, the Fonds Wetenschappelijk Onderzoek (FWO-Vlaanderen) (project number G051120N) to PVD, and by Ghent University Concerted Research Actions (grant BOF23/GOA/001) to PVD and (grant BOF14/GOA/013) to KG. PW and MDM authors received a salary from the PROPHECY grant. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Salmonella enterica serovar Typhimurium (S. Typhimurium) is an enteric facultative intracellular pathogen with wide-ranging disease outcomes grading from self-limiting gastroenteritis to systemic infection. As a foodborne pathogen, S. Typhimurium invades the intestinal tissue via different routes through infection of a diverse range of host cells, including epithelial cells (e.g., M cells and absorptive epithelial cells) and immune cells (e.g., dendritic cells and macrophages), in which they proliferate [1, 2]. Infection by Salmonella is mediated by its two type III secretion systems (T3SSs); T3SS-1 and -2, encoded within Salmonella pathogenicity island 1 (SPI-1) and -2 (SPI-2), respectively [3]. Along with the T3SS-1 components, the 40-kb SPI-1 gene cluster encodes chaperones, type III effector proteins (T3Es) and transcriptional regulators that modulate expression of many virulence genes within, but also outside SPI-1 [4]. Whereas T3SS-1 and its effectors are mainly associated with the invasion of non-phagocytic epithelial cells [5], T3SS-2 delivers effectors that promote Salmonella replication inside phagocytic as well as non-phagocytic cells [6–8]. To invade epithelial cells, Salmonella delivers through its T3SS-1 a first batch of effectors into the host cytoplasm to provoke actin cytoskeleton rearrangements that allow for bacterial uptake by macropinocytosis. More specifically, lamellipodia and filopodia-containing protrusions, called membrane ruffles, emerge and enclose the pathogen [9, 10]. This active invasion of epithelial cells mediated by T3SS-1 and its cognate effectors is thought to play a key role in gut inflammation during salmonellosis [1, 5]. A first set of pro-inflammatory responses (e.g., NF-κB signalling) are detected upon entry of the pathogen owed to both manipulation of cell death pathways and/or tight junctions via effectors, and by recognition of peptidoglycan, flagellin or lipopolysaccharide by host pattern recognition receptors (PRRs) [11]. Following internalization, Salmonella resides inside a phagosome-like compartment, dubbed the Salmonella-containing vacuole (SCV), where it further rewires host cell processes to create a replicative niche mainly through the action of T3SS-2 effectors [12]. In essence, T3SS-2 effectors maintain the SCV membrane, navigate juxtanuclear SCV positioning and elicit filamentous membrane extensions from the SCV along microtubules, termed Salmonella-induced filaments (SIFs) [13]. SIFs have been observed in various cell types and exhibit endosomal markers like LAMP1, Rab7 and vacuolar ATPase [13]. Interestingly, the onset of SIF biogenesis at 4–8 hours post-infection coincides with intercellular bacterial replication and has therefore been postulated to facilitate nutrient acquisition from the host, amongst others [14–17]. Remarkably, in about 20% of infected epithelial cells, bacterial replication also occurs outside the SCV in the cytosol where flagellated bacteria hyper-replicate and account for the majority of net replication [18]. Moreover, some SCVs are displaced from their juxtanuclear positioning in the host cell towards the cell periphery [19]. Both phenomena, i.e., vacuolar escape and centrifugal SCV movement, have been linked to dissemination of the pathogen to other cells to repeat the infectious cycle [19]. Moreover, the various (subcellular) environments encountered by these bacterial subpopulations during infection further lie at the basis of the observed heterogeneity of bacterial subpopulations in vitro [20] and in vivo [21]. Specific bacterial subpopulations were for example found to reprogram infected hosts to promote long-term bacterial survival [22].
As such, it is clear that over the course of an infection, an intricate and complex interplay between Salmonella and its host occurs that ultimately defines the infection outcome. As diverse successive environmental cues shape both the host and bacterial proteome, this highlights the importance of studying both proteomes in an integrative manner to unwire this multifaceted interaction and to obtain holistic insights. The characterization of molecular mechanisms underlying interactions between pathogen and host has extensively advanced over the last two decades, mainly through omics-informed infection studies in cellular as well as animal infection models.
Staples et al. [22] characterized non-growing, metabolically active subsets of Salmonella associated with persistence [23, 24] by dual RNA sequencing (RNA-seq). This simultaneous analysis of the transcriptomes of the host and the pathogen revealed that this specific bacterial subpopulation secrete effectors implicated in reprograming macrophages to promote anti-inflammatory macrophage polarization linked with bacterial long-term survival [25]. While the sensitivity of transcriptome analysis outperforms proteome studies, especially in the context of infection, the lower growth rates typically associated with infection [26], bacterial virulence [27, 28] and persistence, are associated with a considerably higher gene expression noise [29] and extensive post-transcriptional regulatory control. Therefore, mRNA levels in bacterial populations with reduced growth rates often correlate poorly with protein levels, stressing the need for protein-based studies. However, as the theoretical Salmonella proteome (~4,500 proteins) is vastly underrepresented in the context of the infected host (~20,000 proteins), especially when considering an up to ~1500 fold difference in the Salmonella versus human cellular total proteome content in the context of infection [30], studying both proteomes simultaneously and interactively entails proven technical challenges [30, 31] that are further complicated by the considerable heterogeneity in bacterial and host subpopulations as mentioned before.
Accordingly, proteome profiling of Salmonella and its infected host has thus far mainly focused on studying the reprogramming and signalling responses of the host or bacterial adaptive responses disjointedly [32–38], or alternatively, studied the interactions between host and pathogen to provide insights into the complex network of effector/host interactions during infection [39].
By significantly improving peptide detection and quantification, we recently reported on the hybrid use of data-dependent and data-independent acquisition spectral libraries to empower what we called dual proteome profiling [31]. In our current study, we used this optimized workflow to perform time-dependent dual proteome profiling of infection-relevant S. Typhimurium pathogen encounters with its host, without the need of a priori bacterial pathogen enrichment. More specifically, through the use of an intracellular epithelial infection model (i.e., S. Typhimurium infected HeLa cells), dual proteome profiling reports for the first time on the extensive protein-level dynamics during Salmonella infection thereby unveiling a detailed and integrative proteomic perspective on host-pathogen interactions over the time course of infection. This way, the regulation of host proteins associated with Salmonella-modified membranes and/or the formation of SCVs or SIFs was revealed besides the induction of epithelial cell differentiation in response to Salmonella infection. The later potentially representing a pathogenic strategy of intracellular bacteria to reprogram the host cell thereby enabling survival and pathogenic spread [25].
Results
Epithelial salmonella infection model
In line with previous reports in HeLa cells, while at lower multiplicities of infections (MOI or ratio of bacteria to cells of 10) only ~5% of HeLa cells were infected and since at modest MOI increases, higher bacterial loads rather than an increase in the overall number of infected cells has been reported due to Salmonella cooperative entry, an MOI of 100 resulted in the infection of over half of the cell population [18, 40–42] and was opted for dual proteome profiling (see below results). As bacterial load has proven to impact host signalling and host responses [2, 18, 20, 42], we assessed host cell cytotoxicity and host cell counts in function of bacterial load over the time course of the infection in real-time (S1 Fig). In control cultures (non-infected cells and cells infected with a non-invasive ΔprgH mutant strain [43]) as well as cultures infected at low MOI (MOI 10), no significant impact on cell viability and proliferation was apparent (i.e., inferred from the expected ~2-fold increase in cell counts over the time course of measurement (24 h) considering the 28.2 h doubling time of HeLa CCL2 cells [44]). At an MOI of 100, host cell counts and viability was not significantly affected up till 10 hpi when compared to the control setups (S1A and S1B Fig), while at 24 hpi a 16% increase in cytotoxicity was observed (i.e., when considering maximum cytotoxicity (total lysate reference) compared to the control setups)(S1C and S1D Fig). Using a constitutively eGFP-expressing SL1344 strain [45], bacterial load could be inferred from the total fluorescence (RFU) and green object counts. Here, from the two-step progression curve observed, a first steep rise in fluorescence–and total green object counts–was observed between 2 and 5 hpi, followed by a plateauing phase and a less steep increase at ~8 hpi which plateaued again at ~16–17 hpi (S1E Fig). This second phase coincides with the occurrence of hyper-replicating Salmonella in the cytosol [46]. Further, green object sizes also indicated that a maximum green object size was observed at ~8 hpi before a ~2-fold decrease in size (S1F Fig), both phases aligning with the accumulation of Salmonella in SCVs before vacuolar escape and more diffuse bacterial spread in the cytosol, respectively [46].
Dual proteome profiling of salmonella-infected epithelial cells
Proteome samples of S. Typhimurium infected human HeLa cells were prepared to monitor differences in steady-state protein expression levels at 2, 4, 8, 16 and 24 hpi in biological quadruplicates. Following trypsin digestion, each sample was analysed by LC-MS/MS in both DDA and DIA mode and data analysed respectively using MaxQuant [47] against a composite database containing Salmonella and human UniProtKB protein entries, and EncyclopeDIA [48] using our optimized hybrid spectral library workflow [31]. The latter combines experimental libraries with predicted spectral libraries to optimize the search space for identifications, shown to facilitate dual proteome profiling [31] and suited for DIA-based exploration of S. Typhimurium infected epithelial host cells.
When considering all samples analysed in both DDA and DIA acquisition modes, a total of 7,821 proteins were identified, comprising 6,830 human proteins (87% of all protein identifications) and 991 S. Typhimurium (13% of all protein identifications and 21% of the annotated S. Typhimurium proteome) proteins.
In a bit more detail, by DDA, 6,696 unique proteins (i.e., 767 S. Typhimurium and 5,929 human proteins) (S1 Table) and, by DIA, 6,734 unique DIA proteins (i.e., 853 S. Typhimurium and 5,881 human proteins) (S2 Table) were identified, implying that the number of human proteins identified remained by and large unaffected when comparing DDA versus DIA data, while a notable increase in S. Typhimurium proteins identified (~10%) was observed (Fig 1A). The latter is further accentuated when considering the 1.6-fold increase in the unique number of Salmonella peptides identified (i.e., 3,851 versus 6,306 unique peptides by DDA and DIA, respectively) (S1 and S2 Tables).
(A) Venn diagram created using Venny (http://bioinfogp.cnb.csic.es/tools/venny/index.html) showing the total number of human (left Venn diagram) and S. Typhimurium (right) proteins identified using DDA (orange) and DIA (grey) data (both at FDR ≤ 0.01). Variability of DDA (B) and DIA (C) replicate samples represented in PCA plots. Green, cyan, blue, orange, purple and pink circles represent control (Ctrl), 2, 4, 8, 16 and 24 hpi samples, respectively.
From the host perspective, global or individual principal component analysis (PCA) revealed a clear variability between control versus S. Typhimurium infected HeLa cells (component 1) as well as a temporal response (component 2) where proteomes from earlier time-points post infection (2, 4 and 8 hpi) can clearly be distinguished from later timepoints (16 and 24 hpi) (Figs 1B, 1C, S2A and S2B). This temporal response is also reflected in the Salmonella recordings (component 1, 53% for DDA and 34.1% for DIA data) (S2C and S2D Fig).
To monitor differences in protein levels between all DIA and DDA setups and replicates analysed, multiscatter plots of protein intensity values (log2) were generated. When grouping replicate samples per setup, the average intra-setup Pearson correlations appeared higher for DDA versus DIA (average and minimum correlations of 97% for DDA samples, and average and minimum correlations of 95% and 92%, respectively for DIA samples) with correlations ranging between 92% and 98% for all DDA datasets, and between 84% and 97% for DIA datasets when considering average correlations among all different setups (inter-setup) analysed (see correlation plots in S3 and S4 Figs).
Further, considering the protein expression ranges observed, a dynamic range spanning over 5 orders of magnitude (max. ~365,000-fold change) for DIA data, while only over 4 orders for DDA data was detected (max. ~57,000-fold change) (S1, S2, S3 and S4 Tables).
Improved sensitivity of DIA to study Salmonella protein downregulation during infection
After filtering for valid values (minimum 3 valid values in at least one group/setup), median normalized expression per species and imputation of missing values, human and S. Typhimurium protein expression values were compared in a pairwise manner with the HeLa control and the S. Typhimurium proteomes at 2 hpi, respectively. This selection left 4,666 human (S3 Table) and 479 S. Typhimurium (S4 Table) protein groups for comparison in case of the DDA data, and 5,818 human (S5 Table) and 844 S. Typhimurium (S6 Table) protein groups in case of the DIA data. In line with our dual proteome profiling results reported using artificial proteome mixtures [31], and while the S. Typhimurium protein identification rates in DIA only modestly increased (767 vs. 853 S. Typhimurium protein identifications), the number of reliably quantified S. Typhimurium proteins drastically increased with 76% (1.8-fold) when using DIA data (479 vs. 844 quantified proteins). Further, from the differential expression analysis, it is clear from the corresponding profile plots that both up- as well down-regulation (compared to 2 hpi) of S. Typhimurium proteins can comprehensively be studied with DIA data, while expression analysis is mostly confined to upregulated S. Typhimurium proteins with DDA data (S5 Fig). This discrepancy can be explained by the limited DDA over DIA sensitivity and the low S. Typhimurium protein content in the proteome samples of infected cells, especially at the earliest time points post infection.
Consequently, studying DDA-based S. Typhimurium protein downregulation during infection is rather limited to only a few [8] highly expressed S. Typhimurium proteins already identified and quantified at 2 hpi. Because of this finding, we further focussed our dual proteome expression analysis to DIA data. Nonetheless, generally similar trends and overlapping regulations could be observed among the DDA and DIA data (S1, S3 and S4 Tables). The intensities of significantly regulated proteins after multiple sample ANOVA testing (FDR ≤ 0.01 and an S0 of 0.1) are shown as heat maps and/or profile plots for human and S. Typhimurium regulated proteins, respectively (S5 and S6 Figs, and corresponding S5 and S6 Tables). Overall, 1,742 human and 436 S. Typhimurium protein groups showed a significantly regulated temporal response in DIA data during the time-course of infection (FDR ≤ 0.01), with proteins showing similar time-specific responses indicated in assigned clusters (S5 and S6 Figs).
Host cell responses to Salmonella infection reveal induction of host immunity and epithelial cell differentiation
To map the epithelial host cell response to S. Typhimurium infection over time, a gene ontology (GO) enrichment analysis (GOBP (Biological Process), GOCC (Cellular Component), GOMF (Molecular Function), general GO and UniProt keywords (FDR 0.02)) of the significantly regulated proteins (FDR 0.02) identified in the DIA-MS analysis was performed (Fig 2 and S7 Table).
Term enrichment was determined using the 1D annotation enrichment algorithm embedded in the Perseus software suite and p-values were corrected for multiple hypotheses testing using the Benjamini and Hochberg false discovery rate. Enriched annotation terms (FDR ≤ 0.02) in the categories (A) UniProt keywords, (B) GOBP (Biological Process), (C) general GO and (D) GOCC (Cellular Component) are shown as heatmaps and colour coded according to the 1D enrichment scores calculated [49], with red and green colouring indicating enriched and depleted annotations, respectively.
At early time points post-infection (i.e., 2 hpi), interleukin-1 mediated signalling and protein polyubiquitination are upregulated indicative of the induction of host immunity, while intriguingly, at late time points post infection, cell adhesion, extracellular matrix organizations and related terms, e.g., ‘collagen-containing extracellular matrix’ (e.g., including collagen α-3, thrombospondin-1 and Galectin-1) [50] and ‘cell adhesion mediated by integrin’ (e.g., with up to over 7-fold upregulation of the integrins α-2 (ITGA2), α-5 (ITGA5), α-V and β, besides ICAM-1, ADAM9 and vitronectin)) represent upregulated processes opposed to intracellular actin cytoskeleton downregulation. Some selected representative protein profile plots of the aforementioned significantly regulated categories are shown in Fig 3. While the marked induction of host immunity, host integrin signalling and extracellular matrix organization is in line with previous proteomic findings of infected epithelial cells [32], we are the first to theorize that jointly, these cellular and phenotypic hallmarks reflect epithelial cellular differentiation induced early upon infection [51].
Term enrichment was determined using the 1D annotation enrichment algorithm embedded in the Perseus software suite and p-values were corrected for multiple hypotheses testing using the Benjamini and Hochberg false discovery rate. Only corrected p-values ≤ 0.02 were considered. Selected GO terms and keyword annotations for with significantly altered protein abundancies were observed (i.e., extracellular matrix organization (GOBP, #26); upregulation at 24 hpi, cell adhesion (GOBP, #45); upregulation at 24 hpi, IL1-mediated signalling (GOBP, #27); upregulation at 2 hpi, integrin complex (keyword, #6); upregulation 24 hpi, actin cytoskeleton (GOCC, #26); downregulation 24 hpi, Ficolin-1-rich granule lumen (GOCC, #40); upregulation 2 hpi and downregulation 16 hpi, IFN-gamma signalling (GO, #8); upregulation 16 hpi, collagen binding (GO, #10); upregulation 24 hpi, proteasome complex (keyword, #22); down in control and upregulation 2 hpi) are shown for all setups analysed. Red and black asterisk above the profile plots means significant downregulation and upregulation of the corresponding proteins in the annotation group, respectively.
Overall, besides demonstrating a specific and complex response to bacterial infection over time, our findings seem to indicate that cell differentiation may generate physiological heterogeneity in S. Typhimurium infected epithelial cells as previously shown for the S. Typhimurium mediated reprogramming of macrophage hosts towards alternative anti-inflammatory (i.e., M2) polarized macrophages, a pathogenic strategy enabling (a subpopulation of) intracellular bacteria to survive and reprogram their host cell to enable bacterial survival and pathogenic spread [25].
Regulated expression of Salmonella-modified membranes host proteins implicated in formation of SCVs or SIFs
Since the creation of an extensive membrane network within its host lies at the basis of Salmonella’s infection strategy, we turned to a study reporting on a comparative proteome analysis of S. Typhimurium infected HeLa cells to obtain more detailed insight into expression changes of the proteome of internal membrane systems [52]. More specifically, Reuter et al. identified host proteins associated with Salmonella-modified membranes (SMM) shown to be enriched for transport proteins, GTPases, cytoskeleton and membrane linker proteins as well as proteins implicated in clathrin and coatomer-mediated transport [52]. From the high-confidence set of host proteins found associated with SMM (247 proteins), 213 proteins (86%) were also identified in our DIA analyses. Interestingly, of these, 84 (39%) were found to be significantly regulated after multiple sample ANOVA testing (FDR ≤ 0.01 and an S0 of 0.1) (see columns ‘SMM proteome (PMID: 25348832)’ and ‘ANOVA regulated SMM protein’ S5 Table). By itself, regulation of SMM protein expression over the time-course of infection was found to be highly statistically significant as determined by a chi-square test of independency (p < 0.01). Regulated SMM proteins included among others transport proteins (Phosphate carrier protein MPCP, Translational endoplasmic reticulum ATPase (VCP), ATP synthases F(0) complex subunits 1 and 0), proteins implicated in clathrin/vesicle-mediated transport (AP2A1, AP2B1 and TMED9), cytoskeleton and membrane linker proteins (Ezrin, Plastin-3, Myosin I, the Arp2/3 complex and Catenin alpha-1) besides several GTPases (including the over 10-fold upregulation of the Ras-related protein Ral-A). In line, annotation enrichment analysis showed significantly reduced protein abundancies of SMM proteins in the control setup, pointing to an overall general increase in SMM proteins upon Salmonella infection (p-value < 0.01). It is noteworthy that these regulated SMM targets hold several components of host pathways previously shown to be implicated in formation of SCVs or SIFs (e.g., endosomal and recycling pathways). These regulated targets include the regulated small Rab GTPases Rab7a [53], Rap-1b [54] and Rab10 [52]–all with known fundamental roles in cellular membrane dynamics besides representing hub targets of bacterial effector proteins [55]–next to the observed regulations of the AP-2 complex subunit β (AP2B1) [54], the F-actin binding protein utrophin [52], and the cytoskeleton(-linked) proteins desmoplakin [56] and Filamin [57]. Overall, our data demonstrates that a multitude of infection-relevant host proteins from the SMM are regulated in expression over the time course of infection, thereby representing interesting targets for further follow-up studies. The 29 ANOVA regulated SMM proteins (from the 84 regulated SMM proteins identified in total) signifying GTPases or proteins implicated in transport or cytoskeleton and membrane organizations (selected form keywords) are listed in S8 Table. In conclusion, in the context of infection, our work reports the regulation of host proteins associated with Salmonella-modified membranes including targets implicated in formation of SCVs or SIFs.
Temporal Salmonella proteome adaptations to adjust to the cytosolic lifestyle in infected epithelial cells
Reflecting the virulence lifestyle of infecting Salmonella, the pronounced activity of the two-component system phoPQ known to be exerted at later time-points post infection was evidenced by the upregulation of its two components PhoP and PhoQ at later time points and its downstream impact on SPI-1 and SPI-2 expression more generally [58] (Fig 4). More specifically, an enrichment of ΔphoPQ regulon targets [59] (i.e., a group of genes regulated as a unit, here genes regulated upon deletion of phoPQ) found upregulated (i.e, SipC, SopB, SipB, InvB, PrgH, InvG, PrgI) or downregulated (SodCa, PagC, PhoN, WrbA, UgpB, PdgL, AroQ, PotD, Rna, MsrA, YdcW, SseJ, Tal, DkgA, YdcR, PhnO, SseL) was observed at early (2 hpi) and late (16–24 hpi) times post-infection, respectively (FDR ≤ 0.05) (S6 Table and Fig 4A). These time-dependent PhoPQ dependencies are in line with the well-established positive regulation of PhoPQ on SPI-2 expression. Other temporal Salmonella proteome adaptations were indicative of other affected regulons, with an overall positive regulation of SPI-1 (e.g. hilA, hilC, fliZ) or SPI-2 (e.g. ssrA, slyA) at early and later time points of the infection, respectively (Fig 4).
(A) Using a recently reported compilation of highly curated regulon gene targets identified by means of microarray or NGS-based gene expression studies [59], enrichment of specific regulons was determined using the 1D annotation enrichment algorithm embedded in the Perseus software suite. P-values were corrected for multiple hypotheses testing using the Benjamini and Hochberg false discovery rate. Enriched annotation terms (FDR ≤ 0.05) in the categories are shown as heatmaps and colour coded according to the 1D enrichment scores calculated [49], with red and green colouring indicating enriched and depleted annotations, respectively. ESP (early stationary growth phase), MEP (mid exponential phase), upregulated (UP) and downregulated (DN) genes as reported in [59]. (B) Heatmap representation of significantly regulated S. Tyhimurium regulators (FDR ≤ 0.01, corresponding normalized averaged z-scores) from the DIA analysis and corresponding to the regulated regulons shown in A. SsrB was also identified, but expression was not significantly regulated.
Further, indicative of the metal ion shortage encountered by the bacterial pathogen when residing in the host [59, 60], significant upregulation of metal ion transporters and uptake systems for zinc (ZnuA, up to over 3-fold upregulation at 24 hpi), iron (IroBN (> 8-fold), EntBF (> 6-fold), SitAB (> 17-fold), Fur (> 13-fold), Mrp (> 2-fold), FepAB (> 4-fold) and CirA (>6-fold)) and manganese (SitAB) was observed (S6 Table). Also, expression of the ferric uptake regulator Fur—which binds to promoters of target genes implicated in iron acquisition and transport in Fe2+-rich conditions thereby preventing target transcription–was lowered (Fig 4B). In accordance, a significant impact on the fur regulon (including ugpB, sufA, aldB, ynhA and tal as well as the aforementioned sitAB, cirA, entF, fepA, iroB, entB targets) was detected, as expression of their corresponding target protein products were found enriched at 24 hpi (FDR ≤ 0.01) (Fig 4A and S6 Table). In line with the studies by Liu et al. [36] and Li et al. [33], bacterial proteome adaptations revealed downregulation of chemotaxis (e.g., CheAMW) and upregulation of His biosynthesis (HisABCDFGHI). In agreement with the increased proliferation rate of cytosolic Salmonella, ribosomal proteins (rplBEKMU and rpsCDEJLMRST) and proteins implicated in cell division (MinE, MpI, MukB, MurAEFG, Pal, TolB) were found upregulated at later timepoints of the infection (S6 Table and Fig 5). Since extensive transcriptional reprogramming was shown to accompany the successful colonization of the epithelial cytosol, a niche occupied by a hyper-replicating Salmonella sub-population especially at later times post-infection, it is noteworthy that upregulation of cytosol signature genes previously reported to be transcriptionally upregulated in Gram-negative pathogens colonizing the cytosol were also regulated in our proteome study (i.e., bioA, entF, fepA, fepB, iroB, iroN and sitAB). Furthermore, Powers et al. [59] demonstrated that a S. Typhimurium ΔsitAΔmntH mutant was compromised for growth in the cytosol of epithelial cells, linking this to the need for acquiring Mn2+ when residing in this subcellular location. Altogether, proteome profiling of infecting salmonellae is heavily biased towards its virulence repertoire and provides a proteomic perspective on host-pathogen interactions and (sub-)cellular pathogen localisation over the time course of infection.
Term enrichment was determined using the 1D annotation enrichment algorithm embedded in the Perseus software suite and p-values were corrected for multiple hypotheses testing using the Benjamini and Hochberg false discovery rate. Only corrected p-values ≤ 0.05 were considered. Selected GO terms and keyword annotations for with significantly altered protein abundancies were observed (i.e., SPI-1 (#4); downregulation at 16 and 24 hpi, histidine biosynthesis (keyword, #8); upregulation at 16 and 24 hpi) are shown for the 5 infection setups analysed. A black asterisk above the profile plots means significant upregulation (by 1D annotation enrichment of the corresponding proteins in the annotation group. Besides, some other categories (SPI-2 (#4), chemotaxis (GO, #4), structural constituent of the ribosome (GOMF, #14) and cell division (GOBP, #9) of which the corresponding protein members were all found to be significantly regulated based on t-testing are shown as profile plots.
Discussion
Given the generally acknowledged multifactorial complexity of host-pathogen interactions, the data variability in infection studies reported, and the high temporal aspect of the interactions, multiplex omics-based analyses deliver a comprehensive understanding of the complexity of such systems and the interplay between pathogen and host. To understand bacterial adaptation to host encounters, to date, the vast majority of Salmonella centred proteomics research relied on axenic grown Salmonella exposed to various (infection-relevant) growth conditions [61]. Only more recently, extensive adaptation of S. Typhimurium to infected host epithelial cells was shown by proteomics efforts making use of enriched Salmonella isolated from infected epithelial host cells [33, 36, 60]. The aforementioned strategies however requiring a rather lengthy and potentially biased enrichment procedure for the isolation of intracellular bacteria. Here, we report for the first time a dual proteome profile of Salmonella-infected epithelial host cells in function of time, both in DIA and DDA modes, overall providing a proteomic perspective on the bacterial pathogen encounter with its epithelial host and permitting the simultaneous study of host and pathogen proteomes.
Noteworthy, while DIA outperformed DDA both in terms of identifications and quantifications, especially for the bacterial pathogen, for DDA data, correlation of quantifications was shown to be higher than for DIA data. This finding is mainly attributed to the increased DIA-based quantification of low abundant peptides and due to the higher variability of quantification data for low abundance species given the interference with random noise [31].
This observation is also apparent from the generally less discrete (differential) patterns of Salmonella protein expression as compared to the distinctive human expression clusters. The former finding is also likely (in part) due to the biological variability of proteome sampling at different timepoints post-infection.
Bacterial proteome sampling in dual proteome profiling nevertheless remains less comprehensive as compared to previously reported protein identifications making use of pure Salmonella cultures or of Salmonella isolated from infected epithelial host cells. For example, when considering axenic S. Typhimurium cultures and triplicate shotgun proteome analysis, typically a 2-fold increase in the number of S. Typhimurium proteins identified and quantified [62] (around ~1,500 to 1,600) are found using similar MS instrumentation in DDA mode [30]. Noteworthy, while a significant higher number of Salmonella proteins were reported (~1,800) by Li et al making use of enriched Salmonella [33], it is important to note that samples were extensively fractionated by means of 1D SDS-PAGE separation and slicing of the gel into 36 gel pieces per biological replicate sample, while in our study unfractionated samples were analysed using a single LC-MS run per replicate sample. As such, offline sample fractionation in combination with DIA can provide a useful asset to improve DIA identification rates, especially in the case of complex proteome mixes with components of low abundance [31]. Alternatively, as demonstrated in the EncyclopeDIA workflow, making use of parallel DIA runs with narrow m/z windows in consecutive m/z ranges [48], a significant increase in the identification of Salmonella proteins can be expected in case of dual proteome profiling. Altogether, this indicates that sensitivity does not represent an inherent major shortcoming of dual proteome profiling.
Further, the high quantification sensitivity obtained by DIA permitted the reliable capture of over 100 regulated bacterial proteins over the time course of infection, generally exceeding the numbers of regulated bacterial genes reported in other studies, and, importantly, also including the identification of downregulated Salmonella proteins upon infection.
In line with the relative comparisons of Salmonella protein expression at various time-points post-infection previously reported [36] and by comparing protein expression regulation across infection conditions only, initial bacterial adaptations to the host might be missed. However, upregulation of e.g., phosphate utilization, previously shown to be highly induced in the Salmonella proteome upon medium switching occurring prior to bacterial internalization [60], was not observed in our study. This indicates that likely phosphate shortage is not experienced within the host itself but rather dependent on bacterial cultivation steps preceding the actual infection, necessitating the need for detailed experimental descriptions and uniformity when comparing omics datasets, hence the value of dual omics strategies.
On a related note, upon systematically comparing our dual proteomics data with previously reported transcriptome changes making use of dual RNA-sequencing and FACS-enriched S. Typhimurium infected HeLa cells [41], for the about 5,700 matching identifications (DIA data), transcriptome versus proteome changes were found to poorly correlate, especially at later timepoints post-infection (Pearson correlation below 0.2 at 24 hpi). Annotation enrichment analysis even pointed to oppositely regulated gene expressions when comparing protein versus transcript levels. In particular, while proteome data indicates downregulation of the GO term ‘actin cytoskeleton’ at 24 hpi, upregulation was observed at the transcript level. Opposingly at 16 and 24 hpi, upregulated ribosome biogenesis and mitochondrial gene expression was observed at the proteome level, while transcript levels were down (S7 Fig). Overall, this indicates that besides the lag phase between transcription and translation to be considered, translational and/or post-translational control significantly impacts bacterial gene expression especially in the context of (prolonged) infections, further emphasizing the need to study gene expression changes at the proteome level.
Further noteworthy in this context is that S. Typhimurium T3E host target proteins were found enriched in the category of regulated proteins as determined by a chi-square test of independency (p < 0.01; p-value 0.002203), indicating that bacterial effector functioning greatly shapes the proteome of the host. As a representative example, it has been shown that Salmonella remodels the host actin cytoskeleton to force its entry into non-phagocytic cells. These cytoskeletal rearrangements are mediated by T3SS-1 effectors such as SopB, SopE and SopE2, with indirect actions that are essential for efficient invasion [63]. Several other effectors were shown to directly interact with actin(-binding) proteins including effector SspH2 that has established interactions with profilin (PFN)-1 and -2 inside the host [57, 64, 65]. In our DIA data, PFN2 showed to be ~3-fold upregulated at 2 hpi. With its predicted inhibitory effect on the formation of membrane ruffles, PFN2 upregulation at 2 hpi supports the predicted downregulation of “formation of cellular protrusions” when performing Ingenuity Pathway Analysis (IPA) (S8A Fig).
To restore the host cell cytoskeleton, Salmonella effector SptP uses its GTPase-activating protein (GAP) activity to inactivate RAC1 [66], thereby reversing the Rho GTPase activating effects of SopE [67]. This restoration of the host cytoskeleton is also evident from the predicted activation of the “formation of actin filaments” at 2 hpi and the regulation of contributing genes (S8A and S8B Fig).
Further, IPA pathway analysis substantiated the observation that apoptosis of epithelial cells infected with Salmonella is delayed until 8 hpi (S8A Fig) [68], an observation in line with our host cell viability data. This delay in apoptosis is thought to be attributed to sustained AKT signalling that is accredited to the multifaceted effector SopB, a lipid phosphotransferase [69].
Further, in order for Salmonella to manipulate the host endocytic pathway, several ras-associated binding (RAB) proteins have been reported to be directly targeted by effectors (e.g., GtgE, SopD and SopD2 [70–72]. Based on our IPA analysis, autophagy has a positive activation z-score at 8 hpi (S8A Fig) with a predicted contribution of the RAB1A, RAB7A and RAB9A, all found at higher abundancies at this time point in our DIA data.
Next to demonstrating the fluctuations among functions affected during Salmonella infection, IPA revealed 60 putative upstream regulators to be either activated or inhibited over the course of infection (S8C Fig). Among others, immune system signalling regulators such as toll-like receptor signalling proteins, e.g. STAT1, IL1B, IFNA2 and TLR3 and MAPK3, were shown to be activated at 16–24 hpi.
Viewing the complex response to bacterial infection observed over time, and with some of the expression differences linked to intracellular host localization of the bacteria and shown to promote intracellular Salmonella growth [36], studies on the causal links between host/pathogen expression interdependencies remain an interesting area of future research. For this, strain-specific responses of the host to Salmonella infection (considering an e.g., SPI-2 mutant) could be interrogated to uncover unique profiles of pathogen/host interdependencies over the time course of the infection.
Moreover, with the observed variability of Salmonella subpopulations in diverse host cells [23, 24], future DIA-based dual proteome profiling infection studies performed at the sub-population or even single-cell level may further advance our understanding of host-pathogen interactions at an unprecedented resolution.
Materials and methods
Bacterial strain and growth cultivation conditions
The Salmonella enterica serovar Typhimurium (S. Typhimurium) wild-type (WT) strain SL1344 (Hoiseth and Stocker, 1981) (Genotype: hisG46, Phenotype: His(-); biotype 26i) was obtained from the Salmonella Genetic Stock Center (Salmonella Genetic Stock Center (SGSC), Calgary, Canada; cat n° 438 [73]). The GFP-expressing Salmonella SL1344 strain JVS-3858 with Ptet::gfp integrated into the put locus of the chromosome [45] was a gift from Prof. Jörg Vogel and Prof. Alexander Westermann. Bacterial growth was performed in Luria Beltrami (LB)-Miller broth (10 g/L Bacto tryptone, 5 g/L Bacto yeast extract, 10 g/L NaCl). For bacterial cultivation, single colonies were picked from LB plates, inoculated in 3 ml liquid Lennox (L) growth medium (L-broth) in round-bottom falcon tubes (Corning, cat n° 352059) and grown for ~18 h at 37°C with agitation (180 rpm). Subsequently, the cultures (~OD600 2.0) were diluted 1:50 in T25 flasks with ventilated cap in 8 ml L-medium and grown at 37°C and 180 rpm to early stationary phase (OD600 3.0) (corresponding to ~2 x 109 bacteria per ml and ~3.5 h of culture after 1/50 dilution of an overnight culture).
Cell culture
Human epithelial HeLa cells (epithelial cervix adenocarcinoma cell line, American Type Culture Collection, Manassas, VA, USA; ATCC CCL2) were cultured in GlutaMAX containing Dulbecco’s Modified Eagle Medium (DMEM) (Gibco, cat n° 31966–047) supplemented with 10% foetal bovine serum (FBS) (Gibco, cat n° 10270–106) and 50 units/ml penicillin and 50 μg/ml streptomycin (Gibco; cat n° 5070–063) (0.22 μm filtered in case of real-time imaging in 96-well plates (see below methods)). Cells were cultured at 37°C in a 5% CO2 gas-equilibrated humidified incubator and passaged every 3–4 days.
For infection, HeLa cells were seeded in a T175 (Greiner Bio One, cat n° 660160) at a density of 107 cells in 19 ml of medium for dual-proteome profiling, or at 1.5 x 104 cells per well in 96-well plates (Tecan 96 Flat Black, cat n° 30122306) in 150 μl of medium for cell-free cell-counting and fluorescence detection. The culture medium was supplemented with 10% heat-inactivated FBS (heat inactivation was done for 30 min at 56°C) and without antibiotics. In both set-ups, HeLa cells were seeded one day prior to infection thereby reaching a final confluence of ~85–90%, corresponding to approx. 1.25 x 107 HeLa cells per T175 cell culture flask and 1,875 x 104 HeLa cells per well in 96-well plates at the day of infection.
Bacterial infection of cultured HeLa cells
For bacterial invasion, the bacteria were harvested for infection at early stationary phase (when studying Salmonella spp. Invasion of epithelial cell lines, the early stationary phase was shown to be the phase during which Salmonella pathogenesis island-1 (SPI-1) is highly induced, and therefore a suitable growth phase for infection [74]) after reaching an optical density at 600 nm of 3.0. Bacterial cells were collected by centrifugation (6,000 × g, 10 min) at room temperature, washed once with pre-heated (37°C) Dulbecco’s Phosphate Buffered Saline (DPBS) (Gibco; cat n° 14190–144), re-spun and resuspended in pre-heated DMEM without serum.
Salmonella infection was allowed to proceed in DMEM at a multiplicity of infection (MOI) of 100 (i.e., per T175 cell culture flask of 1,25 x 107 HeLa cells, 1,25 x 109 bacterial cells (or 1.25 ml of a 109 bacteria bacterial cell suspension)), or in case of 96-wells, 50 μl containing 1,875 x 105 (MOI 10), 1,875 x 106 (MOI 100) or 3,75 x 106 bacterial cells (MOI 200) were added for 30 min at 37°C in a 5% CO2 gas-equilibrated humidified incubator. After infection, the medium was removed from the culture flasks or the 96-well plates, the monolayers washed with DPBS pre-heated at 37°C and respectively, 20 ml or 0,2 ml of pre-warmed DMEM supplemented with heat-inactivated FBS and 100 μg/ml gentamicin (Sigma-Aldrich, G1264-1G) was added to kill extracellular bacteria after which the infection was allowed to proceed for an additional 2 h. When the infection was allowed to proceed for longer times, after 2 h, cells were washed and DMEM supplemented with 10 μg/ml gentamicin was added. For dual proteome profiling, 4 replicate samples corresponding to 2, 4, 8, 16 and 24 hours post-infection (hpi) were collected, besides a HeLa control setup. After a DPBS wash (37°C) of the cell monolayers, cells were harvested by trypsinization (Trypsin-EDTA; Gibco, cat n° 25300054), collected by centrifugation (1000 × g, 5 min) at 4°C and the cell pellets washed 2 times with DPBS, flash frozen in liquid nitrogen and stored at -80°C until further processing.
Real-time assessment of host cell count and death, besides monitoring of Salmonella infection
For real-time measurement of label-free host cell count (indicative of the overall viability of infected cell cultures) and cell death, besides monitoring of Salmonella, 4 replicates samples infected with either WT SL1344, a constitutively eGFP-expressing derivative strain of SL1344 (i.e., JVS-3858 [45]), and a non-invasive ΔprgH mutant at different MOIs (MOI 10,100 and 200) were analysed over a 24 hours’ time course of infection. Replicates to which CellTox Green Dye was added (Promega; Cat. n° G873A) at 4x concentration according to the manufacturer’s instructions were analysed in parallel. To calibrate fluorescence read-out, a lysis control was prepared by adding 8 μl lysis buffer (Promega; Lysis Solution; G182B) (1/25) to control cell cultures. Plates were sealed with sterile, gas-permeable, optical adhesive sealing film for microplates and incubated in a small humidity cassette of a Spark Cyto 400 multi-well optical plate reader, controlled by SparkControl V3.2 software (Tecan, Mechelen, Belgium). Cultures were monitored for 24 hours and the image acquisition within the Spark Cyto performed using a user-defined application in SparkControl. A bright field center image was captured every 30 min for each well using a 10x objective (center grid, user-defined area) and fluorescence was measured every ~30 min (sensitivity 70%, object length and width ranging between 8–30 μm, analysis type: segmentation), and the count and confluence algorithms used to calculate the cell count and cell-covered area, respectively. The sensitivity of these algorithms were further refined post-acquisition using the Spark Cyto’s image analysis software; Image Analyzer. GFP and CellTox Green fluorescence was measured using bottom reading and an excitation and emission wavelength of 485 nm (width 20 nm) and 535 nm (width 20 nm), respectively.
prgH deletion mutant generation by λ Red recombineering
A S. Typhimurium SL1344 prgH mutant (ΔprgH) was constructed by λ Red recombineering [75]. Briefly, ~2.4 x 109 of λ red-induced (0.2% L–(+)–arabinose w/v) electrocompetent and pKD46 (accession number #7669, Coli Genetic Stock Center (CGSS), Yale University, USA) transformed S. Typhimurium WT cells (culture grown till OD600 0.6) were electroporated with a linear PCR-editing substrate designed to replace prgH with the kanamycin-resistance cassette from pKD4 (accession number #7632, CGSS, Yale, USA). PCR was used to amplify the antibiotic-resistance cassette with 5’ and 3’ 50 bp homology arms complementary to the flanking regions of prgH using primer sequences caatggggatgatggttcttttaatatgtgttgagacgcattatacagaatgtgtaggctggagctgcttc and acgccgccagtagcgccggatcggagggttttgctgctaatttatccagctatgaatatcctccttagttcctatt). Immediately following electroporation, bacterial cells were recovered in pre-warmed (28°C) Super Optimal broth with Catabolite repression (SOC, consisting of 2% tryptone, 0.5% yeast extract, 10 mM NaCl, 2.5 mM KCl, 5 mM MgCl2, 10 mM MgSO4 and 20 mM glucose) media, incubated at 180 rpm for 1 h at 28°C and subsequently at 37°C for 3 h. Mutant colonies were selected after plating the cell suspension and overnight incubation (upside down) on LB agar plates supplemented with 25 μg/ml kanamycin at 37°C. Allelic replacement of prgH was confirmed by colony PCR using primer sequences ggatgatggttcttttaatatgt and ggcaagaaagccatccagtttactttgca, and acgccgccagtagcgccggatcg and gacgagttcttctgagcgggact, and the resulting PCR products sequence verified by Sanger sequencing (Applied Biosystems 3730xl, Thermo, VIB Genomic Service Facility, University of Antwerp, Belgium).
Proteome extractions and sample mixes
Cell pellets of 4 replicate samples and 6 setups (control HeLa + 2, 4, 8, 16 and 24 hpi samples) were resuspended in guanidinium-hydrochloride (Gu.HCl) lysis buffer at 20 x 106 cells/ml (4 M Gu.HCl, 50 mm ammonium bicarbonate (pH 7.9)) and lysed by three rounds of freeze-thaw lysis in liquid nitrogen. The lysates were sonicated (Branson probe sonifier output 40, 50% duty cycle, 3×30 s, 1 s pulses) followed by centrifugation (16,100 x g, 10 min at 4°C) to remove cellular debris. The protein concentration of the supernatant was determined by Bradford measurement [76] according to the manufacturer’s instructions (Bio-Rad, cat n° 5000006).
An aliquot adjusted with lysis buffer to 2 mg/ml, and equivalent of 400 μg of total protein (200 μl) was transferred to a clean Eppendorf tube, twice diluted with HPLC-grade water, and precipitated with 4 volumes of acetone (at-20°C) overnight. The precipitated protein material was recovered by centrifugation for 15 min at 16,000 x g at 4°C, pellets washed twice with cold 80% acetone, and air dried upside down for ~10 min at room temperature or until no residual acetone odour remained. Protein pellets were resuspended in 200 μl TFE (2,2,2-trifluoroethanol) digestion buffer (10% TFE, 100 mM ammonium bicarbonate pH 7.9) with sonication (Branson probe output 20; 1 s pulses) until a homogenous suspension was obtained. All samples were digested overnight at 37°C using MS-grade trypsin/Lys-C Mix (Promega, Madison, WI) (enzyme/substrate of 1:50 w/w) while mixing (550 rpm). Samples were acidified with TFA to a final concentration of 0.5%, cleared from insoluble particulates by centrifugation for 10 min at 16,000 x g at 4°C and the supernatant transferred to new tubes. Methionine oxidation was performed by addition of hydrogen peroxide to a final concentration of 0.5% for 30 min at 30°C. Solid phase extraction of peptides was performed using C18 reversed phase sorbent containing 100 μl pipette tips Bond Elut OMIX 100 μl C18 tips (Agilent, Santa Clara, CA, USA, cat n° A57003100K) according to the manufacturer’s instructions. The pipette tip was conditioned by aspirating the maximum pipette tip volume of water:acetonitrile, 50:50 (v/v) and the solvent discarded. After equilibration of the tip by washing three times with the maximum pipette tip volume in 0.1% TFA in water, 100 μl of the acidified samples (original input of ~200 μg) were dispensed and aspirated for 10 cycles for maximum binding efficiency. The tip was washed three times with the maximum pipette tip volume of 0.1% TFA in water:acetonitrile, 98:2 (v/v) and the bound peptides eluted in LC-MS/MS vials with the maximum pipette tip volume of 0.1% TFA in water:acetonitrile, 30:70 (v/v). The samples were vacuum-dried in a SpeedVac concentrator and re-dissolved in 50 μl of 2 mM tris(2-carboxyethyl)phosphine (TCEP) in 2% acetonitrile spiked with an indexed Retention Time or iRT peptide mix (i.e., a mixture of eleven non-naturally occurring synthetic peptides added according to manufacturer’s instructions (Biognosys)), to enable retention time (RT) prediction (see below). Samples were stored at -20°C until LC-MS/MS analysis.
LC-MS/MS data acquisition
For mass spectrometry analyses, 10 μl was injected from each sample for LC-MS/MS analysis on an Ultimate 3000 RSLCnano system in-line connected to a Q Exactive HF Hybrid Quadrupole-Orbitrap BioPharma mass spectrometer (Thermo). Trapping was performed at 10 μL/min for 4 min in loading solvent A (0.1% TFA in water/I (98:2, v/v) on a 20 mm trapping column (made in-house, 100 μm internal diameter (I.D.), 5 μm beads, C18 Reprosil-HD, Dr. Maisch, Germany). After flushing from the trapping column, the peptides were loaded and separated on an analytical 200 cm μPAC column with C18-endcapped functionality (PharmaFluidics, Belgium) kept at a constant temperature of 50°C. Peptides were eluted by a non-linear gradient reaching 9% MS solvent B (0.1% FA in water/acetonitrile (2:8, v/v)) in 15 min, 33% MS solvent B in 105 min, 55% MS solvent B in 125 min and 99% MS solvent B in 135 min at a constant flow rate of 300 nL/min, followed by a 5-minutes wash at 99% MS solvent B and re-equilibration with MS solvent A (0.1% FA in water). For both analyses (data-dependent acquisition (DDA) and data-independent acquisition (DIA), a pneu-Nimbus dual column ionization source was used (Phoenix S&T), at a spray voltage of 2.6 kV and a capillary temperature of 275°C. For the first analysis (DDA mode), the mass spectrometer automatically switched between MS and MS2 acquisition for the 16 most abundant ion peaks per MS spectrum. Full-scan MS spectra (375–1500 m/z) were acquired at a precursor resolution of 60,000 at 200 m/z in the Orbitrap analyser after accumulation to a target value of 3,000,000. The 16 most intense ions above a threshold value of 13,000 were isolated for higher-energy collisional dissociation (HCD) fragmentation at a normalized collision energy of 28% after filling the trap at a target value of 100,000 for maximum 80 ms injection time using a dynamic exclusion of 12 s. MS2 spectra (200–2,000 m/z) were acquired at a resolution of 15,000 at 200 m/z in the Orbitrap analyser. Another equivalent 10 μL aliquot from each sample was analysed using the same mass spectrometer now operated in DIA mode. Nano LC conditions and gradients were the same as used for DDA. Full-scan MS spectra ranging from 375–1,500 m/z with a target value of 5E6 were followed by 30 quadrupole isolations with a precursor isolation width of 10 m/z for HCD fragmentation at a normalized collision energy of 30% after filling the trap at a target value of 3E6 for maximum injection time of 45 ms. MS2 spectra were acquired at a resolution of 15,000 at 200 m/z in the Orbitrap analyser without multiplexing. The isolation intervals ranged from 400–900 m/z with an overlap of m/z.
Data-dependent acquisition (DDA) data processing
For DDA-based peptide identification and quantification, Raw data files were searched by MaxQuant [47] (version 1.6.9.0). The protein search database comprises the UniProtKB Salmonella reference proteome concatenated to the human UniProtKB reference proteome (UP000008962 and UP000005640, together 74,449 proteins), as well as MaxQuant built-in contaminants. In addition, contaminant proteins present in the contaminants.fasta file that comes with MaxQuant and the 11 iRT peptide sequences (Biognosys-11) were added to the search database. Methionine oxidation was set as fixed modification and N-terminal acetylation was set as variable modification. To augment protein/peptide quantification, we enabled built-in matching-between-run and label-free quantitation (LFQ) algorithms using default settings in MaxQuant. We used the enzymatic rule of trypsin/P with a maximum of 2 missed cleavages. The main search peptide tolerance was set to 4.5 ppm and the ion trap MS/MS match tolerance was set to 0.5 Da. Default PSM, peptide and protein level FDR thresholds of 1% were used, with an additional minimal Andromeda score of 40 for modified peptides. Protein FDR was set at 1% and estimated by using the reversed search sequences. The maximal number of modifications per peptide was set to 5. For protein quantification in the proteinGroups.txt file, razor peptides were considered and further allowed all modifications.
Data-independent acquisition (DIA) data processing
DDA-based spectral library construction.
The ‘msms.txt’ files outputted by MaxQuant DDA searches were used as input for creation of redundant BLIB spectral libraries using BiblioSpec (version 2.1) [77]. Redundant spectra were subsequently filtered using the ‘BlibFilter’ function, requiring entries to have at least 5 peaks (‘-n 5’). Peptide sequences uniquely identified in only one of the DDA searches were appended to the spectral library to extend the library and to augment the detection capacity of Salmonella peptides. We transformed the peptide RTs present in the BLIB library to iRT using the spiked-in iRT peptides (Biognosys-11). To this end, empirical RTs of the top-scoring iRT peptide identifications (lowest posterior error probability, ‘msms.txt’) in the DDA samples, were used to fit a linear trendline and scale RTs. For the artificial mixtures and Salmonella pre-fractionation analyses [31], the corresponding trendlines were iRT = 1.220 RT– 74.566 and iRT = 1.189 RT– 75.039. The updated BLIB files were then converted to DLIB format by EncyclopeDIA, using the combined human-Salmonella UniProtKB proteome FASTA as background [48].
Library-free searching of proteome FASTA by PECAN/Walnut.
DIA raw data files were converted to mzML by MSConvert using vendor peakPicking. Pre-processed DIA samples were searched against a compilation of the Salmonella UniProtKB proteome (UP000008962, 4,657 proteins) and human Swiss-Prot proteome (UP000005640, 20,367 proteins) using the EncylopeDIA built-in PECAN algorithm [48, 78]. We opted to solely search the human Swiss-Prot protein database, which resulted in a ~3-fold reduction of the protein database search space (25,024 versus 74,449 proteins when combining Salmonella [UP000008962] and human [UP000005640] UniProtKB references proteomes), in order to minimize the theoretical search space. This was desired to limit the size of a predicted spectral library for all possible tryptic peptides and overall runtime and memory usage. Since 36,494 out of 36,668 (i.e., 99.53%) of all human peptides identified by MaxQuant matched a Swiss-Prot protein entry, no drastic loss in identifications is anticipated. Default settings were used, except for methionine oxidation (to methionine-sulfoxide) being set as fixed modification, and considering a maximum length of 25 amino acids and HCD as fragmentation type.
Construction of an MS2PIP-based spectral library.
MS2 spectra were predicted by MS2PIP (version 20190312) [79] for tryptic peptides derived from an in silico digest of the Salmonella UniProtKB proteome and human Swiss-Prot proteome (trypsin/P, peptide length 7–25 AA, mass 500–5,000 Da, one missed cleavage, N-terminal initiator methionine removal considered) in case of 2+ and or 3+ peptide precursor fit within 400 to 900 m/z (scanned range DIA). This yields a total of 1,586,777 predicted MS2 spectra for 1,151,386 peptides solely matching human proteins, 197,782 spectra for 144,156 peptides matching Salmonella proteins, and 117 spectra for 110 peptides matching both species. We set methionine oxidation (to methionine-sulfoxide) as a fixed modification for MS2 prediction by MS2PIP. Predicted spectra were supplemented with DeepLC predicted RTs using a model trained on RTs of 35,206 non-redundant peptides identified in DIA PECAN searches (peptide Q-value < 0.01) (as described in the above section).
Hybrid library construction.
The chromatogram libraries (ELIB) generated by EncyclopeDIA after PECAN and MS2PIP-library searching were combined into a single redundant library. After conversion to BLIB format, and adding the ‘:redundant:’ tag in library info (sqlite3), the ‘BlibFilter’ function was used to create a non-redundant DIA spectral library as described above. This DIA spectral library was further extended with spectra from the extended DDA spectral library (see above) corresponding to peptides not yet contained within the DIA spectral library. The resulting hybrid DDA/DIA spectral library (i.e., joining DDA- and DIA-based identifications) was then converted to DLIB format as described above.
EncyclopeDIA spectral library searching and peptide quantification.
The resulting mzML files were searched against the DDA, MS2PIP, or hybrid spectral DLIB libraries using EncylopeDIA software (version 0.90) [48] with default settings. Sample-specific Percolator output files and chromatogram libraries were stored. Per setup, a combined chromatogram library was created consisting of the three replicates. This performs a Percolator re-running of the combined results and provides peptide and protein quantifications at a 1% peptide and protein Q-value, respectively. For quantification, the number of minimum required and quantifiable ions were set at 5 with aligning between samples enabled.
MS2PIP, Elude and EncyclopeDIA are open source, licensed under the Apache-2.0 License, and are hosted on https://github.com/compomics/ms2pip_c, https://github.com/percolator/percolator and https://bitbucket.org/searleb/encyclopedia/wiki/Home.
Software packages for statistics and data visualization
For basic data handling, normalization, statistics (if not stated otherwise) and annotation enrichment analysis, we used the freely available open-source bioinformatics platform Perseus (http://141.61.102.17/perseus_doku/doku.php?id=start) (v1.6.5.0). Perseus was used to analyse data from principal component analysis, non-supervised hierarchical clustering, and scatter plots. Furthermore, the 1D and 2D annotation algorithms and Fisher’s exact tests implemented in Perseus (Cox & Mann, 2012) were used for annotation enrichment analysis. Heat maps, PCA plots and bar charts were generated using GraphPad Prism software (www.graphpad.com).
IPA Pathway analysis
For the different time points post-infection, significantly regulated human proteins (t-test, p-value ≤ 0.01; described above) were characterized by core analysis in IPA software (version 68752261, QIAGEN Inc., https://www.qiagenbioinformatics.com/products/ingenuity-pathway-analysis) using the t-test differences as directional expression values. Upstream regulators were determined using a p-value of overlap (cutoff set at 0.01) and z-scores calculated using the algorithm implemented in IPA [80].
Supporting information
S1 Fig. Characterization of epithelial Salmonella infection model.
HeLa cells (1,875 x 104 HeLa cells) were non-infected or infected with S. Typhimurium (WT or ΔprgH SL1344) at MOIs 10 (blue), 100 (red) or 200 (green) in quadruplicate and cell count and cytotoxicity (or GFP fluorescence) measured in a multi-well optical plate reader in real-time. Label-free cell counting (10x objective) of HeLa cell cultures infected with WT Salmonella (A) or non-invasive ΔprgH mutant Salmonella (B). Monitoring cell death in real-time using the CellTox Green Cytotoxicity Assay (C-D). Using a constitutively eGFP-expressing strain of SL1344 [45] (WT eGFP) green objects were counted and their average size (μm2) determined (F). Due to technical measurement errors, the 17.5 and 18 hpi recordings were omitted.
https://doi.org/10.1371/journal.ppat.1011183.s001
(TIF)
S2 Fig. Variability of DDA and DIA replicate samples represented in PCA plots.
Individual DDA (A-C) and DIA (B-D) PCA plots for Salmonella (C-D) as well as HeLa (A-B) over the time-course of infection. Green, cyan, blue, orange, purple and black circles represent control (Ctrl), 2, 4, 8, 16 and 24 hpi samples, respectively.
https://doi.org/10.1371/journal.ppat.1011183.s002
(TIF)
S3 Fig. Correlations (Pearson) between all setups and replicate samples analysed.
Ctrl HeLa, 2 hpi, 4 hpi, 8 hpi, 16 hpi, and 24 hpi.
https://doi.org/10.1371/journal.ppat.1011183.s003
(TIF)
S4 Fig. Average mean correlations (Pearson) between setups analysed.
Ctrl HeLa, 2 hpi, 4 hpi, 8 hpi, 16 hpi, and 24 hpi.
https://doi.org/10.1371/journal.ppat.1011183.s004
(TIF)
S5 Fig. Profile plots of regulated S. Typhimurium proteins over the time-course of infection.
Significant regulation (FDR ≤ 0.01) was determined across the 5 timepoints post-infection using multiple t-testing and corresponding normalized averaged z-scores of S. Typhimurium DDA (left panels) and DIA data (right panels) plotted, with corresponding members of the two main clusters (i.e., up- and downregulated cluster) shown and its corresponding number of regulated S. Typhimurium protein groups indicated.
https://doi.org/10.1371/journal.ppat.1011183.s005
(TIF)
S6 Fig. Profile plots and heat map representation of regulated human proteins over the time-course of infection.
Regulation was determined across the 5 timepoints post-infection and control cells using t-testing for normalized DIA data. The intensities of significantly (FDR ≤ 0.01) regulated human proteins (#1,742) from the DIA analysis are shown. The averaged expression values of corresponding members of the 7 main clusters after ANOVA are shown as profile plots (A) and as a heat map (B), with the corresponding number of regulated protein groups indicated per cluster (see also S5 Table). In panel B, green indicates low intensities while red indicates high intensities.
https://doi.org/10.1371/journal.ppat.1011183.s006
(TIF)
S7 Fig. 2D annotation enrichment of human proteome and transcriptome expression changes.
The 2D-annotation enrichment scores for representative GO biological processes (GOBP), GO cellular component (GOCC), GO molecular function (GOMF) and keywords were plotted for the proteome (DIA data) and transcriptome expression changes (GEO series accession number project GSE60144, [41]) at 2, 4, 16 and 24 hpi (no 2D enriched terms were observed at 8 hpi). The fold change of log2 transformed means of corresponding LFQ and RPKM values were plotted and only corrected p values ≤0.01 were considered.
https://doi.org/10.1371/journal.ppat.1011183.s007
(TIF)
S8 Fig. Selection of functions and regulators predicted to be activated or inhibited using Ingenuity Pathway Analysis (IPA).
(A) Heatmap visualization (IPA activation z-score) of selected IPA functions regulated over the course of the infection. (B) Overview of regulated genes (#85) (log FC) contributing to the IPA function “formation of actin filaments” (see also S5 Table). (C) Overview of activated and inhibited upstream regulators. Heatmap representation of 60 predicted proteinaceous (activated or inhibited for at least one of the timepoints post-infection analysed) upstream regulators (p-value ≤ 0.01), colour coded according to the IPA activation z-score.
https://doi.org/10.1371/journal.ppat.1011183.s008
(TIF)
S1 Table. List of 6,696 proteins identified in the dual proteome shotgun samples of S. Typhimurium infected human HeLa cells by means of data-dependent acquisition (DDA).
S. Typhimurium infected human HeLa cells were analysed to monitor quantitative differences in steady-state protein expression levels at 2, 4, 8, 16 and 24 hpi in biological quadruplicates. UniProt database primary accession number, UniProt entry (name), gene name, corresponding protein description and sequence, molecular protein weight, number of (razor/unique) peptides identified, (razor/unique) sequence coverage, score, log2-transformed LFQ protein intensities, GO annotation and keywords, MS/MS counts and sequence length(s) are given. Ranking is done alphabetically according to protein names.
https://doi.org/10.1371/journal.ppat.1011183.s009
(XLSX)
S2 Table. List of 6,734 proteins identified in the dual proteome shotgun samples of S. Typhimurium infected human HeLa cells by means of data-independent acquisition (DIA).
S. Typhimurium infected human HeLa cells were analysed to monitor quantitative differences in steady-state protein expression levels at 2, 4, 8, 16 and 24 hpi in biological quadruplicates. UniProt database primary accession number, UniProt entry (name), gene name, corresponding protein description and sequence, sequence length, number (sequence) of peptides identified, species, log2-transformed protein intensities, GO annotation and keywords are given. Ranking is done alphabetically according to protein names.
https://doi.org/10.1371/journal.ppat.1011183.s010
(XLSX)
S3 Table. List of 4,666 human proteins quantified with a minimum of three valid values in at least one setup of the dual proteome shotgun samples of S. Typhimurium infected human or control HeLa cells by means of data-dependent acquisition (DDA).
Comparisons were analysed by t-testing and significant hits determined using as cut-off values a permutation based false discovery rate (FDR) of 0.01 and 0.001 (100 permutations) and a background variance parameter S0 of 0.1. Table headers are as in S1 Table.
https://doi.org/10.1371/journal.ppat.1011183.s011
(XLSX)
S4 Table. List of 479 S. Typhimurium proteins quantified with a minimum of three valid values in at least one setup of the dual proteome shotgun samples of S. Typhimurium infected human HeLa cells by means of data-dependent acquisition (DDA).
Comparisons were analysed by t-testing and significant hits determined using as cut-off values a permutation based false discovery rate (FDR) of 0.05 and 0.01 (100 permutations) and a background variance parameter S0 of 1. Table headers are as in S1 Table.
https://doi.org/10.1371/journal.ppat.1011183.s012
(XLSX)
S5 Table. List of 5,818 human proteins quantified with a minimum of three valid values in at least one setup of the dual proteome shotgun samples of S. Typhimurium infected human or control HeLa cells by means of data-independent acquisition (DIA).
Comparisons were analysed by t-testing and significant hits determined using as cut-off values a permutation based false discovery rate (FDR) of 0.01 and 0.001 (100 permutations) and a background variance parameter S0 of 0.1. (Regulated) host proteins previously found associated with Salmonella-modified membranes (SMM) [52] are indicated in indicated in columns ‘SMM proteome (PMID: 25348832)’ and/or ‘ANOVA regulated SMM protein’, respectively. Remaining table headers are as in S2 Table and cluster indication is as in S6 Fig.
https://doi.org/10.1371/journal.ppat.1011183.s013
(XLSX)
S6 Table. List of 844 S. Typhimurium proteins quantified with a minimum of three valid values in at least one setup of the dual proteome shotgun samples of S. Typhimurium infected human HeLa cells by means of data-independent acquisition (DIA).
Comparisons were analysed by t-testing and significant hits determined using as cut-off values a permutation based false discovery rate (FDR) of 0.05 and 0.01 (100 permutations) and a background variance parameter S0 of 1. Table headers are as in S2 Table with regulon and genome locus and strand info added.
https://doi.org/10.1371/journal.ppat.1011183.s014
(XLSX)
S7 Table. Significantly regulated (FDR ≤ 0.02) normalized human annotation enrichment scores across all 6 conditions (averaged expression values) assayed by data-independent acquisition (DIA).
Term enrichment was determined using the 1D annotation enrichment algorithm embedded in the Perseus software suite and p-values were corrected for multiple hypotheses testing using the Benjamini and Hochberg false discovery rate.
https://doi.org/10.1371/journal.ppat.1011183.s015
(XLSX)
S8 Table. Regulated host proteins associated with Salmonella-modified membranes, implicated in transport, membrane or cytoskeleton organization, or representing GTPases.
From the 84 regulated SMM proteins identified in total according to [52], 29 ANOVA regulated host proteins associated with Salmonella-modified membranes, implicated in transport, membrane or cytoskeleton organization, or representing GTPases (selected from keywords) are listed. Table headers are as in S2 Table and cluster indication as in S6 Fig.
https://doi.org/10.1371/journal.ppat.1011183.s016
(XLSX)
Acknowledgments
We would like to acknowledge An Staes for her help performing LC-MS sample analyses and thank Prof. Jörg Vogel and Prof. Alexander Westermann for providing us the GFP-expressing Salmonella SL1344 strain JVS-3858 (Ptet::gfp) [22].
References
- 1. Hume PJ, Singh V, Davidson AC, Koronakis V. Swiss Army Pathogen: The Salmonella Entry Toolkit. Front Cell Infect Microbiol. 2017;7:348. pmid:28848711
- 2. Castanheira S, Garcia-Del Portillo F. Salmonella Populations inside Host Cells. Front Cell Infect Microbiol. 2017;7:432. pmid:29046870
- 3. Haraga A, Ohlson MB, Miller SI. Salmonellae interplay with host cells. Nat Rev Microbiol. 2008;6(1):53–66. pmid:18026123
- 4. Darwin KH, Miller VL. Molecular basis of the interaction of Salmonella with the intestinal mucosa. Clin Microbiol Rev. 1999;12(3):405–28. pmid:10398673
- 5. Que F, Wu S, Huang R. Salmonella pathogenicity island 1(SPI-1) at work. Curr Microbiol. 2013;66(6):582–7. pmid:23370732
- 6. Figueira R, Holden DW. Functions of the Salmonella pathogenicity island 2 (SPI-2) type III secretion system effectors. Microbiology (Reading). 2012;158(Pt 5):1147–61. pmid:22422755
- 7. LaRock DL, Chaudhary A, Miller SI. Salmonellae interactions with host processes. Nat Rev Microbiol. 2015;13(4):191–205. pmid:25749450
- 8. McGhie EJ, Brawn LC, Hume PJ, Humphreys D, Koronakis V. Salmonella takes control: effector-driven manipulation of the host. Curr Opin Microbiol. 2009;12(1):117–24. pmid:19157959
- 9. Jones BD, Paterson HF, Hall A, Falkow S. Salmonella typhimurium induces membrane ruffling by a growth factor-receptor-independent mechanism. Proc Natl Acad Sci U S A. 1993;90(21):10390–4. pmid:8234304
- 10. Perrett CA, Jepson MA. Regulation of Salmonella-induced membrane ruffling by SipA differs in strains lacking other effectors. Cell Microbiol. 2009;11(3):475–87. pmid:19046340
- 11. Fattinger SA, Sellin ME, Hardt WD. Epithelial inflammasomes in the defense against Salmonella gut infection. Curr Opin Microbiol. 2021;59:86–94. pmid:33128958
- 12. Malik-Kale P, Jolly CE, Lathrop S, Winfree S, Luterbach C, Steele-Mortimer O. Salmonella—at home in the host cell. Front Microbiol. 2011;2:125. pmid:21687432
- 13. Brumell JH, Goosney DL, Finlay BB. SifA, a type III secreted effector of Salmonella typhimurium, directs Salmonella-induced filament (Sif) formation along microtubules. Traffic. 2002;3(6):407–15. pmid:12010459
- 14. Birmingham CL, Jiang X, Ohlson MB, Miller SI, Brumell JH. Salmonella-induced filament formation is a dynamic phenotype induced by rapidly replicating Salmonella enterica serovar typhimurium in epithelial cells. Infect Immun. 2005;73(2):1204–8. pmid:15664965
- 15. Garcia-del Portillo F, Zwick MB, Leung KY, Finlay BB. Salmonella induces the formation of filamentous structures containing lysosomal membrane glycoproteins in epithelial cells. Proc Natl Acad Sci U S A. 1993;90(22):10544–8. pmid:8248143
- 16. Rajashekar R, Liebl D, Seitz A, Hensel M. Dynamic remodeling of the endosomal system during formation of Salmonella-induced filaments by intracellular Salmonella enterica. Traffic. 2008;9(12):2100–16. pmid:18817527
- 17. Liss V, Swart AL, Kehl A, Hermanns N, Zhang Y, Chikkaballi D, et al. Salmonella enterica Remodels the Host Cell Endosomal System for Efficient Intravacuolar Nutrition. Cell Host Microbe. 2017;21(3):390–402. pmid:28238623
- 18. Malik-Kale P, Winfree S, Steele-Mortimer O. The bimodal lifestyle of intracellular Salmonella in epithelial cells: replication in the cytosol obscures defects in vacuolar replication. PLoS One. 2012;7(6):e38732. pmid:22719929
- 19. Szeto J, Namolovan A, Osborne SE, Coombes BK, Brumell JH. Salmonella-containing vacuoles display centrifugal movement associated with cell-to-cell transfer in epithelial cells. Infect Immun. 2009;77(3):996–1007. pmid:19103768
- 20. Knodler LA, Nair V, Steele-Mortimer O. Quantitative assessment of cytosolic Salmonella in epithelial cells. PLoS One. 2014;9(1):e84681. pmid:24400108
- 21. Laughlin RC, Knodler LA, Barhoumi R, Payne HR, Wu J, Gomez G, et al. Spatial segregation of virulence gene expression during acute enteric infection with Salmonella enterica serovar Typhimurium. mBio. 2014;5(1):e00946–13. pmid:24496791
- 22. Stapels DAC, Hill PWS, Westermann AJ, Fisher RA, Thurston TL, Saliba AE, et al. Salmonella persisters undermine host immune defenses during antibiotic treatment. Science (New York, NY). 2018;362(6419):1156–60. pmid:30523110
- 23. Helaine S, Thompson JA, Watson KG, Liu M, Boyle C, Holden DW. Dynamics of intracellular bacterial replication at the single cell level. Proc Natl Acad Sci U S A. 2010;107(8):3746–51. pmid:20133586
- 24. Helaine S, Cheverton AM, Watson KG, Faure LM, Matthews SA, Holden DW. Internalization of Salmonella by macrophages induces formation of nonreplicating persisters. Science (New York, NY). 2014;343(6167):204–8. pmid:24408438
- 25. Majstorovic A, Hill PWS, Helaine S. Salmonella antibiotic persistence and macrophage polarisation. Access Microbiology. 2020;2(1).
- 26. Kaiser P, Regoes RR, Dolowschiak T, Wotzka SY, Lengefeld J, Slack E, et al. Cecum lymph node dendritic cells harbor slow-growing bacteria phenotypically tolerant to antibiotic treatment. PLoS Biol. 2014;12(2):e1001793. pmid:24558351
- 27. Mangan MW, Lucchini S, Danino V, Croinin TO, Hinton JC, Dorman CJ. The integration host factor (IHF) integrates stationary-phase and virulence gene expression in Salmonella enterica serovar Typhimurium. Mol Microbiol. 2006;59(6):1831–47. pmid:16553887
- 28. Diard M, Garcia V, Maier L, Remus-Emsermann MN, Regoes RR, Ackermann M, et al. Stabilization of cooperative virulence by the expression of an avirulent phenotype. Nature. 2013;494(7437):353–6. pmid:23426324
- 29. Keren L, van Dijk D, Weingarten-Gabbay S, Davidi D, Jona G, Weinberger A, et al. Noise in gene expression is coupled to growth rate. Genome Res. 2015;25(12):1893–902. pmid:26355006
- 30. Fels U, Gevaert K, Van Damme P. Proteogenomics in Aid of Host-Pathogen Interaction Studies: A Bacterial Perspective. Proteomes. 2017;5(4). pmid:29019919
- 31. Willems P, Fels U, Staes A, Gevaert K, Van Damme P. Use of Hybrid Data-Dependent and -Independent Acquisition Spectral Libraries Empowers Dual-Proteome Profiling. J Proteome Res. 2021. pmid:33467856
- 32. Qi L, Hu M, Fu J, Liu Y, Wu M, Yu K, et al. Quantitative proteomic analysis of host epithelial cells infected by Salmonella enterica serovar Typhimurium. Proteomics. 2017;17(13–14). pmid:28544771
- 33. Li Z, Liu Y, Fu J, Zhang B, Cheng S, Wu M, et al. Salmonella Proteomic Profiling during Infection Distinguishes the Intracellular Environment of Host Cells. mSystems. 2019;4(2). pmid:30984873
- 34. Vogels MW, van Balkom BW, Heck AJ, de Haan CA, Rottier PJ, Batenburg JJ, et al. Quantitative proteomic identification of host factors involved in the Salmonella typhimurium infection cycle. Proteomics. 2011;11(23):4477–91. pmid:21919203
- 35. Noster J, Chao TC, Sander N, Schulte M, Reuter T, Hansmeier N, et al. Proteomics of intracellular Salmonella enterica reveals roles of Salmonella pathogenicity island 2 in metabolism and antioxidant defense. PLoS Pathog. 2019;15(4):e1007741. pmid:31009521
- 36. Liu Y, Zhang Q, Hu M, Yu K, Fu J, Zhou F, et al. Proteomic Analyses of Intracellular Salmonella enterica Serovar Typhimurium Reveal Extensive Bacterial Adaptations to Infected Host Epithelial Cells. Infect Immun. 2015;83(7):2897–906. pmid:25939512
- 37. Liu Y, Yu K, Zhou F, Ding T, Yang Y, Hu M, et al. Quantitative Proteomics Charts the Landscape of Salmonella Carbon Metabolism within Host Epithelial Cells. J Proteome Res. 2017;16(2):788–97. pmid:28152601
- 38. Rogers LD, Brown NF, Fang Y, Pelech S, Foster LJ. Phosphoproteomic analysis of Salmonella-infected cells identifies key kinase regulators and SopB-dependent host phosphorylation events. Sci Signal. 2011;4(191):rs9. pmid:21934108
- 39. D’Costa VM, Coyaud E, Boddy KC, Laurent EMN, St-Germain J, Li T, et al. BioID screen of Salmonella type 3 secreted effectors reveals host factors involved in vacuole positioning and stability during infection. Nat Microbiol. 2019;4(12):2511–22. pmid:31611645
- 40. Beuzon CR, Salcedo SP, Holden DW. Growth and killing of a Salmonella enterica serovar Typhimurium sifA mutant strain in the cytosol of different host cell lines. Microbiology (Reading, England). 2002;148(Pt 9):2705–15. pmid:12213917
- 41. Westermann AJ, Förstner KU, Amman F, Barquist L, Chao Y, Schulte LN, et al. Dual RNA-seq unveils noncoding RNA functions in host–pathogen interactions. Nature. 2016;529(7587):496–501. pmid:26789254
- 42. Voznica J, Gardella C, Belotserkovsky I, Dufour A, Enninga J, Stevenin V. Identification of Parameters of Host Cell Vulnerability during Salmonella Infection by Quantitative Image Analysis and Modeling. Infect Immun. 2018;86(1). pmid:29084895
- 43. Behlau I, Miller SI. A PhoP-repressed gene promotes Salmonella typhimurium invasion of epithelial cells. J Bacteriol. 1993;175(14):4475–84. pmid:8392513
- 44. Liu Y, Mi Y, Mueller T, Kreibich S, Williams EG, Van Drogen A, et al. Multi-omic measurements of heterogeneity in HeLa cells across laboratories. Nat Biotechnol. 2019;37(3):314–22. pmid:30778230
- 45. Hautefort I, Proenca MJ, Hinton JC. Single-copy green fluorescent protein gene fusions allow accurate measurement of Salmonella gene expression in vitro and during infection of mammalian cells. Appl Environ Microbiol. 2003;69(12):7480–91. pmid:14660401
- 46. Knodler LA. Salmonella enterica: living a double life in epithelial cells. Curr Opin Microbiol. 2015;23:23–31. pmid:25461569
- 47. Cox J, Mann M. MaxQuant enables high peptide identification rates, individualized p.p.b.-range mass accuracies and proteome-wide protein quantification. Nat Biotechnol. 2008;26(12):1367–72. pmid:19029910
- 48. Searle BC, Pino LK, Egertson JD, Ting YS, Lawrence RT, MacLean BX, et al. Chromatogram libraries improve peptide detection and quantification by data independent acquisition mass spectrometry. Nat Commun. 2018;9(1):5128. pmid:30510204
- 49. Cox J, Mann M. 1D and 2D annotation enrichment: a statistical method integrating quantitative proteomics with complementary high-throughput data. BMC Bioinformatics. 2012;13 Suppl 16:S12. pmid:23176165
- 50. Naba A, Clauser KR, Hoersch S, Liu H, Carr SA, Hynes RO. The matrisome: in silico definition and in vivo characterization by proteomics of normal and tumor extracellular matrices. Mol Cell Proteomics. 2012;11(4):M111 014647. pmid:22159717
- 51. Gomez-Lamarca MJ, Cobreros-Reguera L, Ibanez-Jimenez B, Palacios IM, Martin-Bermudo MD. Integrins regulate epithelial cell differentiation by modulating Notch activity. J Cell Sci. 2014;127(Pt 21):4667–78. pmid:25179603
- 52. Vorwerk S, Krieger V, Deiwick J, Hensel M, Hansmeier N. Proteomes of host cell membranes modified by intracellular activities of Salmonella enterica. Mol Cell Proteomics. 2015;14(1):81–92. pmid:25348832
- 53. Smith AC, Heo WD, Braun V, Jiang X, Macrae C, Casanova JE, et al. A network of Rab GTPases controls phagosome maturation and is modulated by Salmonella enterica serovar Typhimurium. J Cell Biol. 2007;176(3):263–8. pmid:17261845
- 54. Reuter T, Vorwerk S, Liss V, Chao TC, Hensel M, Hansmeier N. Proteomic Analysis of Salmonella-modified Membranes Reveals Adaptations to Macrophage Hosts. Mol Cell Proteomics. 2020;19(5):900–12. pmid:32102972
- 55. Spano S, Galan JE. Taking control: Hijacking of Rab GTPases by intracellular bacterial pathogens. Small GTPases. 2018;9(1–2):182–91. pmid:28632996
- 56. Auweter SD, Bhavsar AP, de Hoog CL, Li Y, Chan YA, van der Heijden J, et al. Quantitative mass spectrometry catalogues Salmonella pathogenicity island-2 effectors and identifies their cognate host binding partners. J Biol Chem. 2011;286(27):24023–35. pmid:21566117
- 57. Miao EA, Brittnacher M, Haraga A, Jeng RL, Welch MD, Miller SI. Salmonella effectors translocated across the vacuolar membrane interact with the actin cytoskeleton. Mol Microbiol. 2003;48(2):401–15. pmid:12675800
- 58. Colgan AM, Kroger C, Diard M, Hardt WD, Puente JL, Sivasankaran SK, et al. The Impact of 18 Ancestral and Horizontally-Acquired Regulatory Proteins upon the Transcriptome and sRNA Landscape of Salmonella enterica serovar Typhimurium. PLoS Genet. 2016;12(8):e1006258. pmid:27564394
- 59. Powers TR, Haeberle AL, Predeus AV, Hammarlof DL, Cundiff JA, Saldana-Ahuactzi Z, et al. Intracellular niche-specific profiling reveals transcriptional adaptations required for the cytosolic lifestyle of Salmonella enterica. PLoS Pathog. 2021;17(8):e1009280. pmid:34460873
- 60. Hannemann S, Gao B, Galan JE. Salmonella modulation of host cell gene expression promotes its intracellular growth. PLoS Pathog. 2013;9(10):e1003668. pmid:24098123
- 61. Kroger C, Colgan A, Srikumar S, Handler K, Sivasankaran SK, Hammarlof DL, et al. An infection-relevant transcriptomic compendium for Salmonella enterica Serovar Typhimurium. Cell Host Microbe. 2013;14(6):683–95. pmid:24331466
- 62. Fijalkowski I, Peeters MKR, Van Damme P. Small Protein Enrichment Improves Proteomics Detection of sORF Encoded Polypeptides. Front Genet. 2021;12:713400. pmid:34721520
- 63. Zhou D, Chen LM, Hernandez L, Shears SB, Galan JE. A Salmonella inositol polyphosphatase acts in conjunction with other bacterial effectors to promote host cell actin cytoskeleton rearrangements and bacterial internalization. Mol Microbiol. 2001;39(2):248–59. pmid:11136447
- 64. De Meyer M, Fijalkowski I, Jonckheere V, De Sutter D, Eyckerman S, Van Damme P. Capturing Salmonella SspH2 Host Targets in Virus-Like Particles. Front Med (Lausanne). 2021;8:725072. pmid:34568381
- 65. Fiskin E, Bionda T, Dikic I, Behrends C. Global Analysis of Host and Bacterial Ubiquitinome in Response to Salmonella Typhimurium Infection. Mol Cell. 2016;62(6):967–81. pmid:27211868
- 66. Fu Y, Galan JE. A salmonella protein antagonizes Rac-1 and Cdc42 to mediate host-cell recovery after bacterial invasion. Nature. 1999;401(6750):293–7. pmid:10499590
- 67. Friebel A, Ilchmann H, Aepfelbacher M, Ehrbar K, Machleidt W, Hardt WD. SopE and SopE2 from Salmonella typhimurium activate different sets of RhoGTPases of the host cell. J Biol Chem. 2001;276(36):34035–40. pmid:11440999
- 68. Paesold G, Guiney DG, Eckmann L, Kagnoff MF. Genes in the Salmonella pathogenicity island 2 and the Salmonella virulence plasmid are essential for Salmonella-induced apoptosis in intestinal epithelial cells. Cell Microbiol. 2002;4(11):771–81. pmid:12427099
- 69. Knodler LA, Finlay BB, Steele-Mortimer O. The Salmonella effector protein SopB protects epithelial cells from apoptosis by sustained activation of Akt. J Biol Chem. 2005;280(10):9058–64. pmid:15642738
- 70. Lian H, Jiang K, Tong M, Chen Z, Liu X, Galan JE, et al. The Salmonella effector protein SopD targets Rab8 to positively and negatively modulate the inflammatory response. Nat Microbiol. 2021;6(5):658–71. pmid:33603205
- 71. D’Costa VM, Braun V, Landekic M, Shi R, Proteau A, McDonald L, et al. Salmonella Disrupts Host Endocytic Trafficking by SopD2-Mediated Inhibition of Rab7. Cell Rep. 2015;12(9):1508–18. pmid:26299973
- 72. Savitskiy S, Wachtel R, Pourjafar-Dehkordi D, Kang HS, Trauschke V, Lamb DC, et al. Proteolysis of Rab32 by Salmonella GtgE induces an inactive GTPase conformation. iScience. 2021;24(1):101940. pmid:33426511
- 73. Hoiseth SK, Stocker BA. Aromatic-dependent Salmonella typhimurium are non-virulent and effective as live vaccines. Nature. 1981;291(5812):238–9. pmid:7015147
- 74. Ibarra JA, Knodler LA, Sturdevant DE, Virtaneva K, Carmody AB, Fischer ER, et al. Induction of Salmonella pathogenicity island 1 under different growth conditions can affect Salmonella–host cell interactions in vitro. Microbiology (Reading, England). 2010;156(Pt 4):1120–33. pmid:20035008
- 75. Datsenko KA, Wanner BL. One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products. Proc Natl Acad Sci U S A. 2000;97(12):6640–5. pmid:10829079
- 76. Bradford MM. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Analytical biochemistry. 1976;72:248–54. pmid:942051
- 77. Frewen B, MacCoss MJ. Using BiblioSpec for creating and searching tandem MS peptide libraries. Current protocols in bioinformatics. 2007;Chapter 13:Unit 13.7. pmid:18428681
- 78. Ting YS, Egertson JD, Bollinger JG, Searle BC, Payne SH, Noble WS, et al. PECAN: library-free peptide detection for data-independent acquisition tandem mass spectrometry data. Nature methods. 2017;14(9):903–8. pmid:28783153
- 79. Degroeve S, Martens L. MS2PIP: a tool for MS/MS peak intensity prediction. Bioinformatics (Oxford, England). 2013;29(24):3199–203. pmid:24078703
- 80. Kramer A, Green J, Pollard J Jr., Tugendreich S. Causal analysis approaches in Ingenuity Pathway Analysis. Bioinformatics. 2014;30(4):523–30. pmid:24336805
- 81. Perez-Riverol Y, Csordas A, Bai J, Bernal-Llinares M, Hewapathirana S, Kundu DJ, et al. The PRIDE database and related tools and resources in 2019: improving support for quantification data. Nucleic Acids Res. 2019;47(D1):D442–d50. pmid:30395289
- 82. Foster ED, Deardorff A. Open Science Framework (OSF). J Med Libr Assoc. 2017;105(2):203–6.