SEARCH

SEARCH BY CITATION

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Growth of Salmonella within cultured eukaryotic cells
  5. Salmonella virulence and the repression of pathogen functions in the host
  6. Active role of Salmonella in restricting intracellular growth in vivo
  7. Conclusions
  8. Acknowledgements
  9. References

The literature refers to Salmonella enterica as an intracellular bacterial pathogen that proliferates within vacuoles of mammalian cells. However, recent in vivo studies have revealed that the vast majority of infected cells contain very few intracellular bacteria (three to four organisms). Salmonella intracellular growth is also limited in cultured dendritic cells and fibroblasts, two cell types abundant in tissues located underneath the intestinal epithelium. Recently, a Salmonella factor previously known for its role as a negative regulator of intracellular growth has been shown to tightly repress certain pathogen functions upon host colonization and to be critical for virulence. The connection between virulence and the negative control of intracellular growth is further sustained by the fact that some attenuated mutants overgrow in non-phagocytic cells located in the intestinal lamina propria. These findings are changing our classical view of Salmonella as a fast growing intracellular pathogen and suggest that this pathogen may trigger responses directed to reduce the growth rate within the infected cell. These responses could play a critical role in modulating the delicate balance between disease and persistence.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Growth of Salmonella within cultured eukaryotic cells
  5. Salmonella virulence and the repression of pathogen functions in the host
  6. Active role of Salmonella in restricting intracellular growth in vivo
  7. Conclusions
  8. Acknowledgements
  9. References

Many successful microbial pathogens have evolved sophisticated strategies to prevent overt-damage to the host and in this way guarantee their transmissibility (Young et al., 2002; Ficht, 2003; Monack et al., 2004a). The characterization of the mechanisms underlying these strategies is fundamental for understanding virulence. One of the most remarkable examples of host–pathogen coexistence is that of Mycobacterium tuberculosis. This intracellular pathogen is carried by about one-third of the human population. Only 5–10% of the infected individuals develop the disease. M. tuberculosis harbours ‘persistence’ genes that are used exclusively to remain latent in host tissues. Strains carrying mutations in these functions overgrow inside macrophages and are attenuated for virulence (Zahrt and Deretic, 2001), which suggests that persistence and disease are intimately related.

Salmonella enterica is another intracellular pathogen prone to cause persistent infections. This bacterial species causes diseases ranging from mild gastroenteritis to life-threatening systemic diseases. Serovars causing these pathologies have been linked to persistent infections in humans and animals. A relevant example is that of the serovar Enteritidis, which causes a self-limited gastroenteritis in humans but establishes a long-lasting asymptomatic colonization upon infection of the chicken caecum (Sadeyen et al., 2004). About ∼5% of humans developing typhoid fever, a systemic disease caused by the human-adapted serovars Typhi and Paratyphi, become chronic carriers (Young et al., 2002).

A host factor that influences the capacity of Salmonella to establish persistent infections is Nramp1 (Slc11a1) (Caron et al., 2002; Beaumont et al., 2003; Monack et al., 2004b). Nramp1 acts as a transporter of divalent cations that mediates natural resistance to infections by several intracellular pathogens, including Salmonella (Blackwell et al., 2001). Of interest, Nramp1–/– mice, which succumb rapidly to serovar Typhimurium infections after developing a systemic disease, are able to clear sublethal doses of the serovar Enteritidis more efficiently than Nramp1+/+ mice (Caron et al., 2002). The serovar Typhimurium has been shown to persist in Nramp1+/+ mice for as long as 1 year (Monack et al., 2004b). These data suggest that although Nramp1 controls Salmonella proliferation during the early phase of the infection, it may also elicit in the pathogen responses that prevent eradication of the bacteria at the late phase of the infection. Consistent with this hypothesis, several recent reports have shown that Nramp1 upregulates Salmonella virulence genes (Zaharik et al., 2002; 2004).

Virulence-related genes shared by all salmonellae include the Salmonella-pathogenicity islands-1 (SPI-1) and -2 (SPI-2), which encode type III protein secretion systems (TTSS) that deliver specific effectors to the host cell (Galán, 2001; Hansen-Wester and Hensel, 2001; Waterman and Holden, 2003). SPI-1 is essential for invasion of non-phagocytic cells while SPI-2 is required for intracellular survival and proliferation. Given the essential role played by these two TTSS in virulence, it is not surprising that both systems had been linked to persistence. The polynucleotide phosphorylase (PNP), an enzyme that controls RNA maturation and degradation, has been proposed to alter the balance between acute infection and persistency states by modulating the expression of SPI-1 and SPI-2 genes (Clements et al., 2002). More recently, Mig-14, a serovar Typhimurium membrane protein regulated by the two-component system PhoP/PhoQ, was also shown to be required for persistence (Brodsky et al., 2005). Mig-14 mediates resistance to the cathelin-related antimicrobial peptide (CRAMP), a molecule produced by activated macrophages. Salmonella persistence might then be programmed by pathogen functions that alter the pattern of expression of distinct virulence traits together with factors required to cope with host defences.

An important distinction between a state of acute infection (disease) and persistence is that only in the former case the bacteria are assumed to undergo massive grow in the host. As Salmonella is an intracellular pathogen, these differences should be manifested at a single cell level. However, recent studies performed in the mouse model have contradicted this notion since a low number of bacteria per infected cell has been observed in both acute and chronic infections (Sheppard et al., 2003; Monack et al., 2004b). Thus, the killing of susceptible Nramp1–/– mice by Salmonella, which exemplifies a classical ‘disease’ state accompanied by a marked increase of the bacterial load in target organs, should be interpreted with caution as it may not be the consequence of exacerbated intracellular growth of the pathogen inside the infected cell. In this review, we address this apparent paradox, revisiting what it is known on the factors controlling the rate at which Salmonella grow within eukaryotic cells.

Growth of Salmonella within cultured eukaryotic cells

  1. Top of page
  2. Summary
  3. Introduction
  4. Growth of Salmonella within cultured eukaryotic cells
  5. Salmonella virulence and the repression of pathogen functions in the host
  6. Active role of Salmonella in restricting intracellular growth in vivo
  7. Conclusions
  8. Acknowledgements
  9. References

Epithelial and macrophage cells

Upon invasion of cultured macrophages and epithelial cells, Salmonella undergoes rapid intracellular growth within a membrane-bound compartment, known as the Salmonella-containing vacuole (SCV). The SCV is a highly dynamic compartment that transiently interacts with the early endosome to further mature as an specialized compartment displaying limited interactions with the endocytic route (Gorvel and Meresse, 2001; Holden, 2002; Knodler and Steele-Mortimer, 2003; Brumell and Grinstein, 2004). Intracellular bacterial proliferation rates are often estimated by gentamicin-protection assays and increases as high as 100-fold in less than 24 h have been reported. Activation of cultured macrophages with interferon (INF)-γ results in a substantial reduction of the intracellular bacterial yield (Brodsky et al., 2005). Bacterial proliferation in epithelial cells and intramacrophage survival are considered as major hallmarks of Salmonella virulence as in vitro-selected mutants defective for these processes are strongly attenuated. The SPI-2-encoded TTSS and the two-component system PhoP-PhoQ play a pivotal role in promoting intracellular growth (Groisman, 2001; Waterman and Holden, 2003; Kuhle and Hensel, 2004). The SPI-1-encoded TTSS, which remains functional in the SCV long after bacterial entry (Drecktrah et al., 2005), also seems to contribute to bacterial growth in epithelial cells (Steele-Mortimer et al., 2002). Whether SPI-1, SPI-2 and PhoP-PhoQ act in a sequential, co-ordinate or independent manner has not been clarified yet. Another important issue is the effect of the host cell type. While a deficiency in SPI-2 results in severe reduction of bacterial growth in macrophages and epithelial cells (Suvarnapunya et al., 2003; Waterman and Holden, 2003; Brodsky et al., 2005), the lack of a functional PhoP-PhoQ system compromises the final bacteria yield only in macrophages (Groisman, 2001). The basis of this difference may be linked to the fact that macrophages are endowed with more robust and varied antimicrobial defences (Rosenberger and Finlay, 2003). These include reactive oxygen/nitrogen intermediates (ROI and RNI), antimicrobial peptides and the Raf/MEK/ERK cascade (Vazquez-Torres and Fang, 2001; Chakravortty et al., 2002; Eriksson et al., 2003; Rosenberger and Finlay, 2003; Rosenberger et al., 2004). SPI-2 and PhoP-PhoQ are absolutely required to withstand these defences.

The onset of Salmonella replication within cultured epithelial cells is accompanied by a pronounced redistribution of endosomes in the infected cell. This alteration ends in the formation of filamentous structures, named Sifs. Although initially thought to be exclusive associated to the epithelial cell infection, Sifs have been observed in IFN-γ-activated macrophages infected with Salmonella (Knodler et al., 2003). This result is intriguing as bacteria growing actively within non-primed macrophages do not induce Sifs. The presence of Sifs correlates with a higher content of intracellular bacteria and specific bacterial products may also downregulate its formation (Birmingham et al., 2005). For unknown reasons, Sifs are not formed in all infected cells, which is in line with the mixed populations of bacteria documented in macrophages (Holden, 2002). It has also been reported that a small percentage of intracellular Salmonella residing within cultured macrophages and epithelial cells reach the cytosol (Perrin et al., 2004 and references therein). These bacteria are recognized by the ubiquitin system, a mechanism thought to contribute to bacterial clearance in the infected cell. Whether these processes affect the proliferation of Salmonella in host tissues is at present unknown.

The fact that distinct pathogen–host interactions occur simultaneously in different cells of the same culture raises important considerations for the interpretation of biochemical and gene-expression data. Thus, it is common to find studies estimating intracellular growth rates exclusively from gentamicin-protection assays, which do not provide information on the fate of intracellular bacteria at the single cell level. This widely used practice, together with lack of uniform criteria in parameters such as the growth state of bacteria (stationary versus late log phase, SPI-1-inducing conditions), opsonized versus non-opsonized bacteria, the multiplicity of infection, the density of cell culture, and the incubation time of cultured cells with bacteria, might explain some of the contradictory data regarding the capacity of distinct Salmonella mutants to grow within cultured cells. The control of these variables is even more critical in the macrophage infection model, given the capacity of Salmonella to induce cell death in this cell type by multiple mechanisms (Hueffer and Galán, 2004). Beside the inherent different physiology linked to a concrete cell type (e.g. macrophage versus epithelial), another variable that complicates interpretation of the data is the different origin (human, rat, mouse) of the cell lines used. In addition, although the exact impact of Nramp1 in the intracellular growth of Salmonella remains to be characterized, there is a tendency to use macrophages freshly isolated from Nramp1–/– mice. The RAW264.7 macrophage cell line, which is often used, also carries a loss-of-function mutation in Nramp1 (D169G). The growth kinetics in organs of Nramp1+/+ and Nramp1–/– mice is distinguishable only a few days after infection (Govoni et al., 1999). Whether or not this difference in bacterial load results from an increase in proliferation at the single cell level has not yet been examined.

Dendritic and fibroblast cells

Unlike the disparity of data found in vitro with macrophage and epithelial cells, the infection of cultured dendritic cells (DCs) and fibroblasts by Salmonella results in a homogeneous population of intracellular bacteria that remain in a non-growing resting state (García-del Portillo et al., 2000; Niedergang et al., 2000; Cano et al., 2001; Jantsch et al., 2003). DCs play a pivotal role in antigen presentation since they, unlike macrophages, are capable of stimulating naive T cells (Niedergang et al., 2004). Unexpectedly, SPI-2 and PhoP-PhoQ were shown to be dispensable for bacterial survival or proliferation within the DCs (García-del Portillo et al., 2000; Niedergang et al., 2000; Dreher et al., 2001; Jantsch et al., 2003). These observations are intriguing considering that the SPI-2 system is induced within DCs (Jantsch et al., 2003) and that the PhoP-PhoQ system is involved in preventing efficient antigen presentation by the infected DC (Svensson et al., 2000). Thus, it seems that the control exerted by Salmonella on the DC is primarily focused in preventing antigen presentation and the attenuation of the intracellular growth rate must facilitate such manipulation of the host cell.

Despite of the ubiquity of fibroblasts in all tissues, these cells have been rarely used in cell culture models to analyse Salmonella–host interactions. A probable reason for this apparent lack of interest in fibroblasts is that these cells are supposed not to be targeted by Salmonella during the course of the infection. However, no investigation has formally discarded this possibility. Of interest, cultured fibroblasts have been very useful for unravelling a novel Salmonella intracellular response dedicated to reduce the intracellular growth rate within the infected cell (Cano et al., 2001). As in DCs, the infection of fibroblasts with virulent strains results in a homogeneous population of intracellular bacteria, with an average of two to three organisms per cell. The infected fibroblasts retain integrity and viability for long periods of time, up to 3 weeks after bacterial entry (Cano et al., 2003), denoting a clear distinction with the macrophage model. Most of the intracellular bacteria recovered at these times display a ‘small-colony’ phenotype, defects in respiratory metabolism, increased intracellular persistence rates and are strongly attenuated for virulence (Cano et al., 2003). These results consistently support the idea that the establishment of persistence and the progression to disease are interconnected processes.

Another striking feature of the fibroblast infection model is that, when compared with epithelial or macrophages, the bacterial functions that control intracellular growth seem to act in an opposite manner. The inactivation of the PhoP-PhoQ system, which results in reduced intramacrophage survival, causes an abrupt increase (> 10-fold) in the intracellular bacterial yield in fibroblasts (Cano et al., 2001). This is the only example in which the PhoP-PhoQ system has been shown to be involved in reducing the intracellular bacterial load. Other functions playing a similar role include the regulators SvpR and SlyA, the alternate sigma factor RpoS, and a novel protein named ‘intracellular-growth attenuator-A’ (IgaA) (Cano et al., 2001). IgaA controls negatively the activity of the RcsC-YojN-RcsB regulatory system (Cano et al., 2002), suggesting that the repression of the RcsC-YojN-RcsB system may be a requisite for establishing a low growth rate within the host cell. The SPI-2 TTSS also plays a relevant role in fibroblasts as abrogation of its activity leads to pronounced killing of the intracellular bacteria (Cano et al., 2001). A key aspect that remains unexplored is the timing at which these distinct pathogen functions act during the long-standing residence of bacteria within the fibroblast.

Salmonella-directed responses that downregulate intracellular growth in non-fibroblast cells

The exact significance for virulence of the response uncovered in the Salmonella fibroblast model is at present unknown. However, several studies suggest that this response, probably mounted to stimulate ‘on purpose’ host cell defences, might also operate in other cell types. Eriksson et al. (2000) identified mutants that overgrow within the macrophage due to their incapacity to stimulate the host nitric-oxide synthase (iNOS). Interestingly, most of these mutants are attenuated for virulence. Pharmacological inhibition of iNOS has also been shown to increase the number of intracellular bacteria in DCs (Eriksson et al., 2003). A more recent study has shown that the SPI-1 effector SopB stimulates the production of nitric oxide in epithelial cells (Drecktrah et al., 2005). Noteworthy, the same effector is also responsible for stimulating the phospho-inositide-3-kinase (PI3K)/Akt pro-survival route, thus preventing the initiation of apoptotic cascades (Knodler et al., 2005). Are these defence/survival mechanisms finely orchestrated by intracellular Salmonella to reduce its growth rate and ensure the integrity of the new niche colonized? In support to this line of thinking, Roy et al. have recently demonstrated in epithelial and fibroblast cells that the activity of a bacterial ‘weapon’, the SPI-1 TTSS, signals the infected cell to compromise bacterial growth (Roy et al., 2004). Perforation of the host cell membranes by the SPI-1 TTSS leads to changes in intracellular Ca2+ levels, which are sensed by synaptogamin (Syt)-VII to increase the fusion rate of the SCV with lysosomes. This model predicts that mutants unable to assemble the SPI-1 TTSS translocon, such as invA mutant, should overgrow within the SCV. Such a phenotype has been observed in cultured fibroblasts (M. Elias-García and F. García-del Portillo, unpubl. obs.). However, a lower proliferation rate for the same invA mutant has been reported in epithelial cells (Steele-Mortimer et al., 2002). These differences need to be clarified by further analysis of the exact mode of entry of the bacteria in both cell types. The possibility that the SPI-2-encoded TTSS may also perforate the phagosomal membrane and activate host functions affecting the bacterial growth has not yet been investigated.

Salmonella virulence and the repression of pathogen functions in the host

  1. Top of page
  2. Summary
  3. Introduction
  4. Growth of Salmonella within cultured eukaryotic cells
  5. Salmonella virulence and the repression of pathogen functions in the host
  6. Active role of Salmonella in restricting intracellular growth in vivo
  7. Conclusions
  8. Acknowledgements
  9. References

Virulence regulators tightly co-ordinate expression of virulence genes for promoting pathogen proliferation upon host colonization. One of the most intensively studied regulators is the Salmonella PhoP-PhoQ two-component system, which is induced in intracellular bacteria (Groisman, 2001). A recent study proposed that a gene regulated positively by PhoP-PhoQ, pcgL, is involved in stimulating the host immune system to prevent bacterial overgrowth in mouse organs (Mouslim et al., 2002). This observation is intriguing when considering that a defect in the PhoP-PhoQ system results in bacterial overgrowth within fibroblasts. In fact, pcgL mutants also overgrow upon infection of cultured fibroblasts (F. García-del Portillo, unpubl. obs.). As in the case of Nramp1, whether the pcgL mutation affects the growth rate of Salmonella within individual host cells upon colonization of animal tissues remains unknown. Nonetheless, these observations reinforce the hypothesis that Salmonella may intentionally limit the proliferation in host tissues by stimulating host defences.

More recently, it has been demonstrated that the Salmonella IgaA protein tightly represses the RcsC-YojN-RcsB system during host colonization (Domínguez-Bernal et al., 2004). Any failure in the silencing of this phosphorelay causes a strong attenuation in virulence (Domínguez-Bernal et al., 2004; Mouslim et al., 2004; Garcia-Calderon et al., 2005). Thus, the same regulatory circuit involved in restricting intracellular growth within cultured fibroblasts was shown to be critical for the progression of the infection. Why mutants that overgrow within cultured fibroblasts are attenuated when tested in animal models is at present unknown. It is possible that massive intracellular growth results in destruction of a privileged niche, the non-phagocytic cell itself, leading to exposure of the pathogen to bystander phagocytes. These phagocytes might have even been ‘primed’ by the cell initially infected. In the case of overgrowing mutants highly susceptible to activated phagocytic cells, such as phoP- or phoQ-defective mutants, the loss of the non-phagocytic cell niche would constitute their ‘suicide’ and inability to further progress in the infection cycle. Under this condition, the host immune system could be also stimulated by the spread of bacterial products derived from the overgrowing bacteria. The release to the extracellular milieu of other type of overgrowing mutants, such as the igaA, might have a different outcome as this specific strain is not defective for intramacrophage survival. This postulate agrees with the observation that, upon an oral challenge, igaA mutants colonize systemic sites of mice such as spleen and liver although they are unable to kill the animals.

Active role of Salmonella in restricting intracellular growth in vivo

  1. Top of page
  2. Summary
  3. Introduction
  4. Growth of Salmonella within cultured eukaryotic cells
  5. Salmonella virulence and the repression of pathogen functions in the host
  6. Active role of Salmonella in restricting intracellular growth in vivo
  7. Conclusions
  8. Acknowledgements
  9. References

Despite the existing evidence for a Salmonella-directed response that restrains growth within certain cultured cells, no study has still addressed whether this phenomenon also occurs in host tissues. Salmonella colonizes predominantly macrophages in the liver and CD18+ cells (macrophages and/or DCs) in the blood (Richter-Dahlfors et al., 1997; Vazquez-Torres and Fang, 2000). Phagocytes also account for the majority of cells containing intracellular bacteria in the spleen (Salcedo et al., 2001). This latter study provided a detailed estimation of the number of intracellular bacteria present in the infected cells. About 25% of these macrophages contained 1 single bacterium and similar percentages were estimated for populations containing 1–5, 5–10 or > 10 intracellular bacteria respectively. A defect in SPI-2 resulted in a bias to the macrophage population carrying one single bacterium (approximately 80% of total).

It is worth noting that some studies have based the identification of Salmonella-infected cells in vivo in the analysis of markers that do not distinguish between different types of phagocytes as macrophages and DCs. An example is CD18, which besides these phagocytes is also present in polymorphonuclear neutrophils and lymphocytes. F4/80, initially thought to be macrophage-specific, is also a marker detectable in immature DCs. More specific markers, such as CD11c (DCs) or MOMA-2 (macrophages), should be widely used to differentiate the phagocyte populations targeted by Salmonella.

Infection of neutrophils, macrophages and DCs by Salmonella has also been documented in spleen and mesenteric lymph nodes (MLN) (Yrlid et al., 2001; Cheminay et al., 2004). Bystander DCs have been shown to present antigens derived from neighbouring Salmonella-infected macrophages that undergo apoptosis (Sundquist et al., 2004; Winau et al., 2004). No evidence has been found in vivo for DCs containing actively growing bacteria. Similarly to the situation in the spleen, a low number of intracellular bacteria has been reported in the liver. After four days post-infection, more than 90% of the infected cells (CD18+ phagocytes) contained ≤ 3 bacteria (Sheppard et al., 2003). Interestingly, the bacterial load increases by 3-logs during this time. This apparent contradiction was resolved by the demonstration that Salmonella expands its population in the liver by increasing the number of infection foci rather than undergoing massive intracellular growth in individual host cells. Bacteria spreading from the initial infection foci to nearby phagocytes may be facilitated by apoptosis of the infected cells (Richter-Dahlfors et al., 1997; Monack et al., 2000; Mastroeni and Sheppard, 2004). Preferential targeting of macrophages by Salmonella has also been shown in the chronic infection model optimized with low bacterial doses and resistant Nramp1+/+ mice. One year after infection most of the bacteria (85% of infected cells) are visualized in macrophages (MOMA-2-positive cells) located in the MLN. These macrophages also carry an average of three to four bacteria (Monack et al., 2004b). The identity of the remaining MOMA-2-infected cells was not reported. Taken together, these evidences indicate that massive Salmonella intracellular growth is not a hallmark of the virulence of this pathogen, neither in systemic disease nor in persistence. Moreover, considering that low numbers of intracellular Salmonella are observed in susceptible and resistant mice, it is tempting to postulate that Nramp1 could play a pivotal role in diverting the infection to persistence (Nramp1+/+ mice) or disease (Nramp1–/– mice) by a mechanism unrelated to the control of pathogen intracellular growth. Attempts to unravel this putative mechanism might certainly add in gaining new insights on how Salmonella establishes a persistent infection.

Unlike the preference shown by Salmonella to target macrophages in acute and chronic infections, the initial stages of the infection of the intestinal epithelium are apparently driven by different bacteria–host interactions. Immature DCs having high phagocytic capacity are the first cells that internalize the bacteria upon passage of the M cells located in the Peyer's patches (Hopkins et al., 2000). DC maturation stimulated by the pathogen may facilitate the migration of the infected cells to T cell areas of secondary lymph organs (Niedergang et al., 2004). Uptake of bacteria present in the intestinal lumen is also carried out efficiently by DCs located in the lamina propria (Niedergang et al., 2004; Niess et al., 2005). This uptake is directed by the DCs, which form extensions (dendrites) between the epithelial cells to capture the luminal bacteria in a SPI-1-indepedent manner (Vazquez-Torres and Fang, 2000). Expression of the chemokine receptor CX3CR1 by the lamina propria DCs is required for efficient bacterial capture (Niess et al., 2005). Mice lacking CX3CR1 are more susceptible to Salmonella infection, suggesting that the rapid uptake of bacteria taking place in the lamina propria is important for eliciting an effective immune response.

The lamina propria is also filled with distinct subpopulations of fibroblast cells (Adegboyega et al., 2002). Fibroblasts have also been shown to be pivotal cells for the modulation of immune responses (Buckley et al., 2001). Unlike phagocytes and lymphocytes, no marker exclusive for all types of fibroblasts is currently known. Differential staining of alpha-smooth-muscle actin (α-SMA), vimentin, desmin and the proto-oncogene c-Kit has allowed to differentiate certain fibroblast subtypes such as myofibroblasts and interstitial cells of Cajal (ICC) (Powell et al., 1999). However, another important population of fibroblasts, known as interstitial fibroblasts located in the lamina propria, are classified as vimentin+, c-Kit, α-SMA, desmin (Powell et al., 1999). This marker profile is shared with phagocytic cells, which forces the investigator to infer the presence of this specific fibroblast subtype by the ‘lack’ of phagocyte-specific markers. This caveat certainly hampers the unequivocal detection of some fibroblasts subtypes in infected tissues and may explain why this cell type is often unexplored in bacterial infections of the intestine.

Recently, it has been claimed that Salmonella does not target fibroblasts upon penetration of the intestinal barrier (Santos and Baumler, 2004). However, this conclusion was based exclusively in morphological criteria without providing a detailed analysis of markers specific of phagocytic cells. So, the possibility that Salmonella may target fibroblasts in vivo cannot be formally discarded. In fact, phoP-defective mutants have been observed overgrowing in a subpopulation of CD18-negative cells located in the lamina propria(Fig. 1). This phenotype is reminiscent of what is seen in fibroblast cultures and raises the novel concept of Salmonella as a highly specialized pathogen that actively restrains the intracellular growth rate shortly after passage of the epithelial barrier. A more complete set of markers will be required to discern the exact identity of the non-phagocytic cell in which Salmonella reduces the intracellular growth rate. Primary cultures obtained from intestinal tissue could also be a valuable material to address this hypothesis. In addition, the model of bacterial infection of human intestinal explants may also offer new information on the Salmonella cell tropism occurring at the level of the intestine (Haque et al., 2004). A relevant feature of this model, also known as in vitro organ culture (IVOC), is that it provides a more physiological and heterogeneous environment than immortalized cell cultures.

image

Figure 1. Salmonella enterica serovar Typhimurium restricts the intracellular growth rate in CD18-negative cells located in the lamina propria of the mouse ileum. BALB/c mice were infected orally with wild type (A and C) or the isogenic phoP-defective mutant (B and D). The tissue was processed for confocal laser immunofluorescence microscopy using antibodies against bacterial lipopolysaccharide (green) and CD18 (red). Micrographs were taken 6 h after oral administration of bacteria. Nuclei in (B) and (D) were stained with the cyanine fluorochrome TO-PRO3. (C) and (D) are enlarged images of the infected cells shown within boxes in images (A) and (B) respectively.

Download figure to PowerPoint

Conclusions

  1. Top of page
  2. Summary
  3. Introduction
  4. Growth of Salmonella within cultured eukaryotic cells
  5. Salmonella virulence and the repression of pathogen functions in the host
  6. Active role of Salmonella in restricting intracellular growth in vivo
  7. Conclusions
  8. Acknowledgements
  9. References

Besides mycobacteria, other non-commensal bacteria such as Chlamydia spp., Brucella spp., Helicobacter pylori and S. enterica are among the pathogens more frequently linked to persistent infections (Young et al., 2002; Ficht, 2003; Monack et al., 2004a). The current knowledge of the mechanisms used by these pathogens to establish a persistence state is still very limited. These pathogens may have evolved with sophisticated strategies to finely orchestrate the defences in the infected cell to reach a perdurable and low intracellular growth rate. The recent findings collected in animal tissues suggest that intracellular Salmonella may not experience more than two to three doublings along its lifetime within the infected cell. Cells harbouring more than 10 bacteria represent a very minor percentage of the infected cells, even in conditions of an acute lethal infection. The fact that mutations in a subset of pathogen functions result in bacterial overgrowth in macrophages or fibroblasts reinforces the concept of Salmonella as a pathogen that preferentially remains in a state of low proliferation within the host cell (Fig. 2). The bacteria could regulate this process from the very early stages of the infection upon perforating the host cell membranes. The occurrence at later times of events such as apoptosis and/or stimulation of distinct host cell defences may result in different modes of intracellular growth arrest. The data accumulated with specific host factors as Nramp1 and the virulence attenuation of overgrowing mutants suggest that disease and persistence might be intimately regulated. The data collected in the fibroblast infection model also raise the question of whether Salmonella persistence is established exclusively in macrophages. Unlike differentiated cells of the immune system and cells of epithelia, fibroblasts are not subjected to rapid turnover. Is it then possible that Salmonella targets this cell type for ensuring a ‘persistence’ niche shortly after the penetration of the intestinal barrier? This exciting hypothesis may certainly attract future studies focused to unravel how Salmonella has evolved to reduce the growth rate within the infected cell.

image

Figure 2. Salmonella–host interactions reported in the intestinal tissue, mesenteric lymph nodes and target organs (spleen, liver). Intracellular bacterial proliferation occurs at a certain extent only in splenic phagocytes (macrophages and probably DCs). MLN, mesenteric lymph node; DC, dendritic cell; Mø, macrophage; PMN, polymorphonuclear neutrophil. See text for details.

Download figure to PowerPoint

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Growth of Salmonella within cultured eukaryotic cells
  5. Salmonella virulence and the repression of pathogen functions in the host
  6. Active role of Salmonella in restricting intracellular growth in vivo
  7. Conclusions
  8. Acknowledgements
  9. References

We apologize to those colleagues whose work has not been cited due to space limitations. A.T. is recipient of a predoctoral fellowship of the ‘Consejería de Educación de la Comunidad de Madrid’. Work in the laboratory is supported by Grants BIO2004-03455-C02-01 and GEN2003-20234-C06-01 from the Spanish Ministry of Education and Science.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Growth of Salmonella within cultured eukaryotic cells
  5. Salmonella virulence and the repression of pathogen functions in the host
  6. Active role of Salmonella in restricting intracellular growth in vivo
  7. Conclusions
  8. Acknowledgements
  9. References