Published online before print April 7, 2006
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Laboratory of Microbiology and Immunology of Infection, IBMC-Institute for Molecular and Cell Biology and ICBAS-Instituto de Ciências Biomédicas de Abel Salazar, University of Porto, Portugal
1Correspondence: IBMC, Rua do Campo Alegre 823, 4150-180 Porto, Portugal. E-mail: rappelb{at}ibmc.up.pt
|
|
|---|
Key Words: nutrient phagosome iron intracellular parasites immunity
|
|
|---|
The lifestyles of intracellular pathogens vary considerably. Some appear to block vacuole maturation, arresting it at a certain stage (e.g., mycobacteria), some allow full maturation (e.g., Coxiella and certain Leishmania), and others appear to "construct" their own replicative organelle (e.g., Legionella, Brucella, and Salmonella). Mycobacteria and Salmonellae prevent the fusion of the pathogen-containing vacuole with the lysosome. Whereas the former remain in a compartment with early/late endosomal characteristics and continuously interact with endosomes [12 13 14 ], the latter only transiently inhabit such a compartment (containing Rab5 and early endosomal antigen 1), moving onto a more modified, microtubule-asssociated vesicular system, the Salmonella-induced filaments, which acidify through the action of vacuolar adenosinetriphosphatases (ATPases) and contain Rab7, its effector Rab-interacting lysosomal protein, and lysosomal-associated membrane protein-1 (LAMP-1) [15 ]. In contrast, parasitophorous vacuoles containing certain species of Leishmania mature into lysosomal-like compartments rich in cathepsins and contain Rab7, vacuolar ATPases, LAMP-1 and -2, and the mannose-6-phosphate receptor [16 , 17 ]. These parasitophorous vacuoles are extremely fusogenic, acquiring material obtained by fluid-phase endocytosis [16 , 18 ], from autophagic vacuoles [17 ], and even large structures such as phagocytosed zymosan [19 , 20 ]. In contrast, the L. pneumophila phagosome has been shown to interact with the endoplasmic reticulum (ER) [21 ], but as replication of the bacteria occurs, replication vacuoles become acidic and begin interacting with the lysosomal compartment [22 ]. Clearly distinct is the vacuole occupied by Toxoplasma gondii: It does not interact with endosomes or lysosomes, does not express the vacuolar ATPases, and has unclear associations with the ER and mitochondria, likely mediated by the parasite rhoptry protein (ROP) 2 [23 , 24 ]. The Brucella vacuole, conversely, avoids fusion with the lysosome and interacts with, or at least acquires markers from, the ER [25 , 26 ], whereas the Chlamydia replication compartment, named inclusion, does not fuse with lysosomes and resides close to the Golgi apparatus and mitochondria [23 ]. Some pathogens use a more dramatic approach of lysing the vacuole membrane and invading the cytoplasm. Examples of the latter include L. monocytogenes [27 ], T. cruzi [28 ], and Shigellae [29 ].
Much of the information regarding how a microbe accesses nutrients inside the macrophage is still lacking: What are those nutrients? How are they obtained inside the host cell? Do the subversion tactics used by intracellular microbes, namely those that affect the normal trafficking of an ingested particle, play a role in the acquisition of nutrients? Conversely, the antimicrobial machinery of the phagocytes leading to restricted access to nutrients (nutriprive mechanisms) is still unclear: Does the macrophage limit access to nutrients at all? What are the mechanisms involved? Are they constitutive, or can they be induced upon activation by the immune system? Does the macrophage play an active role in sabotaging the subversion strategies of the ingested microbe, namely after activation?
Here, we will analyze some of these questions. We will define nutrient in a relatively broad senseany molecule acting as a source of energy and reducing power as a source of basic elements (carbon, sulfur, nitrogen, iron, and others) or as a required growth factor, such as any organic molecule the microbe is unable to synthesize, namely vitamins, amino acids, or nucleotide bases. Electron acceptor molecules (in most instances, oxygen) and physicochemical properties such as the pH will also affect microbial growth and may be manipulated by the phagocyte or the structures they form in vivo (e.g., the granulomas). As we need to understand which nutrients are accessed by proliferating intracellular pathogens to evaluate whether they are limiting in a situation of restricted microbial growth, the first part of this article will review what is known about the nutritive potential of the macrophage habitat.
|
|
|---|
Direct analysis
Wagner et al. [31
] used the hard X-ray microprobes to detect X-ray fluorescence and to directly estimate the presence of 10 elements inside phagosomes containing mycobacteria. They paid particular attention to iron and showed an enrichment in vacuoles with time as well as a decrease following activation with interferon-
(IFN-
) or tumor necrosis factor (TNF). Furthermore, they showed that a mutant deficient in the synthesis of the siderophore, mycobactin, was defective in its ability to accumulate iron, showing that iron was supplied to the mycobacterial phagosome as it was used by the microbe, as expected. The study of iron is also amenable to analysis by ultrastructural autoradiography [31
] or by scintillation counting on isolated bacterial cells [32
] following the addition of 59Fe to infected cells. The presence of a labile iron pool was also detected in the cytoplasm using an iron-sensitive fluorescent probe, calcein [33
]. The pH in vesicular compartments can be determined by the accumulation of certain weak bases upon protonation in the low pH compartments followed by immunochemical or fluorescence detection of the amount of probe concentrated in such cellular compartments. This technique was used to determine the lack of acidification in mycobacterial vacuoles [34
]. The pH can also be assessed in vacuoles by using fluorescent probes with a pH-sensitive emission spectrum coupled to the microbial cells, and in this way, reduced acidification of mycobacterial phagosomes has been confirmed [35
].
Indirect indications of the presence of nutrients
One indirect way to look at the availability of nutrients is to study gene expression signatures. Another is to analyze the natural requirement for growth factors (natural auxotrophy) or to compare the growth of auxotrophic mutants with the growth of wild-type microorganisms.
Study of auxotrophy
Salmonella does not assimilate folate. Instead, it synthesizes it from para-amino-benzoic acid (PABA). As the host cannot provide PABA, the microbe will not synthesize folate if the synthesis of PABA is inhibited, hence the efficacy of sulfonamides in chemotherapy. In addition, Salmonella requires 2,3-dihydroxybenzoate (DHB) for enterochelin synthesis, and again, this precursor cannot be provided by the host. Chorismate is a precursor for PABA and DHB. Hoiseth and Stocker [36
] produced a Salmonella mutant (aroA), which was unable to synthesize chorismate. This model system is thus the ideal situation to analyze how the lack of a nutrient (in this case, two, none of which can be supplied by the host and thus represent a situation of nutrient deprivation) will affect growth of a pathogen inside its host. These authors reported an increase in 50% lethal dose (LD50) from less than 20 bacteria for the wild-type strain to more than 3 x 105 bacteria for the aro strain. Therefore, we can postulate that limitation of a nutrient in a host would make an auxotroph for that nutrient to increase its LD50 to such an extent and thus set the read-out for the assessment of the presence of a specific nutrient or the existence of putative, nutriprive mechanisms in the host in the context of salmonellosis. It should be kept in mind, however, that other explanations may underlie an increase in LD50 of a particular mutant and that even intracellular pathogens may at some point be extracellular and thus have different nutrient requirements. Table 1
summarizes the impact of auxotrophy on the ability of intracellular pathogens to proliferate in vivo or in cultured macrophages.
|
View this table: [in a new window] |
Table 1. The Effect of Auxotrophy on the Ability of Pathogens to Proliferate in vivo or in Cultured Macrophages
|
M. tuberculosis, albeit its slow replication rate, has no known growth factor requirements [52 ]. Mycobacteria auxotrophs have also been characterized, and the data are similar to those obtained with Salmonella, inasmuch as a lack of purines and certain amino acids appears to make the M. tuberculosis vacuole inhospitable to auxothophs, such as the purC [41 , 42 ], the leuD [43 ], and the lysA [44 ]. However, when proline and tryptophan auxotrophs were compared [53 ], the latter appeared to be less able to survive than the former, suggesting that the deficiency in certain amino acids is more severe than others or that the mutations may in some way affect the fitness of the mycobacteria to survive in the macrophage. In the case of mycobacteria, it is not clear whether the auxotrophy leads to an increase in susceptibility to antimicrobial (e.g., oxidative) mechanisms or to stress or if an inability to run salvage pathways for the required nutrient does not compensate for its presence in the vacuole. Similar caution must be used in interpreting the low virulence of panthotenate auxotrophs [45 ]. Diminished amounts of panthotenic acid may lead to defects in fatty acid metabolism and polyketide synthesis with possible deficiencies at the level of resistance to oxidative damage and reduced synthesis of virulence factors. When comparing the growth characteristics of a leucine BCG auxotroph, Jacobs and colleagues [46 , 47 ] found that the lack of in vivo virulence exhibited by the mutant was related to its inability to proliferate inside a macrophage cell line and thus, concluded that leucine was not available to the mycobacteria inside the host cell. It is thus safe to speculate that the concentration of amino acids, purines, and vitamins is likely to be reduced severely in the mycobacterial vacuole.
L. pneumophila is quite demanding in regards to its nutritional requirements. However, it was shown that tryptophan auxotrophs are still able to proliferate in human monocytes, suggesting no limitation of this nutrient in the pathogen vacuole [48 ]. In contrast, thymidine auxotrophs did not grow in monocytes [48 ], supporting the notion that phagosomes lack this nucleotide. Likewise, B. melitensis mutants, deficient in purine biosynthesis, exhibit reduced virulence in mice [49 ]. An extensive analysis of Brucella suis mutants unveiled 15 amino acid biosynthetic genes and five purine/pyrimidine biosynthetic genes whose mutation would lead to attenuation of the pathogen in cultured macrophages, suggesting that the Brucella-containing vacuole is poor in such nutrients [54 ]. It should be noted that the Legionella and the Brucella vacuoles possess markers of the ER, and yet, only the latter appears to be limiting in amino acids, showing that they are indeed distinct compartments.
In contrast, the cytoplasm probably presents no such restrictions, and even naturally auxotrophic L. monocytogenes can proliferate in this locale. In fact, this bacterium was shown to be quite demanding in regards to its growth-factor requirements: Seven amino acids and the vitamins riboflavin, thiamine, biotin, and thioctic acid are required for in vitro growth, and it is quite dependent on glucose (or some substituting sugar) for proliferation [55 ]. If on top of this natural auxotrophy, auxotrophic mutants are induced for additional amino acids, adenine, uracil, or niacin, listeriae are still able to proliferate intracellularly [50 ], suggesting no restriction of all these nutrients in the cytoplasm of macrophages. As already mentioned, the cytoplasm also has iron available as a labile iron pool [33 ]. It is not surprising that when made unable to egress the vacuole, L. monocytogenes loses its ability to replicate [56 , 57 ], probably as it no longer obtains the required nutrients.
Toxoplasma requires cholesterol [58 , 59 ] and purines [60 ] to proliferate. It is thus fair to suppose that the isolated vacuole it inhabits has access to these two groups of molecules. Cholesterol is obtained from low-density lipoproteins obtained through receptor-mediated endocytosis and recycled through the lysosomal compartment [58 ], whereas purines may reach the parasitophorous vacuole (PV) from the cytosol through functional pores in the PV membrane [61 ]. Similarly, T. cruzi, a cytosol inhabitant, and Leishmania, an inhabitant of an almost mature phagosome, also require purines [62 ] and must obtain them from these locations. Whereas no restrictions are apparent in a cytosolic location, the intravacuolar Leishmania parasites should resort to some strategy to obtain such metabolites. It has been reported that these parasitophorous vacuoles are extremely fusogenic and thus prone to obtain cargo from many intracellular vesicular compartments [16 , 18 19 20 ], as they also contain an organic anion transporter, which pumps small anions from the cytoplasm [17 ].
The data discussed above strongly suggest that intracellular pathogens encounter macrophage niches, which can supply selected nutrients, and as might be expected from the variety of interactions these microbes establish with their host cell, this supply of nutrients varies according to the pathogen under study. These studies do not always allow us to tell whether certain nutrients are available to the pathogen, as they subverted the function of the phagocyte, or whether they are not restricted at all, as they would also not be to nonpathogenic microbes. Additional studies enlarge this picture as will be discussed next.
Study of (other) metabolic mutants
Following the identification of the enzyme isocitrate lyase, an enzyme involved in the anaplerotic pathway known as the glyoxylate cycle, as a major protein produced during mycobacterial infection of macrophages [63
], it was shown that this enzyme is required for the virulence of M. tuberculosis [64
]. Today, it has also been shown to be essential for the replication of Salmonella [65
] and fungi such as Candida albicans [66
] inside their hosts. The current interpretation of these mutants states that the lack of a functional glyoxylate cycle impedes the survival on a one- or two-carbon source and therefore, that the vacuole inhabited by the microbes only has access to these carbon sources, namely fatty acids. In the case of mycobacteria and salmonellae, this was shown to occur late in infection in a chronic phase where granulomas are expected to be fully organized. It should be stressed, however, that other interpretations are possible as well to explain the phenotype of the isocitrate lyase mutants. This enzyme is induced in anaerobic conditions and may participate in recycling reduced nicotinamide adenine dinucleotide (NADH) into NAD+ with production of glycine from the reductive amination of glyoxylate by glycine dehydrogenase [67
].
Mutations affecting iron acquisition were shown to reduce virulence of different intracellular pathogens: L. pneumophila mutants with defective iron acquisition or assimilation were impaired in their ability to proliferate in a macrophage cell line [68 ], and mutations in M. tuberculosis leading to disruption of the mycobactin synthesis [69 ] or to overexpression of an iron-dependent repressor [70 ] reduced considerably the proliferative capacity of this microbe in vivo. Thus, although iron is available inside vacuoles inhabited by intracellular pathogens, it is likely present in suboptimal concentrations or must be removed from host chelating molecules such that these pathogens rely on efficient iron acquisition systems for proliferation.
S. typhimurium mutants deficient in the MgtCB genes are unable to proliferate in macrophages, a defect that can be mimicked in axenic cultures by inoculating these mutants in low Mg2+ media [71 ]. Similarly, the analysis of M. tuberculosis MgtC mutants has led to the suggestion that the vacuole harboring this bacterium is mildly acidic and with limiting levels of magnesium [72 ]. Recently, it was also shown that MgtC is required for intra-macrophage growth of B. suis [73 ], illustrating the possibility that Mg2+ is in general limiting inside macrophage vacuolar compartments.
The reduction in the ability to proliferate in mice of a narG mutant of M. bovis BCG suggests that mycobacteria, residing in granuloma macrophages, use nitrate respiration and therefore, that oxygen is limiting in these cells [74 ].
The mutation of the gene encoding an amino acid transporter PhtA in L. pneumophila has allowed the identification of threonine as an essential nutrient acquired by this pathogen inside macrophages [75 ]. Conversely, the blocking of the function or the gene silencing of a neutral amino acid transporter SLC1A5 in macrophages was shown to severely reduce the proliferation of L. pneumophila [76 ]. The expression of this transporter is highly up-regulated upon infection of the host cell [76 ]. Thus, it appears that this pathogen, relying on amino acids as growth-supporting nutrients, recruits host cell machinery to satisfy its nutritive requirements.
Expression profiling
Many genetic techniques have been designed to determine the genes that are regulated upon phagocytosis, namely in vivo expression technologies (IVET) and signature-tagged mutagenesis (STM). For example, using an IVET approach, it was shown that upon infection, S. typhimurium faces an environment poor in iron, magnesium, and copper [77
]. However, with ways to analyze the expression of hundreds of genes through the use of high-throughput analysis technologies, namely microarrays, it is becoming possible to look for signatures in expression related to particular growth conditions. One can, for example, study the up- or down-regulation in expression of many different genes in response to hypoxia, nutrient deprivation, oxidative stress, and other cultural conditions and then compare the changes observed with those that occur once the microbe is ingested by macrophages in culture or even when they infected the host. Several studies have thus defined a series of regulons and compared them with the expression profile induced in macrophages, cultured in vitro or from in vivo experiments. One such study using Salmonella observed changes in the expression of 919 genes (approximately one-fifth of the genome-coding capacity), and changes suggested that the Salmonella-containing vacuole in macrophages cultured in vitro is poor in phosphate (given the strong up-regulation of the phosphate-starvation loci) and magnesium (there was an increase in expression of the genes mgtBC) and is acidic (some of the genes involved in the acid-tolerance response are indeed induced) [78
]. Curiously, data in that report suggest no limitation in amino acids (namely the aromatic ones), iron [although a macrophage with a defective natural resistance-associated macrophage protein 1 (Nramp1)/Slc11a1 transporter was used in the study], or oxygen (albeit the oxygen pressure in the culture medium was likely not the one found in the tissues of the animal). Potassium levels were likely high, and data suggested that gluconate was the carbon and energy source available to vacuolar salmonellae. The differences between this study and the STM analysis may highlight inherent limitations in using a particular method (e.g., low sensitivity regarding the levels of gene expression taking place, existence of redundant or alternative pathways, competition or complementation in the heterogeneous populations analyzed in the STM assays).
Schnappinger et al. [79
] looked at the gene expression of M. tuberculosis and found 601 genes differentially regulated in intraphagosomal microorganisms, accounting for approximately 15% of the mycobacterial genome. It is interesting that a selected number of the genes, whose expression was up-regulated by the phagosomal environment, were also induced in vivo in the mycobacteria that infected the mouse lung. The genes that were induced upon phagosomal residence included those involved in fatty acid catabolism, the glyoxylate cycle, and mycobactin synthesis, suggesting that the mycobacterial phagosome is poor in iron and that mycobacteria rely on lipids for a source of carbon and energy. When they compared the phagosome-induced response with other regulons, they found that many genes were co-regulated in those in vitro conditions and in the intraphagosomal environment. Specifically as to what concerns this analysis, 35 genes were induced in IFN-
-activated macrophages or in broth culture by exposure to low iron concentrations. It is interesting that the magnitude of induction of these genes was higher in macrophages that were able to produce NO than in inducible NO synthase (iNOS)/ cells, suggesting that the response could be partly related to NO damage and not to iron deprivation or alternatively, that NO exacerbates iron deprivation. Many hypoxia-induced genes were also induced inside the macrophage in M. tuberculosis. Finally, there was also a series of genes that was up-regulated by phagosomal residence and nutrient starvation, and these were clearly independent of the presence of NO in the cell. Much of this picture can apply to in vivo-growing M. tuberculosis. Timm et al. [80
] used a quantitative, real-time polymerase chain reaction approach to analyze gene expression in tubercle bacilli isolated from infected mice or human patients. Again, up-regulation of genes responsive to iron deprivation, glucose starvation, hypoxia, and fatty acid metabolism was found in mycobacteria extracted from mouse-infected lungs. The picture for samples taken from human patients was not so clear-cut, and there was less evidence of iron deficiency or use of the glyoxylate cycle. A different approach using a promoter trap strategy confirmed the use of fatty acids in vivo as a carbon source by M. tuberculosis infecting the lungs of mice, as well as the likely iron-restrictive environment faced by the mycobacterium in vivo [81
].
Despite the generation of extremely valuable and extensive information, the regulons defined by microarray techniques for different stimuli overlap. For example, of the 36 genes induced by reactive nitrogen intermediates, 27 were also induced by hypoxia [82 ]. Thus, Voskuil et al. [83 ] have defined a "dormancy regulon", which can be induced by low NO concentrations or by hypoxia and whose expression can be detected in vivo in M. tuberculosis extracted from the lungs of infected mice. High-throughput analysis of gene expression in many model systems will continue to supply extensive amounts of raw data, which will be useful in addressing the questions discussed in this paper.
Manipulation of the intracellular habitat
The manipulation of the intracellular trafficking of certain microbes has shown that the change in the habitat affects the survival or proliferative capacity of the microbe. Thus, SifA mutants of Salmonella, forced into the cytoplasm of macrophages, fail to proliferate [84
]. Curiously, that is not the case if the SifA mutants infect epithelial cells where the cytoplasm can support bacterial growth [85
]. In contrast, when the saprophyte Bacillus subtilis was made to express listeriolysin O, targeting it to the cytoplasm of a macrophage cell line, it became able to proliferate in this cell type [86
]. Using a similar approach, it was shown that B. suis expressing listeriolysin O failed to proliferate after lysing the membrane vacuole in human monocytes [87
]. Lack of proliferation in the cytoplasm might be seen as evidence for the presence of some restricting factor. This might be simply the lack of some nutrient or the presence of some toxic antimicrobial factor. In the latter case, it would obviously have to be specific for some but not all microbes.
Brucella mutants deficient in the virB genes fail to target their final replication niche in nonphagocytic cells, a compartment with ER characteristics, remaining in an intermediate organelle between the ER and autophagic vacuole and losing their ability to replicate [88 ]. Whether these observations can be explained by the lack of suitable, nutritive conditions for the bacteria in the new cellular compartment is yet unclear. However, Brucella do not reach autophagic vacuoles in macrophages [89 ].
Promoting the fusion of phagosomes with lysosomes has different consequences according to the microbe. M. avium vacuoles made fusogenic with the lysosomes by co-infecting macrophages with this mycobacterium and Coxiella remain hospitable to the mycobacteria, despite the acidification and the acquisition of lysosomal characteristics [90 ]. This can be used as an argument to suggest that the subversion tactics of M. avium were selected, not to prevent exposure to the toxic lysosomal environment but to keep the microbe in a nutrient-favorable site. In the case of the co-infection, such nutrient-favorable conditions are likely maintained. In contrast, a similar strategy applied to M. tuberculosis leads to reduced proliferative capacity, suggesting that indeed, the lysosomal environment is too harsh for this mycobacterial species. In this regard, it has been shown that M. avium is more resistant to low pH than M. tuberculosis [91 ]. However, using a different strategy, Anes et al. [92 ] showed that promotion of phagosome/lysosome fusion with specific fatty acids led to reduced proliferation of M. tuberculosis or M. avium. Also, in the case of Salmonella, the promotion of phagosome-lysosome fusion leads to an inability to proliferate [93 , 94 ].
|
|
|---|
Not much is known about the pathways leading to the putative deprivation of specific nutrients. One can envision mechanisms relying on degradative or withholding systems (enzymes that destroy particular nutrients such as amino acids, transporters that remove specific compounds from defined intracellular compartments, scavengers of specific substances such as iron chelators), as well as cellular mechanisms regulating the access of certain molecules to different compartments within the cell or even supracellular mechanisms regulating the access of nutrients to particular cells (e.g., within structures such as granulomas or abcesses). Figure 1 presents a summary of some of the possible mechanisms able to lead to nutrient deprivation.
![]() View larger version (29K): [in a new window] |
Figure 1. Summary of the possible mechanisms leading to nutrient deprivation in infected macrophages. The vacuole containing the pathogen (gray circle) may acquire transporters that remove nutrients (e.g., NRAMP1 may remove iron), enzymes that degrade specific molecules [e.g., indoleamine 2,3-dioxygenase (IDO) may metabolize tryptophan], or chelators that sequester certain compounds (e.g., iron chelators). Signaling initiated by cytokine, pattern recognition, or other receptors may also affect the maturation of the vacuole, namely by preventing its interaction with endosomes (a source of nutrients such as transferrin-bound iron), its interaction with other subcellular compartments with the generation of special replicative vacuoles or the lysis of the vacuolar membrane.
|
Induced mechanisms
A major inducer of macrophage antimicrobial functions is the cytokine IFN-
. It is well known that IFN-
-activated macrophages produce oxidative radicals such as those dependent on the phagocyte NADH phosphate oxidase and the iNOS (type 2). Nevertheless, many reports suggest that the control of microbial growth is not fully explained by the activity of these pathways. Indeed, bacteria such as L. pneumophila, M. avium, A. phagocytophilum, and L. monocytogenes and the protozoa T. cruzi resist the toxicity of NO and may even depend on its production for survival inside mice [3
4
5
6
7
8
9
105
106
]. Thus, it is clear that additional antimicrobial mechanisms dependent on IFN-
are triggered, and one of the possibilities is a series of nutriprive mechanisms.
One of the best-studied of such mechanisms is the effect of IFN-
on iron metabolism. The control of L. pneumophila proliferation in human monocytes induced by IFN-
was associated with its ability to restrict the access to iron [107
]. Activated monocytes were shown to have reduced expression of the receptor for transferrin, a mechanism that can, at least in part, account for the phenomenon of iron withholding [107
, 108
]. Levels of expression of the transferrin receptor on mouse macrophages were also shown to correlate with their ability to support the growth of L. monocytogenes [109
], and mRNA levels for the transferrin receptor were seen to decrease following activation with IFN-
[104
]. Similar iron limitation mechanisms may underlie the anti-Histoplasma capsulatum activity of IFN-
-activated mouse macrophages [110
]. Treatment with the cytokines IFN-
and TNF leads to a decrease in the access to iron by mycobacteria [31
], although acquisition through other chelates such as lactoferrin might increase following IFN-
activation [32
]. However, T. gondii, Chlamydia psittaci, and Leishmania donovani were shown to avoid these antimicrobial mechanisms [111
], and given the lack of iron requirements of Borrelia burgdorferi [112
], it is expected that they will not play a role against this pathogen. Cytokines also affect the expression of the putative iron transporter NRAMP1: It was shown that IFN-
activates macrophages to increase its expression [98
103
113
], and TNF enhances the expression of the Nramp1 gene [104
], a phenomenon that is consistent with the ability of these cytokines to restrict access to iron and to exert antimicrobial activity.
It is curious that in the study of Schnappinger et al. [79
], it was found that the up-regulation of genes in M. tuberculosis, which is specifically found in IFN-
-activated macrophages and not in nonactivated cells, was strictly dependent on the production of NO by the phagocytic cell. The lack of a specific RNA signature for a putative nutriprive mechanism is not, however, an argument against the existence of such a mechanism. The well-documented restriction of iron, present in nonactivated cells but made more intense upon IFN-
activation, is such a case. Conversely, the change in the requirement for the glyoxylate cycle enzymes, which occurs during the chronic stages of the infection by mycobacteria and salmonellae [64
, 65
], suggests that there are metabolic alterations induced by the immune response such that the availability of carbon sources changes.
A second well-known pathway triggered by IFN-
and leading to the deprivation of a nutrient is the activation of the enzyme IDO, which degrades L-tryptophan into N-formylkynurenine and has been shown to play a role in the control of Toxoplasma [114
115
116
], Leishmania [115
], and Chlamydia infections [115
116
117
]. However, the in vivo significance of the tryptophan-metabolizing pathways is confounded by the multiple effects that tryptophan catabolism has on microbes and the immune system: from simple depletion of the amino acid to the generation of immunomodulators (e.g., picolinic acid, shown to promote antimycobacterial activity in macrophages [118
]) and from the deprivation of the microbe to the deprivation of the immune cell [119
].
Toll-like receptors (TLRs) may convey antimicrobial signals to the macrophage following ligation of microbial molecules. It was shown that engagement of TLR2 on macrophages led to killing of M. tuberculosis [120 ]. Although macrophages triggered through TLRs can be induced to produce antimicrobial molecules such as NO [120 ], additional antimicrobial mechanisms are being sought. Blander and Medzhitov [121 ] described deficient phagocytosis of microorganisms in macrophages defective in TLR2 and -4 or their signal transducer myeloid differentiation primary-response protein 88 (MyD88). It is more interesting that such deficient macrophages showed impaired maturation of phagosomes, and pathogens such as salmonellae ended up in a vacuolar compartment different from the one they normally occupy. These data have, however, not been reproduced in an independent study using alternative technical approaches [122 ]. Also, not every microbe may be susceptible to this pathway, such as in the case of L. monocytogenes [123 ].
Counteracting the subversion pathways of parasites
Most of the known mechanisms used by the different microbes to survive inside macrophages have been described in nonactivated phagocytes. It is thus likely that once cytokine-activated macrophages are studied, it will be found that many of these microbial strategies are overcome by the host cell. Specifically, the inhibition of those subversion pathways that lead to acquisition of nutrients will fall under our definition of nutriprive mechanisms of the macrophage. It is known that the inhibition of acidification of mycobacterial phagosomes is reverted upon activation of macrophages with IFN-
[124
125
]. In addition to stimulating the secretion of toxic radicals, IFN-
acts on macrophages to induce the expression of a 47-kDa guanosinetriphosphatase, LRG-47, which promotes the maturation of the mycobacterial phagosome [126
]. In the absence of this effector molecule, the mycobacterial phagosome fails to acidify in response to IFN-
, as there is reduced assembly of the proton v-ATPase on its membrane, while interacting with the recycling endosomes and thus acquiring transferrin. The role of LRG-47 extends to other intracellular microbes such as T. gondii and L. monocytogenes [127
].
The activation of macrophages with IFN-
prevents the normal escape of L. pneumophila through the interaction with the ER, forcing the Legionella-containing phagosome to follow the default route of fusion with the lysosome [128
]. This is associated with restriction of growth of the bacterium. What happens to the use of the SLC1A5 transporter is not yet known. Also, IFN-
activation of macrophages has been shown to prevent the escape of L. monocytogenes into the cytoplasm of the macrophage [129
] and can thus be seen as activating a nutriprive mechanism toward this bacterium.
M. tuberculosis dwells in endosomal compartments with access to transferrin and thus to iron [95
]. The effect of activation, namely by cytokines, on this mechanism has hardly been studied. M. avium phagosomes interact actively with the endosomal compartment [12
13
14
]. We have shown that activation of M. avium-infected macrophages with IFN-
plus picolinic acid effectively restricts mycobacterial growth [118
] and at the same time, severely restricts the fusion of endosomes with the mycobacterial vacuole [130
], thus suggesting a way of restricting access to nutrients via the endosomal pathway. In this system, macrophages exhibit some of the alterations typical of apoptosis and raise the possibility that apoptotic macrophages limit the access of nutrients to the infecting parasites. It is interesting to note that apoptosis can be induced through the TLR signaling pathways, which at least in some cases, were shown to lead to antimicrobial activity [131
132
133
134
]. An alternative way to seclude the intracellular pathogen from the access to nutrients may involve the induction of autophagy, recently described to be associated with BCG and M. tuberculosis killing [135
]. It is interesting to note in this regard that a vitamin D analog can induce autophagic cell death through the phosphatidylinositol-3 kinase subunit, beclin-1 [136
]; it has been known for a long time that vitamin D3 has an antimycobacterial effect, but the pathways involved are yet unknown.
Leishmania mexicana acquires material (presumably nutrients such as the purines they require to grow) from the cytosol through two different pathways: One uses an organic anion transporter assembled in many intracellular vacuoles, namely those that harbor other pathogens such as H. capsulatum or a listeriolysin-deficient mutant of L. monocytogenes or even vacuoles containing inert particles such as immunoglobulin G-covered latex beads; a second pathway relies on the fusion of vesicles deriving from autophagic events with the parasitophorous vacuole [17 ]. Like mycobacteria, the parasitophorous vacuole interacts with other membrane compartments of the cell, thus acquiring cargo from these vesicular compartments [16 , 18 , 20 ]. These materials can then be ingested by the parasite [16 ]. Again, little is known about how the activation of the macrophages affects these events.
|
|
|---|
The understanding of the nutritive requirements of microbial pathogens and the nutriprive mechanisms induced in the macrophage will provide a framework for the identification of new targets for chemotherapy, namely inhibitors of the biosynthetic pathways essential for microbial proliferation and survival. Harth and Horwitz [137 ] presented proof of the concept that interfering with nitrogen metabolism leads to the identification of antimicrobial drugs. Particularly interesting candidates are the anaplerotic pathways turned on by limiting carbon sources. The identification of microbial products turned on after induction of nutrient deprivation and their use as antigenic targets in new vaccines can also be envisioned. Dietary manipulation of phagosome maturation and as we postulate, of access to nutrients may provide yet another singular approach as suggested by Anes and collaborators [92 ].
In summary, nutriprive mechanisms, acting in infected macrophages, exist and need to be better understood so that we gain better insight into infectious diseases caused by intracellular parasites and may one day be able to control them more effectively.
Received February 3, 2006; revised March 6, 2006; accepted March 12, 2006.
|
|
|---|
interferon in permissive A/J mouse macrophages: role of reactive oxygen species, nitric oxide, tryptophan, and iron(III) Infect. Immun. 62,3197-3205
and nitric oxide synthase 2 J. Exp. Med. 192,347-358
purC Mycobacterium tuberculosis auxotrophic mutant in
interferon receptor-deficient mice after aerosol infection Infect. Immun. 73,666-670
-induced T cell loss during infection by virulent Mycobacterium avium Infect. Immun. 73,3577-3586
and IFN-
stimulate a macrophage precursor cell line to kill Listeria monocytogenes in a nitric oxide-independent manner J. Immunol. 153,5141-5147[Abstract]
-activated human monocytes downregulate transferrin receptors and inhibit the intracellular multiplication of Legionella pneumophila by limiting the availability of iron J. Clin. Invest. 83,1457-1465[Medline]
J. Clin. Invest. 91,969-976[Medline]
interferon-mediated antihistoplasma state of murine macrophages Infect. Immun. 59,2274-2278
Interferon-activated human macrophages and Toxoplasma gondii, Chlamydia psittaci, and Leishmania donovani: antimicrobial role of limiting intracellular iron Infect. Immun. 59,4684-4686
blocks the growth of Toxoplasma gondii in human fibroblasts by inducing the host cells to degrade tryptophan Proc. Natl. Acad. Sci. USA 81,908-912
interferon-stimulated human macrophages Infect. Immun. 57,845-849
-mediated antimicrobial response. Indoleamine 2,3-dioxygenase-deficient mutant host cells no longer inhibit intracellular Chlamydia spp. or Toxoplasma growth J. Immunol. 150,5529-5534[Abstract]
in the outcome of chlamydial infection Curr. Opin. Immunol. 14,444-451[CrossRef][Medline]
-inducible LRG-47 Science 302,654-659
-inducible genes with essential, pathogen-specific roles in resistance to infection J. Exp. Med. 194,181-187
interferon-activated macrophages Infect. Immun. 73,3166-3171
Interferon limits access of Listeria monocytogenes to the macrophage cytoplasm J. Exp. Med. 170,2141-2146
Microbiology 150,1507-1518This article has been cited by other articles:
![]() |
O. Chacon, L. E. Bermudez, D. K. Zinniel, H. K. Chahal, R. J. Fenton, Z. Feng, K. Hanford, L. G. Adams, and R. G. Barletta Impairment of D-alanine biosynthesis in Mycobacterium smegmatis determines decreased intracellular survival in human macrophages Microbiology, May 1, 2009; 155(5): 1440 - 1450. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Omsland, D. C. Cockrell, D. Howe, E. R. Fischer, K. Virtaneva, D. E. Sturdevant, S. F. Porcella, and R. A. Heinzen Host cell-free growth of the Q fever bacterium Coxiella burnetii PNAS, March 17, 2009; 106(11): 4430 - 4434. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. R. Fuller, R. R. Craven, J. D. Hall, T. M. Kijek, S. Taft-Benz, and T. H. Kawula RipA, a Cytoplasmic Membrane Protein Conserved among Francisella Species, Is Required for Intracellular Survival Infect. Immun., November 1, 2008; 76(11): 4934 - 4943. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Omsland, D. C. Cockrell, E. R. Fischer, and R. A. Heinzen Sustained Axenic Metabolic Activity by the Obligate Intracellular Bacterium Coxiella burnetii J. Bacteriol., May 1, 2008; 190(9): 3203 - 3212. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. K. Pandey and C. M. Sassetti Mycobacterial persistence requires the utilization of host cholesterol PNAS, March 18, 2008; 105(11): 4376 - 4380. [Abstract] [Full Text] [PDF] |
||||
![]() |
I. J. Glomski, J.-P. Corre, M. Mock, and P. L. Goossens Noncapsulated Toxinogenic Bacillus anthracis Presents a Specific Growth and Dissemination Pattern in Naive and Protective Antigen-Immune Mice Infect. Immun., October 1, 2007; 75(10): 4754 - 4761. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||