Identification of Mycobacterium avium Genes That Affect Invasion of the Intestinal Epithelium
http://www.100md.com
感染与免疫杂志 2005年第7期
Kuzell Institute for Arthritis & Infectious Diseases, California Pacific Medical Center Research Institute, San Francisco, California
Department of Veterinary and Biomedical Sciences, University of Nebraska, Lincoln, Nebraska
Department of Biomedical Sciences, College of Veterinary Medicine, and Department of Microbiology, College of Sciences, Oregon State University, Corvallis, Oregon
ABSTRACT
Invasion of intestinal mucosa of the host by Mycobacterium avium is a critical step in pathogenesis and likely involves several different bacterial proteins, lipids, glycoproteins, and/or glycolipids. Through the screening of an M. avium genomic library in Mycobacterium smegmatis, we have identified a number of M. avium genes that are associated with increased invasion of mucosal epithelial cells. In order to further investigate these genes, we cloned six of them into a plasmid downstream of a strong mycobacterial promoter (L5 mycobacterial phage promoter), resulting in constitutive expression. Bacteria were then evaluated for increased expression and examined for invasion of HT-29 intestinal epithelial cells. The genes identified encode proteins that are similar to (i) M. tuberculosis coenzyme A carboxylase, (ii) M. tuberculosis membrane proteins of unknown function, (iii) M. tuberculosis FadE20, (iv) a Mycobacterium paratuberculosis surface protein, and (v) M. tuberculosis cyclopropane fatty acyl-phopholipid synthase. The constitutive expression of these genes confers to M. avium the ability to invade HT-29 intestinal epithelial cells with a severalfold increase in efficiency compared to both the wild-type M. avium and M. avium containing the vector alone. Using the murine intestinal ligated loop model, it was observed that the constitutive expression of M. avium proteins has a modest impact on the ability to enter the intestinal mucosa when compared with the wild-type control, suggesting that under in vivo conditions these genes are expressed at higher levels. Evaluation of the expression of these invasion-related genes indicated that under conditions similar to the intestinal lumen environment, the genes identified are upregulated. These data suggest that invasion of the intestinal mucosa is an event that requires the participation of several bacterial factors and the expression of the genes that encode them is less observed under standard laboratory growth conditions.
INTRODUCTION
A number of bacterial pathogens, including Salmonella, Yersinia, Shigella, and Escherichia coli, can utilize the intestine as a portal for entry during infection of humans. The molecular mechanisms used by these bacterial pathogens to invade and circumvent the natural defenses of the host are usually quite elegant and complex, involving a large number of bacterial genes (18, 21, 39). The factors encoded by these genes range from components of type III secretion apparatus (21) and outer membrane proteins (18) to enzymes involved in fatty acid biosynthesis (27, 40). Although the molecular mechanisms of invasion by Mycobacterium spp. have been studied to some extent in macrophages (16a), very little is known regarding their mechanisms of intestinal epithelial invasion. This paucity of information exists despite the fact that this route of infection is important for Mycobacterium (7, 23, 24, 32), and invasion of the respiratory epithelium has been proposed to play a role in pathogenesis of Mycobacterium tuberculosis (3, 20, 38).
Mycobacterium avium is an environmental organism that infects humans through both the intestinal tract and the respiratory tract (17, 24). This bacterium is considered a facultative intracellular pathogen that causes disease in both patients with and patients without AIDS. In contrast to infections in non-AIDS populations, where pulmonary infection is most prevalent, immunocompromised patients develop disseminated disease (15, 17, 24). In these individuals, both epidemiologic evidence and clinical evidence suggest that M. avium is predominantly acquired following ingestion of food or water (17, 24). The bacterium can resist acidic conditions in the stomach and colonize the intestinal tract (10, 23). Recent studies have demonstrated that M. avium interacts with the intestinal mucosa, entering primarily enterocytes (7, 35). Invasion of enterocytes appears to occur through the apical membrane rather than the basolateral surface (35). Therefore, it is likely that the abilities to invade and translocate across the intestinal mucosa are important components of M. avium pathogenesis. M. avium invasion of intestinal mucosal cells requires cytoskeleton rearrangements and protein phosphorylation in the host cell (9), indicating an active participation of the bacterium in the process by modulating host proteins. The genes and gene products that play a role in this stage of infection have not been identified.
Several genes and proteins have been suggested as participants in the invasion of intestinal epithelial cells by M. avium, such as fibronectin attachment protein (37), a 27-kDa protein (8), and superoxide dismutase (34). In addition, mce-1, a gene suggested to participate in Mycobacterium tuberculosis invasion of HeLa cells (1) and heparin binding hemagglutinin (29) are present in the M. avium genome. Although some of these proteins may have a role in interactions with the intestinal mucosa, it is likely that this process is multifactorial.
In order to obtain a better understanding of the genes associated with invasion of intestinal epithelial cells and the mechanism(s) of entry used, we set out to identify novel M. avium genes that increase the ability of M. smegmatis (a nonpathogenic mycobacterium) to invade intestinal epithelial cells.
MATERIALS AND METHODS
Strains and growth media. Mycobacterium smegmatis strain mc2155, selected for its high transformation efficiency (25), was grown in Middlebrook 7H9 broth with 0.1% Tween 80 or in Middlebrook 7H11 agar with 0.1% Tween 80. M. avium strains 101 and 104, isolated from the blood of AIDS patients, were cultured on 7H9 broth or 7H11 agar. E. coli XL1-Blue MRF' (Stratagene, LaJolla, CA) transformants were selected on Luria-Bertani agar and in Circlegrow broth (BIO 101) containing 50 μg/ml kanamycin. M. smegmatis transformants were selected on 7H11 agar with 0.1% Tween 80 and 50 μg/ml kanamycin.
Tissue culture. HEp-2 cells, a human epithelial cell line from the American Type Culture Collection (ATTC; CCL-23), Manassas, Virginia, were grown in RPMI 1640 supplemented with 10% fetal bovine serum and used at 80% confluence (approximately 1 x 106 cells) at 37°C under an atmosphere of 10% CO2. The HT-29 human intestinal cell line was also obtained from ATCC. The cells were grown in McCoy's medium (Gibco) supplemented with 10% fetal bovine serum in tissue culture plates and used when at 100% confluence (approximately 1 x 106 cells), as previously reported (9).
Construction of the MAC 101 genomic library. Genomic DNA was isolated from MAC 101 as described previously (28) and partially digested with the restriction enzyme Sau3A. After electrophoresis on a 1% agarose gel, fragments in the 3- to 6-kb range were recovered and cloned into the BamHI site on the mycobacterium-E. coli shuttle plasmid pMV261 (25). The genomic library was transformed into electroporation-competent E. coli and plated on Luria-Bertani agar containing kanamycin. In this manner, approximately 10,000 transformants were generated. Restriction analysis of plasmids isolated from E. coli transformants showed 80% contained insert DNA, the average size being 4 kb. The library in E. coli was stored in Circlegrow broth (Bio 101) plus 30% glycerol, and aliquots were plated for amplification of the library and plasmid extraction. The library DNA was then used to transform M. smegmatis, resulting in approximately 17,000 M. smegmatis transformants pooled in 7H9 broth with 30% glycerol.
Electroporation of M. smegmatis and E. coli. M. smegmatis competent cells were made following a protocol adapted from Bio-Rad and Jacobs et al. (25). The optical density at 600 nm of overnight cultures (log phase) of M. smegmatis was measured and determined to correspond to approximately 5 x 108 bacteria. The bacteria were incubated on ice for 1.5 h, harvested by centrifugation, and washed three times with cold 10% glycerol plus 0.1% Tween 80. Pellets were resuspended in 1 ml 10% glycerol plus 0.1% Tween 80, aliquoted, and frozen in an ice-ethanol bath for storage at –70°C. For electroporation, 100 to 200 μl of cells was combined with a 1- to 5-μg library DNA and transferred to a prechilled 0.2-cm cuvette. Electroporation was carried out using a Gene Pulser (Bio-Rad, Hercules, CA) at the following settings: 25 μF, 1,000 , 2.5 kV. Cells were recovered with 1 ml 7H9 and incubated at 37°C with shaking for 2 h. Electrocompetent E. coli XL1-Blue MRF' cells were either purchased from Stratagene or prepared and transformed as previously described (33).
Invasion assay. Inocula for invasion assays were prepared as follows. M. smegmatis containing the library DNA was recovered from 7H11 agar and suspended in Hanks' balanced salt solution (HBSS), diluted to match McFarland standard 2 (approximately 6 x 108 CFU/ml) (7). The suspensions were passed 10 times through a 26-gauge needle, and large aggregates were allowed to settle. After 5 min, an aliquot was taken from the top half of the bacterial suspension and diluted in HBSS to 107 CFU/ml. One hundred microliters of this suspension was used to infect HEp-2 monolayers. Actual inoculum concentrations were determined by plating serial dilutions as described previously (9). Microscopic examination of inoculum using the LIVE/DEAD assay (Molecular Probes, Eugene, OR) showed the suspensions to contain single, viable cells (5).
The invasion assays were carried out as previously reported (9). HEp-2 and HT-29 monolayers were grown to 80% and 100% confluence (approximately 1 x 106 HEp-2 and HT-29 cells), respectively, in 24-well tissue culture plates (Costar, Pleasanton, CA). Prior to the assay, the culture medium was removed and replaced with fresh medium. Each well was inoculated with 106 bacteria, and the plate was incubated at 37°C in a 5% CO2 incubator for 2 h. Extracellular bacteria were removed by washing monolayers three times with HBSS. Any remaining extracellular bacteria were killed by replacing the medium with medium that contained 200 μg/ml amikacin and incubating for 2 h at 37°C (9). Monolayers were washed three times with HBSS after amikacin treatment and lysed with 1 ml sterile water. The number of viable intracellular bacteria was determined by plating lysates of duplicate wells on 7H11 agar containing 50 μg/ml kanamycin. In some experiments, bacteria were exposed to 0.3 M NaCl (final concentration), anaerobiosis (anaerobic jar), or both for 4 h prior to the assay as previously reported (6). Clones with a percentage of invasion of 1.5 or more over a background of 0.05% (M. smegmatis) were chosen for further study. A control strain of M. smegmatis with the skeleton plasmid pMV261 was used in some assays.
In some assays, infected monolayers were fixed with 2% paraformaldehyde for 30 min at room temperature, washed, and then stained for acid-fast bacilli using the Runyoun method as described previously (9). Intracellular bacteria were counted under the light microscope, and the average number of bacteria per cell and the percentage of infected cells were determined.
Nucleotide sequence and data analysis. M. smegmatis clones that display increased invasion were sequenced at the University of California, San Francisco Biomolecular Resource Center using an Applied Biosystems DNA sequencer and at the University of Nebraska, Lincoln. Sequence comparisons were performed using BLAST 2.0 from the National Center for Biotechnology Information. Sequence data for the M. avium genome were obtained from The Institute for Genomic Research (TIGR; http://www.tigr.org).
Construction of subclones. Clones from the MAC 101 library were further examined by subcloning the insert region of DNA into pFJS8 for overexpression from the mycobacteriophage L5 promoter. The resulting constructs were then transformed into MAC 104 for invasion assays, and RNA dot blots were done to confirm expression of the cloned genes.
The plasmid pFJS8 is an E. coli-mycobacterium shuttle plasmid containing the L5 promoter added for pJ5F8 conduction. The plasmid was constructed as follows. The hsp60 promoter was cut off the plasmid pMV261 (25) by using the enzymes PvuII and DraI, making the plasmid pFJS3. Then pFJS3 was cut with SalI and ClaI, and the mycobacteriophage L5 promoter was cloned in it. In order to subclone each region of interest, sequence data from each of the clones displaying increased invasion were used to design primers to amplify the corresponding region from MAC 101 genomic DNA by PCR. PCR products were ligated directly into pCR2.1 (Invitrogen, Carlsbad, CA) and screened by digestion to determine the direction of the insert. The inserts were then digested from pCR2.1 using a HindIII/NsiI double digest and cloned directionally into pFJS8 cut with HindIII and PstI. All subclones except pEMC7 were constructed in this manner. To construct pEMC7, the homolog to the M. tuberculosis gene Rv3802c was PCR amplified from MAC 101 using primers Rv3802c(NisI)/L (GGATGCATCGGGCCCGCGACAAAG) and Rv3802c(HindIII)U (GGGCAAGCTTGTACCGTGTAGAGC) containing the restriction sites HindIII and NsiI and cloned directly into the vector cut with HindIII and PstI. The primers clone342L (AGGTGGGCAAGGCGCTCAA) and clone342U (CGACGCGGAGAAATACGGTCT) were used amplify the MAC 101 homolog to the M. paratuberculosis protein AF232751, producing construct pEMC3. Primers clone335L (CGAGGCGTACAACACCGTCA) and clone 335U (GCTGCTGCAGAACGAGTTGGG) were used to clone the Rv3720 homolog, creating pEMC4. Plasmid pEMC8 was created as a control, with the Rv3720 gene cloned in opposite orientation to the L5 promoter. Primers fadE20L (CCGGCCGCATCACTCGAT) and fadE20U (GTAGGCCATGCGACGCTCGTT) were used to clone the fadE20 homolog, creating pEMC5. Primers clone 23ORF/L (GGCGGTGACGACGGTGAATCC) and clone 23ORF/U (CGGCCAGCTCCTTGCGGTAG) were used to create pEMC6. Primers L (CTCACCGATCGCTACGTC) and U (GTCCGGATCCGCTACCGC) were used to create pEMC9.
Regulation of invasion genes. To determine whether the genes identified are upregulated when the bacterium is exposed to conditions that model those encountered in the intestinal lumen, 107 M. avium cells were cultured in 7H9 broth, and after the culture reached the turbidity equivalent to 5 x 108 organisms, it was split and placed at 37°C under hyperosmolarity conditions (0.3 M) (6, 33a), anaerobiosis, or a combination of both conditions, as well as under iso-osmolar and aerobic conditions as a control, for 24 h. Total bacterial RNA was obtained as previously reported (41). Total RNA was initially quantified by absorbance at 260 nm, and quality was determined by the 260/280-nm absorbance ratio. Ratios of 1.8 were considered acceptable. RNA was then electrophoresed on an agarose gel to confirm quality. RNA was submitted to reverse transcriptase treatment to obtain cDNA. Briefly, 6 μl of RNA, 6 μl of random hexamer (50 mg/3 μl), 2 μl of deoxynucleoside triphosphate (dNTP) mix (10 mM), and distilled water were mixed and incubated at 65°C for 5 min and subsequently placed on ice for 2 min (reverse transcription [RT] mix). Then to 18 μl of the RT mix we added 5 μl of 10x RT buffer, 8 μl of MgCL2 (25 mM), and dithiothreitol (0.1 M). The mix was incubated at 25°C for 2 min, and 2 μl of Super Script RTII was added to the sample. The sample was incubated for 10 min at 25°C and then transferred to 42°C for 50 min. The reaction mixture was then placed on ice and centrifuged briefly, and 2 μl of RNase H was added at 37°C for 20 min. The sigA gene was used as the constitutively expressed control.
To amplify the cDNA, we used the specific primers for the invasion-associated genes used to clone the genes in pFJS8. PCR amplification was carried out at 95°C for 3 min (1 cycle), 95°C for 3 s, 62°C for 30 s, and 72°C for 2 min in a linear range (35 cycles) and 72°C for 10 min (1 cycle). Equivalent amounts of RNA were used for the RT-PCR. The products obtained by PCR amplification were quantified by using the Kodak EDEAS 290 system and the Kodak id image analysis software (Eastman Kodak Company, Rochester, NY).
Infection using the intestinal loop model. To determine whether the M. avium clones that displayed enhanced invasion in vitro were also associated with increased invasion of intestinal mucosa in vivo, we used the murine intestinal loop model. Bacteria (107) were injected into an intestinal loop of anesthetized mice as previously described (36). After 1 h, the mice were harvested, the intestinal loop was removed and cut open, and bacteria loosely associated with the mucosa were removed by washing. The tissue was then homogenized, and the viable bacteria were plated onto 7H11 agar as described previously (36). The results were calculated as the percentage of the initial inoculum that is present in the tissue, and the experiments were repeated three times.
Statistical analysis. The results shown represent the mean ± standard deviation of at least five experiments. The data were analyzed statistically by comparison with controls at the same time points using the Student's t test. P < 0.05 was considered statistically significant.
RESULTS
Enrichment of M. avium library and identification of invasive clones. Table 1 shows that consecutive passages of the pool of M. smegmatis clones containing M. avium DNA in HEp-2 cells enriched for a more invasive population of bacteria, whereas, this phenomenon was not observed in the wild-type M. smegmatis. After the third passage, the percentage of inoculum that entered HEp-2 cells increased from 0.07% to 4.3%. The cell lysate from this infection was then plated to obtain individual colonies, and they were each evaluated for invasion of HEp-2 cells. Eleven clones that displayed an efficiency of invasion greater than 1.5% (over a background of 0.05%, P < 0.05 as compared to the control; Table 2) were further characterized by purification of the plasmid that they carried and sequencing of the M. avium DNA insert. One gene was represented three times in the pool (fadE20), and three other genes were represented twice each (Af232751, and the homologues to Rv3802c and Rv3720). Since some of the invasion-related clones contained more than one gene (two genes were present in two of the clones), individual genes were cloned into M. smegmatis and screened for the ability to confer enhanced invasion. Each of the six individually subcloned genes conferred comparable levels of invasion to that observed with the original selected clone, indicating that these genes were, in fact, the ones associated with the enhanced invasion phenotype. All of the genes identified have homologues in other mycobacterial species, and most of them are present in M. tuberculosis (Table 2). These results also showed that M. avium genes associated with an increase in M. smegmatis invasion possibly belonged to two groups: surface proteins and enzymes involved in fatty acid synthesis and degradation.
M. smegmatis clones expressing M. avium genes did not show changes in colony morphology. In contrast, M. avium clones expressing constitutively the homologues to Rv3719 and Rv3720 showed significant change in the smoothness of the colony, which would suggest increased quantity of lipids on the cell wall.
Expression of invasion genes under different environmental conditions. Our past observation that M. avium invades both the intestinal mucosa, as well as intestinal epithelial cells, more efficiently under conditions that mimic the intestinal environment (6) suggests that invasion-related genes might be upregulated as a result of intestinal conditions (6). Therefore, if the six genes that we have identified in the current study are similar to those involved in the regulated invasion phenotype we observed previously, one would expect them to be repressed under standard laboratory conditions and upregulated under conditions that mimic those in the intestine, high osmolarity and low oxygen tension (6, 33a). We exposed M. avium 104 (MAC 104) to either low oxygen tension, high osmolarity, or both and performed RT-PCR to evaluate the expression of the invasion genes we identified. As shown in Fig. 1, while open reading frame 23 (carboxylase, pEMC6) was not expressed under laboratory conditions, it was highly expressed under conditions of high osmolarity as well as low oxygen tension. Interestingly, the high level of expression observed under low oxygen tension was also observed when the bacterium was exposed to both conditions simultaneously. The fadE20 (pEMC5) gene had an increased level of expression under low oxygen tension and under high-osmolarity conditions. The gene homologous to Rv3802 (clone 332, pEMC7) showed increased expression under both high-osmolarity and low-oxygen conditions. The gene homologous to Rv3720 (pEMC4) had increased expression under low oxygen tension, and the Af232751 (pEMC3) homologue showed increased expression under both conditions alone and when the conditions were combined. The sodC gene was used as a control because preliminary experiments showed that its expression is not influenced by any of the conditions used (data not shown). These data support the possibility that our selection procedure allowed identification of genes that are involved in invasion of intestinal epithelium because they are upregulated by conditions that mimic those found in the intestinal environment.
Constitutive expression enhances invasion of epithelial cells by M. avium. Although these genes confer enhanced entry to M. smegmatis, it remains possible that in M. avium they do not serve the same function. Our previous studies indicate that invasion of epithelial cells does not occur at very high levels in wild-type M. avium under standard laboratory conditions, but this phenotype can be induced by specific environmental conditions (5). Based on these data, we hypothesized that upregulation of invasion-associated genes would increase the ability to enter intestinal epithelial cells. Thus, if the genes identified are the same as those responsible for the repressed invasion phenotype, wild-type M. avium would be phenotypically similar to a mutant in these genes under standard laboratory growth conditions and the biological role of these genes could then be easily observed by expressing them under these conditions using a constitutive mycobacterial promoter. We have developed a vector that allows constitutive expression of genes from a promoter identified from the mycobacteriophage L5 (2a). In this controlled expression system, the gene being investigated would be cloned downstream of the L5 promoter and transformed into wild-type M. avium. Plasmids pEMC3 (Af232751 homologue), pEMC4 (Rv3720 homologue), pEMC5 (fadE20), pEMC6 (aacD4 homologue), pEMC7 (Rv3802 homologue), pEMC8 (Rv3720 in wrong orientation), and pEMC9 (Rv3273 homologue) were then created and transformed into the wild-type M. avium strain 104. These clones were then evaluated in invasion assays to determine if the constitutive expression of the identified genes is associated with an increased ability to enter epithelial cells. As shown in Fig. 2, all but MAC 104-pEMC8, which has the gene in the opposite orientation to the promoter, display an increased ability to enter epithelial cells. MAC 104-pEMC3, -pEMC4, -pEMC5, -pEMC6, -pEMC7, and -pEMC9 are associated with a significant increase in the ability to enter HT-29 cells as compared to the wild-type M. avium 104. These results were confirmed by light microscopy (Table 3).
Constitutive expression enhances invasion of intestinal epithelium by M. avium. Since the M. avium transformants with the M. avium invasion-related genes expressed constitutively were able to invade HT-29 intestinal cells in vitro with increased efficiency compared to the wild-type strain, we examined their ability to invade the intestinal mucosa in vivo using the mouse intestinal loop model of infection. We chose pEMC4 and pEMC6 for these studies because of their relatively high levels of invasion when expressed constitutively as compared to the other genes identified. M. avium 104-pEMC4 and M. avium MAC 104-pEMC6 were injected in the intestinal loop, and after 1 h, the percentage of the inoculum that invaded the intestine was determined. The results (Table 4) indicate that in vivo the constitutive expression of clone 23 (carboxylase) and the gene homologous to Rv3720 (cyclopropane-fatty acid-phospholipid synthase) had a significant impact on the ability of the bacterium to invade the intestinal mucosa. To determine if exposure to high-osmolarity conditions in vitro would have a similar effect on invasion, we compared invasion of HT-29 cell monolayers by clones exposed to high osmolarity (6, 33a) and to "laboratory conditions" (isotonic). As shown in Table 4, clones that express the "invasion-associated genes" enter HT-29 cells with a significantly higher efficiency than control M. avium cells grown under standard laboratory conditions prior to the assay. Incubation under high-osmolarity conditions increases the ability of wild-type M. avium to enter epithelial cells but had a limited impact on the entry of clones expressing invasion genes constitutively. These data suggest that the level of expression of the "invasion-associated genes" correlates with M. avium invasion of the intact intestinal epithelium as well as intestinal epithelial cells in vitro.
DISCUSSION
Similar to other enteric bacterial pathogens, invasion of the intestinal epithelium by M. avium is multifactorial. We have identified six M. avium genes that are associated with invasion of intestinal epithelial cells in vitro. We also found that these genes are transcribed at higher levels when the bacteria are exposed to conditions that are thought to be similar to those in the intestinal environment, increased osmolarity and/or low oxygen tension.
The sequences of the six genes identified predict for two types of proteins: membrane proteins of unknown function and enzymes involved in fatty acid synthesis, degradation, or transport. Some of the genes involved in fatty acid synthesis or degradation have some degree of homology in the M. smegmatis genome (fadE20), while others do not, but none of the genes coding for proteins has homologues in M. smegmatis (TIGR database). The M. avium fadE20 gene is quite different from the M. smegmatis gene (43% homology), which might help to explain its role in the invasion of intestinal epithelium.
Among the genes, fadE20 is a homologue of a M. tuberculosis acyl coenzyme A (acyl-CoA) dehydrogenase, a participant in the fatty acid degradation/synthesis pathway. The role of mycolic acid and other fatty acids in mycobacterial adherence may be related to their influence on the assembly of the cell wall. Fatty acids may facilitate binding in a nonspecific manner by influencing the glycolipid content or lipid ratios. Alternatively, mycolic acids and enzymes involved in their synthesis or degradation may regulate the presence of other protein or sugar components on the bacterial surface. Alteration of the fatty acid composition of the mycobacterial surface would likely affect the hydrophobic and electrostatic character of the bacteria, thereby, affecting the ability of the bacteria to bind host cell membranes and/or receptors. However, studies by C. Lee and colleagues demonstrated a regulatory role for FadD during invasion by Salmonella (27). The mycobacterial fadE gene identified for its effects on invasion in the current study may have a similar regulatory function. Interestingly, FadR and FadF have been reported to be part of the regulatory system in Salmonella and have been shown to repress hilA, a major regulatory gene of the pathogenicity island I (40). Furthermore, the importance of these fatty acid degradation and synthesis pathways in the pathogenic mechanism is suggested by two recent reports where investigators determined that fadD26 and fadD28 are involved in the virulence of Mycobacterium tuberculosis in vivo (11, 19). Mutations in both genes alter colony morphology as a result of interference with mycolic acid synthesis. Based on the M. avium genome sequence, the fadE20 gene is with a putative operon where the gene just upstream is hflx, a GTP-binding protein. GTP-binding proteins are often associated with regulatory genes and provide energy for their activity. Although speculative, the presence of fadE20 downstream of this gene is also suggestive of a regulatory role for this gene in the invasion of intestinal epithelium. It is for these reasons we consider it more likely that fadE20 plays a regulatory role in M. avium invasion, but further studies are necessary to definitively differentiate this hypothesis from the equally likely possibility of direct effects on invasion through changes in fatty acid composition of the mycobacterial cell surface.
Rv3720 is a cyclopropane fatty acyl-phospholipid synthase that is related to -mycolic acid synthesis and is likely to affect cell wall structure. M. tuberculosis, for example, synthesizes three major types of cyclopropanated mycolic acids (42), and both M. tuberculosis and M. avium have a large number of cyclopropane synthases in their genomes (12; TIGR database, unpublished) which suggests a number of potential functions for cyclopropanate mycolic acids. Recently, inactivation of a gene coding for a cyclopropane synthase, pcaA, was reported to attenuate the virulence of Mycobacterium tuberculosis (22). Our data suggest that cyclopropanated mycolic acid has a role in host cell entry. The mechanism by which cyclopropanated mycolic acid participates in invasion is unclear but is likely to involve fundamental changes in the nature of the bacterial surface that comes in contact with the host cell. The presence of this gene within a group of mycobacterial invasion genes further supports the importance of the lipid composition on the mycobacterial surface during entry into host cells.
In M. paratuberculosis, the accession number AF232752 is for a 35-kDa protein that is a surface protein and that appears to be expressed at higher levels under hyperosmolar conditions as compared to standard laboratory growth conditions (2). M. paratuberculosis cells have been shown to cross the intestinal mucosa by entering M cells in the terminal ileum (30), while M. avium interacts primarily with enterocytes and, to a lesser extent, with M cells (36). It is possible that M. paratuberculosis can cross the intestinal mucosa by either or both pathways, M cells and enterocytes. However, M. paratuberculosis transmission is assumed to occur very early in life, and calves are believed to be infected at birth or soon afterwards. If this assumption is correct, the young calf is likely to not have well-developed Peyer's patches and M cells. Several laboratories have demonstrated that Peyer's patches only appear after some degree of maturation of the intestines and exposure to "contaminated" contents (31).
Two of the other membrane proteins identified are similar to M. tuberculosis proteins, Rv2724c (62% identity and 48% similarity) and Rv3723 (74% identity and 76% similarity). M. tuberculosis is not known to invade the intestinal mucosa, but intestinal tuberculoma can be seen in patients who do not possess a functional stomach barrier (14). Furthermore, M. tuberculosis, in contrast to M. avium, is not tolerant of the acid pH of the stomach (14). Therefore, M. tuberculosis may use similar machinery to invade respiratory epithelial cells (18). Another possible explanation is that the M. avium homologue to Rv3723 is in fact a different protein (only 34% identical to the M. tuberculosis protein) and evolved a different function. Similar modification of existing proteins for novel functions has been previously observed in many other microorganisms (26).
The Rv3802c homologue is a probable membrane protein that contains an N-terminal signal sequence followed by a proline-rich region. These are characteristics encountered in internalin (inlA), a Listeria monocytogenes protein associated with the invasion of epithelial cells (13). Our data indicate that M. avium Rv3802c also has a role in invasion. If this protein has a similar function to that of internalin, in terms of interaction with host cells, its identification would represent an important breakthrough in our understanding of M. avium invasion, but definitive demonstration of this role will require additional more detailed analysis of the role of this specific gene during interactions with host cells.
The application of a system for exogenously controlled expression of genes offers a useful strategy to screen for virulence genes that are only expressed well or at all under conditions found in the host. The previous observation that M. avium exposed to high osmolarity or anaerobiosis had an increased ability to invade intestinal epithelial cells (5) suggests that, under conditions encountered in the intestinal tract, M. avium genes that participate in the invasion process are upregulated. Due to these findings, we hypothesized that invasion-related proteins or lipids would be expressed at a higher level in vivo. Our results support our hypothesis, and only one among the originally identified genes in M. smegmatis was not associated with increased invasion when expressed constitutively. The fact that a couple of the genes that display increased expression when bacteria were exposed to either low oxygen tension or high osmolarity are suppressed when they are exposed simultaneously to both conditions suggests that the bacterium may not be exposed to both of these environmental conditions at the same time in vivo.
Several other genes have been associated with Mycobacterium invasion of epithelial cells, such as fibronectin attachment proteins (to bladder epithelial cells), mce-1 (M. tuberculosis binding to HeLa cells), the M. avium 27-kDa protein (8), and M. tuberculosis hemagglutinin (29) in the binding to alveolar epithelial cells. M. avium binding and invasion of epithelial intestinal cells in vivo appear to require an array of moieties to interact with mucin, microvilli, and membranes of the host cell, triggering cytoskeletal reorganization (9), activation of RhoA (35), and protein phosphorylation (9).
The methods employed in this paper have limitations, although they are the best that can be achieved with M. avium at this point. We have developed genetic systems for M. avium, and future studies will address the importance of the identified genes to invasion.
ACKNOWLEDGMENTS
This work was supported by the grant AI-43199 from the National Institute of Allergy and Infectious Diseases.
We thank Karen Allen and Denny Weber for help in the preparation of the manuscript.
REFERENCES
1. Arruda, S., G. Bomfim, R. Knights, T. Huima-Byron, and L. W. Riley. 1993. Cloning of an M. tuberculosis DNA fragment associated with entry and survival inside cells. Science 261:1454-1457.
2. Bannantine, J. P., J. F. J. Huntley, E. Miltner, J. R. Stabel, and L. E. Bermudez. 2003. Mycobacterium avium subsp paratuberculosis 35 KDa protein plays a role in invasion of bovine epithelial cells. Microbiology 149:2061-2069.
2. Barletta, R. G., D. D. Kim, S. B. Snapper, B. R. Bloom, and J. W. R. Jacobs. 1992. Identification of expression signals of the mycobacteriophages Bxb1, L1 and TM4 using Escherichia-Mycobacterium shuttle plasmids pYUB75 and pYUB76 designed to create translational fusions to the lacZ gene. J. Gen. Microbiol. 138:23-30.
3. Bermudez, L. E., and J. Goodman. 1996. Mycobacterium tuberculosis invades and replicates within type II alveolar cells. Infect. Immun. 64:1400-1406.
4. Reference deleted.
5. Bermudez, L. E., A. Parker, and J. R. Goodman. 1997. Growth within macrophages increases the efficiency of Mycobacterium avium in invading other macrophages by a complement receptor-independent pathway. Infect. Immun. 65:1916-1925.
6. Bermudez, L. E., M. Petrofsky, and J. Goodman. 1997. Exposure to low oxygen tension and increased osmolarity enhance the ability of Mycobacterium avium to enter intestinal epithelial (HT-29) cells. Infect Immun. 65:3768-3773.
7. Bermudez, L. E., M. Petrofsky, P. Kolonoski, and L. S. Young. 1992. An animal model of Mycobacterium avium complex disseminated infection after colonization of the intestinal tract. J. Infect. Dis. 165:75-79.
8. Bermudez, L. E., K. Shelton, and L. S. Young. 1995. Comparison of the ability of Mycobacterium avium, M. smegmatis and M. tuberculosis to invade and replicate within HEp-2 epithelial cells. Tuber. Lung Dis. 76:240-247.
9. Bermudez, L. E., and L. S. Young. 1994. Factors affecting invasion of HT-29 and HEp-2 epithelial cells by organisms of the Mycobacterium avium complex. Infect. Immun. 62:2021-2026.
10. Bodmer, T., E. Miltner, and L. E. Bermudez. 2000. Mycobacterium avium resists exposure to the acidic conditions of the stomach. FEMS Microbiol. Lett. 182:45-49.
11. Camacho, L. R., D. Ensergueix, E. Perez, B. Gicquel, and C. Guilhot. 1999. Identification of a virulence gene cluster of Mycobacterium tuberculosis by signature-tagged transposon mutagenesis. Mol. Microbiol. 34:257-267.
12. Cole, S. T., R. Brosch, J. Parkhill, T. Garnier, C. Churcher, D. Harris, S. V. Gordon, K. Eiglmeier, S. Gas, C. E. Barry III, F. Tekaia, K. Badcock, D. Basham, D. Brown, T. Chillingworth, R. Connor, R. Davies, K. Devlin, T. Feltwell, S. Gentles, N. Hamlin, S. Holroyd, T. Hornsby, K. Jagels, B. G. Barrell et al. 1998. Deciphering the biology of Mycobacterium tuberculosis from the complete genome sequence. Nature. 393:537-544.
13. Cossart, P., and M. Lecuit. 1998. Interactions of Listeria monocytogenes with mammalian cells during entry and actin-based movement: bacterial factors, cellular ligands and signaling. EMBO J. 17:3797-3806.
14. Curtiss, K. J., and M. H. Sleisenger. 1978. Infectious and parasitic diseases, p. 348. In M. H. Sleisenger and J. S. Fordtran (ed.), Gastrointestinal disease, Philadelphia, Pa.
15. Damsker, B., and E. J. Bottone. 1985. Mycobacterium avium-Mycobacterium intracellulare from the intestinal tracts of patients with the acquired immunodeficiency syndrome: concepts regarding acquisition and pathogenesis. J. Infect. Dis. 151:179-181.
16. Reference deleted.
16. Ernst, J. D. 1998. Macrophage receptor for Mycobacterium tuberculosis. Infect. Immun. 66:1277-1279.
17. Falkinham, J. O., III. 1996. Epidemiology of infection by nontuberculous mycobacteria. Clin. Microbiol. Rev. 9:177-215.
18. Finlay, B. B., and S. Falkow. 1997. Common themes in microbial pathogenicity revisited. Microbiol. Mol. Biol. Rev. 61:136-169.
19. Fitzmaurice, A. M., and P. E. Kolattukudy. 1997. Open reading frame 3, which is adjacent to the mycocerosic acid synthase gene, is expressed as an acyl coenzyme A synthase in Mycobacterium bovis BCG. J. Bacteriol. 179:2608-2615.
20. Fujimura, Y. 1986. Functional morphology of microfold cells (M cells) in Peyer's patches—phagocytosis and transport of BCG by M cells into rabbit Peyer's patches. Gastroenterol. Jpn. 21:325-335.
21. Galan, J. E., and D. Zhou. 2000. Striking a balance: modulation of the actin cytoskeleton by Salmonella. Proc. Natl. Acad. Sci. USA 97:8754-8761.
22. Glickman, M. S., J. S. Cox, and W. R. Jacobs, Jr. 2000. A novel mycolic acid cyclopropane synthetase is required for cording, persistence, and virulence of Mycobacterium tuberculosis. Mol. Cell 5:717-727.
23. Gray, J. R., and L. Rabeneck. 1989. Atypical mycobacterial infection of the gastrointestinal tract in AIDS patients. Am. J. Gastroenterol. 84:1521-1524.
24. Inderlied, C. B., C. A. Kemper, and L. E. Bermudez. 1993. The Mycobacterium avium complex. Clin. Microbiol. Rev. 6:266-310.
25. Jacobs, W. R., Jr., G. V. Kalpana, J. D. Cirillo, L. Pascopella, S. B. Snapper, R. A. Udani, W. Jones, R. G. Barletta, and B. R. Bloom. 1991. Genetic systems for mycobacteria. Methods Enzymol. 204:537-555.
26. Jungblut, P. R., D. Bumann, G. Haas, U. Zimny-Arndt, P. Holland, S. Lamer, F. Siejak, A. Aebischer, and T. F. Meyer. 2000. Comparative proteome analysis of Helicobacter pylori. Mol. Microbiol. 36:710-725.
27. Lucas, R. L., C. P. Lostroh, C. C. DiRusso, M. P. Spector, B. L. Wanner, and C. A. Lee. 2000. Multiple factors independently regulate hilA and invasion gene expression in Salmonella enterica serovar Typhimurium. J. Bacteriol. 182:1872-1882.
28. McGarvey, J. A., and L. E. Bermudez. 2001. Phenotypic and genomic analyses of the Mycobacterium avium complex reveal differences in gastrointestinal invasion and genomic composition. Infect. Immun. 69:7242-7249.
29. Menozzi, F. D., R. Bischoff, E. Fort, M. J. Brennan, and C. Locht. 1998. Molecular characterization of the mycobacterial heparin-binding hemagglutinin, a mycobacterial adhesin. Proc. Natl. Acad. Sci. USA 95:12625-12630.
30. Momotani, E., D. L. Whipple, A. B. Thiermann, and N. F. Cheville. 1988. Role of M cells and macrophages in the entrance of Mycobacterium paratuberculosis into domes of ileal Peyer's patches in calves. Vet. Pathol. 25:131-137.
31. Neutra, R. R., and J. P. Kraehenbuhl. 1992. Transepithelial transport and mucosal defence. I. The role of M cells. Trends Cell Biol. 2:134-138.
32. Orme, I. M., S. K. Furney, and A. D. Roberts. 1992. Dissemination of enteric Mycobacterium avium infections in mice rendered immunodeficient by thymectomy and CD4 depletion or by prior infection with murine AIDS retroviruses. Infect. Immun. 60:4747-4753.
33. Parker, A. E., and L. E. Bermudez. 1997. Expression of the green fluorescent protein (GFP) in Mycobacterium avium as a tool to study the interaction between mycobacteria and host cells. Microb. Pathog. 22:193-198.
33. Pepe, J., J. L. Badger, and V. L. Miller. 1994. Growth phase and low pH affect the thermal regulation of the Yersinia enterocolitica inv gene. Mol. Microbiology 11:123-135.
34. Reddy, V. M., and B. Kumar. 2000. Interaction of Mycobacterium avium complex with human respiratory epithelial cells. J. Infect. Dis. 181:1189-1193.
35. Sangari, F. J., J. Goodman, and L. E. Bermudez. 2000. Mycobacterium avium enters intestinal epithelial cells through the apical membrane, but not by the basolateral surface, activates small GTPase Rho and, once within epithelial cells, expresses an invasive phenotype. Cell Microbiol. 2:561-568.
36. Sangari, F. J., J. Goodman, M. Petrofsky, P. Kolonoski, and L. E. Bermudez. 2001. Mycobacterium avium invades the intestinal mucosa primarily by interacting with enterocytes. Infect. Immun. 69:1515-1520.
37. Schorey, J. S., M. A. Holsti, T. L. Ratliff, P. M. Allen, and E. J. Brown. 1996. Characterization of the fibronectin-attachment protein of Mycobacterium avium reveals a fibronectin-binding motif conserved among mycobacteria. Mol. Microbiol. 21:321-329.
38. Teitelbaum, R., W. Schubert, L. Gunther, Y. Kress, F. Macaluso, J. W. Pollard, D. N. McMurray, and B. R. Bloom. 1999. The M cell as a portal of entry to the lung for the bacterial pathogen Mycobacterium tuberculosis. Immunity 10:641-650.
39. Tran Van Nhieu, G., R. Bourdet-Sicard, G. Dumenil, A. Blocker, and P. J. Sansonetti. 2000. Bacterial signals and cell responses during Shigella entry into epithelial cells. Cell Microbiol. 2:187-193.
40. Utley, M., D. P. Franklin, K. A. Krogfelt, D. C. Laux, and P. S. Cohen. 1998. A Salmonella typhimurium mutant unable to utilize fatty acids and citrate is avirulent and immunogenic in mice. FEMS Microbiol. Lett. 163:129-134.
41. Wilson, M., J. DeRisi, H. H. Kristensen, P. Imboden, S. Rane, P. O. Brown, and G. K. Schoolnik. 1999. Exploring drug-induced alterations in gene expression in Mycobacterium tuberculosis by microarray hybridization. Proc. Natl. Acad. Sci. USA 96:12833-12838.
42. Yuan, Y., R. E. Lee, G. S. Besra, J. T. Belisle, and C. E. Barry III. 1995. Identification of a gene involved in the biosynthesis of cyclopropanated mycolic acids in Mycobacterium tuberculosis. Proc. Natl. Acad. Sci. USA 92:6630-6634.(Elizabeth Miltner, Kooros)
Department of Veterinary and Biomedical Sciences, University of Nebraska, Lincoln, Nebraska
Department of Biomedical Sciences, College of Veterinary Medicine, and Department of Microbiology, College of Sciences, Oregon State University, Corvallis, Oregon
ABSTRACT
Invasion of intestinal mucosa of the host by Mycobacterium avium is a critical step in pathogenesis and likely involves several different bacterial proteins, lipids, glycoproteins, and/or glycolipids. Through the screening of an M. avium genomic library in Mycobacterium smegmatis, we have identified a number of M. avium genes that are associated with increased invasion of mucosal epithelial cells. In order to further investigate these genes, we cloned six of them into a plasmid downstream of a strong mycobacterial promoter (L5 mycobacterial phage promoter), resulting in constitutive expression. Bacteria were then evaluated for increased expression and examined for invasion of HT-29 intestinal epithelial cells. The genes identified encode proteins that are similar to (i) M. tuberculosis coenzyme A carboxylase, (ii) M. tuberculosis membrane proteins of unknown function, (iii) M. tuberculosis FadE20, (iv) a Mycobacterium paratuberculosis surface protein, and (v) M. tuberculosis cyclopropane fatty acyl-phopholipid synthase. The constitutive expression of these genes confers to M. avium the ability to invade HT-29 intestinal epithelial cells with a severalfold increase in efficiency compared to both the wild-type M. avium and M. avium containing the vector alone. Using the murine intestinal ligated loop model, it was observed that the constitutive expression of M. avium proteins has a modest impact on the ability to enter the intestinal mucosa when compared with the wild-type control, suggesting that under in vivo conditions these genes are expressed at higher levels. Evaluation of the expression of these invasion-related genes indicated that under conditions similar to the intestinal lumen environment, the genes identified are upregulated. These data suggest that invasion of the intestinal mucosa is an event that requires the participation of several bacterial factors and the expression of the genes that encode them is less observed under standard laboratory growth conditions.
INTRODUCTION
A number of bacterial pathogens, including Salmonella, Yersinia, Shigella, and Escherichia coli, can utilize the intestine as a portal for entry during infection of humans. The molecular mechanisms used by these bacterial pathogens to invade and circumvent the natural defenses of the host are usually quite elegant and complex, involving a large number of bacterial genes (18, 21, 39). The factors encoded by these genes range from components of type III secretion apparatus (21) and outer membrane proteins (18) to enzymes involved in fatty acid biosynthesis (27, 40). Although the molecular mechanisms of invasion by Mycobacterium spp. have been studied to some extent in macrophages (16a), very little is known regarding their mechanisms of intestinal epithelial invasion. This paucity of information exists despite the fact that this route of infection is important for Mycobacterium (7, 23, 24, 32), and invasion of the respiratory epithelium has been proposed to play a role in pathogenesis of Mycobacterium tuberculosis (3, 20, 38).
Mycobacterium avium is an environmental organism that infects humans through both the intestinal tract and the respiratory tract (17, 24). This bacterium is considered a facultative intracellular pathogen that causes disease in both patients with and patients without AIDS. In contrast to infections in non-AIDS populations, where pulmonary infection is most prevalent, immunocompromised patients develop disseminated disease (15, 17, 24). In these individuals, both epidemiologic evidence and clinical evidence suggest that M. avium is predominantly acquired following ingestion of food or water (17, 24). The bacterium can resist acidic conditions in the stomach and colonize the intestinal tract (10, 23). Recent studies have demonstrated that M. avium interacts with the intestinal mucosa, entering primarily enterocytes (7, 35). Invasion of enterocytes appears to occur through the apical membrane rather than the basolateral surface (35). Therefore, it is likely that the abilities to invade and translocate across the intestinal mucosa are important components of M. avium pathogenesis. M. avium invasion of intestinal mucosal cells requires cytoskeleton rearrangements and protein phosphorylation in the host cell (9), indicating an active participation of the bacterium in the process by modulating host proteins. The genes and gene products that play a role in this stage of infection have not been identified.
Several genes and proteins have been suggested as participants in the invasion of intestinal epithelial cells by M. avium, such as fibronectin attachment protein (37), a 27-kDa protein (8), and superoxide dismutase (34). In addition, mce-1, a gene suggested to participate in Mycobacterium tuberculosis invasion of HeLa cells (1) and heparin binding hemagglutinin (29) are present in the M. avium genome. Although some of these proteins may have a role in interactions with the intestinal mucosa, it is likely that this process is multifactorial.
In order to obtain a better understanding of the genes associated with invasion of intestinal epithelial cells and the mechanism(s) of entry used, we set out to identify novel M. avium genes that increase the ability of M. smegmatis (a nonpathogenic mycobacterium) to invade intestinal epithelial cells.
MATERIALS AND METHODS
Strains and growth media. Mycobacterium smegmatis strain mc2155, selected for its high transformation efficiency (25), was grown in Middlebrook 7H9 broth with 0.1% Tween 80 or in Middlebrook 7H11 agar with 0.1% Tween 80. M. avium strains 101 and 104, isolated from the blood of AIDS patients, were cultured on 7H9 broth or 7H11 agar. E. coli XL1-Blue MRF' (Stratagene, LaJolla, CA) transformants were selected on Luria-Bertani agar and in Circlegrow broth (BIO 101) containing 50 μg/ml kanamycin. M. smegmatis transformants were selected on 7H11 agar with 0.1% Tween 80 and 50 μg/ml kanamycin.
Tissue culture. HEp-2 cells, a human epithelial cell line from the American Type Culture Collection (ATTC; CCL-23), Manassas, Virginia, were grown in RPMI 1640 supplemented with 10% fetal bovine serum and used at 80% confluence (approximately 1 x 106 cells) at 37°C under an atmosphere of 10% CO2. The HT-29 human intestinal cell line was also obtained from ATCC. The cells were grown in McCoy's medium (Gibco) supplemented with 10% fetal bovine serum in tissue culture plates and used when at 100% confluence (approximately 1 x 106 cells), as previously reported (9).
Construction of the MAC 101 genomic library. Genomic DNA was isolated from MAC 101 as described previously (28) and partially digested with the restriction enzyme Sau3A. After electrophoresis on a 1% agarose gel, fragments in the 3- to 6-kb range were recovered and cloned into the BamHI site on the mycobacterium-E. coli shuttle plasmid pMV261 (25). The genomic library was transformed into electroporation-competent E. coli and plated on Luria-Bertani agar containing kanamycin. In this manner, approximately 10,000 transformants were generated. Restriction analysis of plasmids isolated from E. coli transformants showed 80% contained insert DNA, the average size being 4 kb. The library in E. coli was stored in Circlegrow broth (Bio 101) plus 30% glycerol, and aliquots were plated for amplification of the library and plasmid extraction. The library DNA was then used to transform M. smegmatis, resulting in approximately 17,000 M. smegmatis transformants pooled in 7H9 broth with 30% glycerol.
Electroporation of M. smegmatis and E. coli. M. smegmatis competent cells were made following a protocol adapted from Bio-Rad and Jacobs et al. (25). The optical density at 600 nm of overnight cultures (log phase) of M. smegmatis was measured and determined to correspond to approximately 5 x 108 bacteria. The bacteria were incubated on ice for 1.5 h, harvested by centrifugation, and washed three times with cold 10% glycerol plus 0.1% Tween 80. Pellets were resuspended in 1 ml 10% glycerol plus 0.1% Tween 80, aliquoted, and frozen in an ice-ethanol bath for storage at –70°C. For electroporation, 100 to 200 μl of cells was combined with a 1- to 5-μg library DNA and transferred to a prechilled 0.2-cm cuvette. Electroporation was carried out using a Gene Pulser (Bio-Rad, Hercules, CA) at the following settings: 25 μF, 1,000 , 2.5 kV. Cells were recovered with 1 ml 7H9 and incubated at 37°C with shaking for 2 h. Electrocompetent E. coli XL1-Blue MRF' cells were either purchased from Stratagene or prepared and transformed as previously described (33).
Invasion assay. Inocula for invasion assays were prepared as follows. M. smegmatis containing the library DNA was recovered from 7H11 agar and suspended in Hanks' balanced salt solution (HBSS), diluted to match McFarland standard 2 (approximately 6 x 108 CFU/ml) (7). The suspensions were passed 10 times through a 26-gauge needle, and large aggregates were allowed to settle. After 5 min, an aliquot was taken from the top half of the bacterial suspension and diluted in HBSS to 107 CFU/ml. One hundred microliters of this suspension was used to infect HEp-2 monolayers. Actual inoculum concentrations were determined by plating serial dilutions as described previously (9). Microscopic examination of inoculum using the LIVE/DEAD assay (Molecular Probes, Eugene, OR) showed the suspensions to contain single, viable cells (5).
The invasion assays were carried out as previously reported (9). HEp-2 and HT-29 monolayers were grown to 80% and 100% confluence (approximately 1 x 106 HEp-2 and HT-29 cells), respectively, in 24-well tissue culture plates (Costar, Pleasanton, CA). Prior to the assay, the culture medium was removed and replaced with fresh medium. Each well was inoculated with 106 bacteria, and the plate was incubated at 37°C in a 5% CO2 incubator for 2 h. Extracellular bacteria were removed by washing monolayers three times with HBSS. Any remaining extracellular bacteria were killed by replacing the medium with medium that contained 200 μg/ml amikacin and incubating for 2 h at 37°C (9). Monolayers were washed three times with HBSS after amikacin treatment and lysed with 1 ml sterile water. The number of viable intracellular bacteria was determined by plating lysates of duplicate wells on 7H11 agar containing 50 μg/ml kanamycin. In some experiments, bacteria were exposed to 0.3 M NaCl (final concentration), anaerobiosis (anaerobic jar), or both for 4 h prior to the assay as previously reported (6). Clones with a percentage of invasion of 1.5 or more over a background of 0.05% (M. smegmatis) were chosen for further study. A control strain of M. smegmatis with the skeleton plasmid pMV261 was used in some assays.
In some assays, infected monolayers were fixed with 2% paraformaldehyde for 30 min at room temperature, washed, and then stained for acid-fast bacilli using the Runyoun method as described previously (9). Intracellular bacteria were counted under the light microscope, and the average number of bacteria per cell and the percentage of infected cells were determined.
Nucleotide sequence and data analysis. M. smegmatis clones that display increased invasion were sequenced at the University of California, San Francisco Biomolecular Resource Center using an Applied Biosystems DNA sequencer and at the University of Nebraska, Lincoln. Sequence comparisons were performed using BLAST 2.0 from the National Center for Biotechnology Information. Sequence data for the M. avium genome were obtained from The Institute for Genomic Research (TIGR; http://www.tigr.org).
Construction of subclones. Clones from the MAC 101 library were further examined by subcloning the insert region of DNA into pFJS8 for overexpression from the mycobacteriophage L5 promoter. The resulting constructs were then transformed into MAC 104 for invasion assays, and RNA dot blots were done to confirm expression of the cloned genes.
The plasmid pFJS8 is an E. coli-mycobacterium shuttle plasmid containing the L5 promoter added for pJ5F8 conduction. The plasmid was constructed as follows. The hsp60 promoter was cut off the plasmid pMV261 (25) by using the enzymes PvuII and DraI, making the plasmid pFJS3. Then pFJS3 was cut with SalI and ClaI, and the mycobacteriophage L5 promoter was cloned in it. In order to subclone each region of interest, sequence data from each of the clones displaying increased invasion were used to design primers to amplify the corresponding region from MAC 101 genomic DNA by PCR. PCR products were ligated directly into pCR2.1 (Invitrogen, Carlsbad, CA) and screened by digestion to determine the direction of the insert. The inserts were then digested from pCR2.1 using a HindIII/NsiI double digest and cloned directionally into pFJS8 cut with HindIII and PstI. All subclones except pEMC7 were constructed in this manner. To construct pEMC7, the homolog to the M. tuberculosis gene Rv3802c was PCR amplified from MAC 101 using primers Rv3802c(NisI)/L (GGATGCATCGGGCCCGCGACAAAG) and Rv3802c(HindIII)U (GGGCAAGCTTGTACCGTGTAGAGC) containing the restriction sites HindIII and NsiI and cloned directly into the vector cut with HindIII and PstI. The primers clone342L (AGGTGGGCAAGGCGCTCAA) and clone342U (CGACGCGGAGAAATACGGTCT) were used amplify the MAC 101 homolog to the M. paratuberculosis protein AF232751, producing construct pEMC3. Primers clone335L (CGAGGCGTACAACACCGTCA) and clone 335U (GCTGCTGCAGAACGAGTTGGG) were used to clone the Rv3720 homolog, creating pEMC4. Plasmid pEMC8 was created as a control, with the Rv3720 gene cloned in opposite orientation to the L5 promoter. Primers fadE20L (CCGGCCGCATCACTCGAT) and fadE20U (GTAGGCCATGCGACGCTCGTT) were used to clone the fadE20 homolog, creating pEMC5. Primers clone 23ORF/L (GGCGGTGACGACGGTGAATCC) and clone 23ORF/U (CGGCCAGCTCCTTGCGGTAG) were used to create pEMC6. Primers L (CTCACCGATCGCTACGTC) and U (GTCCGGATCCGCTACCGC) were used to create pEMC9.
Regulation of invasion genes. To determine whether the genes identified are upregulated when the bacterium is exposed to conditions that model those encountered in the intestinal lumen, 107 M. avium cells were cultured in 7H9 broth, and after the culture reached the turbidity equivalent to 5 x 108 organisms, it was split and placed at 37°C under hyperosmolarity conditions (0.3 M) (6, 33a), anaerobiosis, or a combination of both conditions, as well as under iso-osmolar and aerobic conditions as a control, for 24 h. Total bacterial RNA was obtained as previously reported (41). Total RNA was initially quantified by absorbance at 260 nm, and quality was determined by the 260/280-nm absorbance ratio. Ratios of 1.8 were considered acceptable. RNA was then electrophoresed on an agarose gel to confirm quality. RNA was submitted to reverse transcriptase treatment to obtain cDNA. Briefly, 6 μl of RNA, 6 μl of random hexamer (50 mg/3 μl), 2 μl of deoxynucleoside triphosphate (dNTP) mix (10 mM), and distilled water were mixed and incubated at 65°C for 5 min and subsequently placed on ice for 2 min (reverse transcription [RT] mix). Then to 18 μl of the RT mix we added 5 μl of 10x RT buffer, 8 μl of MgCL2 (25 mM), and dithiothreitol (0.1 M). The mix was incubated at 25°C for 2 min, and 2 μl of Super Script RTII was added to the sample. The sample was incubated for 10 min at 25°C and then transferred to 42°C for 50 min. The reaction mixture was then placed on ice and centrifuged briefly, and 2 μl of RNase H was added at 37°C for 20 min. The sigA gene was used as the constitutively expressed control.
To amplify the cDNA, we used the specific primers for the invasion-associated genes used to clone the genes in pFJS8. PCR amplification was carried out at 95°C for 3 min (1 cycle), 95°C for 3 s, 62°C for 30 s, and 72°C for 2 min in a linear range (35 cycles) and 72°C for 10 min (1 cycle). Equivalent amounts of RNA were used for the RT-PCR. The products obtained by PCR amplification were quantified by using the Kodak EDEAS 290 system and the Kodak id image analysis software (Eastman Kodak Company, Rochester, NY).
Infection using the intestinal loop model. To determine whether the M. avium clones that displayed enhanced invasion in vitro were also associated with increased invasion of intestinal mucosa in vivo, we used the murine intestinal loop model. Bacteria (107) were injected into an intestinal loop of anesthetized mice as previously described (36). After 1 h, the mice were harvested, the intestinal loop was removed and cut open, and bacteria loosely associated with the mucosa were removed by washing. The tissue was then homogenized, and the viable bacteria were plated onto 7H11 agar as described previously (36). The results were calculated as the percentage of the initial inoculum that is present in the tissue, and the experiments were repeated three times.
Statistical analysis. The results shown represent the mean ± standard deviation of at least five experiments. The data were analyzed statistically by comparison with controls at the same time points using the Student's t test. P < 0.05 was considered statistically significant.
RESULTS
Enrichment of M. avium library and identification of invasive clones. Table 1 shows that consecutive passages of the pool of M. smegmatis clones containing M. avium DNA in HEp-2 cells enriched for a more invasive population of bacteria, whereas, this phenomenon was not observed in the wild-type M. smegmatis. After the third passage, the percentage of inoculum that entered HEp-2 cells increased from 0.07% to 4.3%. The cell lysate from this infection was then plated to obtain individual colonies, and they were each evaluated for invasion of HEp-2 cells. Eleven clones that displayed an efficiency of invasion greater than 1.5% (over a background of 0.05%, P < 0.05 as compared to the control; Table 2) were further characterized by purification of the plasmid that they carried and sequencing of the M. avium DNA insert. One gene was represented three times in the pool (fadE20), and three other genes were represented twice each (Af232751, and the homologues to Rv3802c and Rv3720). Since some of the invasion-related clones contained more than one gene (two genes were present in two of the clones), individual genes were cloned into M. smegmatis and screened for the ability to confer enhanced invasion. Each of the six individually subcloned genes conferred comparable levels of invasion to that observed with the original selected clone, indicating that these genes were, in fact, the ones associated with the enhanced invasion phenotype. All of the genes identified have homologues in other mycobacterial species, and most of them are present in M. tuberculosis (Table 2). These results also showed that M. avium genes associated with an increase in M. smegmatis invasion possibly belonged to two groups: surface proteins and enzymes involved in fatty acid synthesis and degradation.
M. smegmatis clones expressing M. avium genes did not show changes in colony morphology. In contrast, M. avium clones expressing constitutively the homologues to Rv3719 and Rv3720 showed significant change in the smoothness of the colony, which would suggest increased quantity of lipids on the cell wall.
Expression of invasion genes under different environmental conditions. Our past observation that M. avium invades both the intestinal mucosa, as well as intestinal epithelial cells, more efficiently under conditions that mimic the intestinal environment (6) suggests that invasion-related genes might be upregulated as a result of intestinal conditions (6). Therefore, if the six genes that we have identified in the current study are similar to those involved in the regulated invasion phenotype we observed previously, one would expect them to be repressed under standard laboratory conditions and upregulated under conditions that mimic those in the intestine, high osmolarity and low oxygen tension (6, 33a). We exposed M. avium 104 (MAC 104) to either low oxygen tension, high osmolarity, or both and performed RT-PCR to evaluate the expression of the invasion genes we identified. As shown in Fig. 1, while open reading frame 23 (carboxylase, pEMC6) was not expressed under laboratory conditions, it was highly expressed under conditions of high osmolarity as well as low oxygen tension. Interestingly, the high level of expression observed under low oxygen tension was also observed when the bacterium was exposed to both conditions simultaneously. The fadE20 (pEMC5) gene had an increased level of expression under low oxygen tension and under high-osmolarity conditions. The gene homologous to Rv3802 (clone 332, pEMC7) showed increased expression under both high-osmolarity and low-oxygen conditions. The gene homologous to Rv3720 (pEMC4) had increased expression under low oxygen tension, and the Af232751 (pEMC3) homologue showed increased expression under both conditions alone and when the conditions were combined. The sodC gene was used as a control because preliminary experiments showed that its expression is not influenced by any of the conditions used (data not shown). These data support the possibility that our selection procedure allowed identification of genes that are involved in invasion of intestinal epithelium because they are upregulated by conditions that mimic those found in the intestinal environment.
Constitutive expression enhances invasion of epithelial cells by M. avium. Although these genes confer enhanced entry to M. smegmatis, it remains possible that in M. avium they do not serve the same function. Our previous studies indicate that invasion of epithelial cells does not occur at very high levels in wild-type M. avium under standard laboratory conditions, but this phenotype can be induced by specific environmental conditions (5). Based on these data, we hypothesized that upregulation of invasion-associated genes would increase the ability to enter intestinal epithelial cells. Thus, if the genes identified are the same as those responsible for the repressed invasion phenotype, wild-type M. avium would be phenotypically similar to a mutant in these genes under standard laboratory growth conditions and the biological role of these genes could then be easily observed by expressing them under these conditions using a constitutive mycobacterial promoter. We have developed a vector that allows constitutive expression of genes from a promoter identified from the mycobacteriophage L5 (2a). In this controlled expression system, the gene being investigated would be cloned downstream of the L5 promoter and transformed into wild-type M. avium. Plasmids pEMC3 (Af232751 homologue), pEMC4 (Rv3720 homologue), pEMC5 (fadE20), pEMC6 (aacD4 homologue), pEMC7 (Rv3802 homologue), pEMC8 (Rv3720 in wrong orientation), and pEMC9 (Rv3273 homologue) were then created and transformed into the wild-type M. avium strain 104. These clones were then evaluated in invasion assays to determine if the constitutive expression of the identified genes is associated with an increased ability to enter epithelial cells. As shown in Fig. 2, all but MAC 104-pEMC8, which has the gene in the opposite orientation to the promoter, display an increased ability to enter epithelial cells. MAC 104-pEMC3, -pEMC4, -pEMC5, -pEMC6, -pEMC7, and -pEMC9 are associated with a significant increase in the ability to enter HT-29 cells as compared to the wild-type M. avium 104. These results were confirmed by light microscopy (Table 3).
Constitutive expression enhances invasion of intestinal epithelium by M. avium. Since the M. avium transformants with the M. avium invasion-related genes expressed constitutively were able to invade HT-29 intestinal cells in vitro with increased efficiency compared to the wild-type strain, we examined their ability to invade the intestinal mucosa in vivo using the mouse intestinal loop model of infection. We chose pEMC4 and pEMC6 for these studies because of their relatively high levels of invasion when expressed constitutively as compared to the other genes identified. M. avium 104-pEMC4 and M. avium MAC 104-pEMC6 were injected in the intestinal loop, and after 1 h, the percentage of the inoculum that invaded the intestine was determined. The results (Table 4) indicate that in vivo the constitutive expression of clone 23 (carboxylase) and the gene homologous to Rv3720 (cyclopropane-fatty acid-phospholipid synthase) had a significant impact on the ability of the bacterium to invade the intestinal mucosa. To determine if exposure to high-osmolarity conditions in vitro would have a similar effect on invasion, we compared invasion of HT-29 cell monolayers by clones exposed to high osmolarity (6, 33a) and to "laboratory conditions" (isotonic). As shown in Table 4, clones that express the "invasion-associated genes" enter HT-29 cells with a significantly higher efficiency than control M. avium cells grown under standard laboratory conditions prior to the assay. Incubation under high-osmolarity conditions increases the ability of wild-type M. avium to enter epithelial cells but had a limited impact on the entry of clones expressing invasion genes constitutively. These data suggest that the level of expression of the "invasion-associated genes" correlates with M. avium invasion of the intact intestinal epithelium as well as intestinal epithelial cells in vitro.
DISCUSSION
Similar to other enteric bacterial pathogens, invasion of the intestinal epithelium by M. avium is multifactorial. We have identified six M. avium genes that are associated with invasion of intestinal epithelial cells in vitro. We also found that these genes are transcribed at higher levels when the bacteria are exposed to conditions that are thought to be similar to those in the intestinal environment, increased osmolarity and/or low oxygen tension.
The sequences of the six genes identified predict for two types of proteins: membrane proteins of unknown function and enzymes involved in fatty acid synthesis, degradation, or transport. Some of the genes involved in fatty acid synthesis or degradation have some degree of homology in the M. smegmatis genome (fadE20), while others do not, but none of the genes coding for proteins has homologues in M. smegmatis (TIGR database). The M. avium fadE20 gene is quite different from the M. smegmatis gene (43% homology), which might help to explain its role in the invasion of intestinal epithelium.
Among the genes, fadE20 is a homologue of a M. tuberculosis acyl coenzyme A (acyl-CoA) dehydrogenase, a participant in the fatty acid degradation/synthesis pathway. The role of mycolic acid and other fatty acids in mycobacterial adherence may be related to their influence on the assembly of the cell wall. Fatty acids may facilitate binding in a nonspecific manner by influencing the glycolipid content or lipid ratios. Alternatively, mycolic acids and enzymes involved in their synthesis or degradation may regulate the presence of other protein or sugar components on the bacterial surface. Alteration of the fatty acid composition of the mycobacterial surface would likely affect the hydrophobic and electrostatic character of the bacteria, thereby, affecting the ability of the bacteria to bind host cell membranes and/or receptors. However, studies by C. Lee and colleagues demonstrated a regulatory role for FadD during invasion by Salmonella (27). The mycobacterial fadE gene identified for its effects on invasion in the current study may have a similar regulatory function. Interestingly, FadR and FadF have been reported to be part of the regulatory system in Salmonella and have been shown to repress hilA, a major regulatory gene of the pathogenicity island I (40). Furthermore, the importance of these fatty acid degradation and synthesis pathways in the pathogenic mechanism is suggested by two recent reports where investigators determined that fadD26 and fadD28 are involved in the virulence of Mycobacterium tuberculosis in vivo (11, 19). Mutations in both genes alter colony morphology as a result of interference with mycolic acid synthesis. Based on the M. avium genome sequence, the fadE20 gene is with a putative operon where the gene just upstream is hflx, a GTP-binding protein. GTP-binding proteins are often associated with regulatory genes and provide energy for their activity. Although speculative, the presence of fadE20 downstream of this gene is also suggestive of a regulatory role for this gene in the invasion of intestinal epithelium. It is for these reasons we consider it more likely that fadE20 plays a regulatory role in M. avium invasion, but further studies are necessary to definitively differentiate this hypothesis from the equally likely possibility of direct effects on invasion through changes in fatty acid composition of the mycobacterial cell surface.
Rv3720 is a cyclopropane fatty acyl-phospholipid synthase that is related to -mycolic acid synthesis and is likely to affect cell wall structure. M. tuberculosis, for example, synthesizes three major types of cyclopropanated mycolic acids (42), and both M. tuberculosis and M. avium have a large number of cyclopropane synthases in their genomes (12; TIGR database, unpublished) which suggests a number of potential functions for cyclopropanate mycolic acids. Recently, inactivation of a gene coding for a cyclopropane synthase, pcaA, was reported to attenuate the virulence of Mycobacterium tuberculosis (22). Our data suggest that cyclopropanated mycolic acid has a role in host cell entry. The mechanism by which cyclopropanated mycolic acid participates in invasion is unclear but is likely to involve fundamental changes in the nature of the bacterial surface that comes in contact with the host cell. The presence of this gene within a group of mycobacterial invasion genes further supports the importance of the lipid composition on the mycobacterial surface during entry into host cells.
In M. paratuberculosis, the accession number AF232752 is for a 35-kDa protein that is a surface protein and that appears to be expressed at higher levels under hyperosmolar conditions as compared to standard laboratory growth conditions (2). M. paratuberculosis cells have been shown to cross the intestinal mucosa by entering M cells in the terminal ileum (30), while M. avium interacts primarily with enterocytes and, to a lesser extent, with M cells (36). It is possible that M. paratuberculosis can cross the intestinal mucosa by either or both pathways, M cells and enterocytes. However, M. paratuberculosis transmission is assumed to occur very early in life, and calves are believed to be infected at birth or soon afterwards. If this assumption is correct, the young calf is likely to not have well-developed Peyer's patches and M cells. Several laboratories have demonstrated that Peyer's patches only appear after some degree of maturation of the intestines and exposure to "contaminated" contents (31).
Two of the other membrane proteins identified are similar to M. tuberculosis proteins, Rv2724c (62% identity and 48% similarity) and Rv3723 (74% identity and 76% similarity). M. tuberculosis is not known to invade the intestinal mucosa, but intestinal tuberculoma can be seen in patients who do not possess a functional stomach barrier (14). Furthermore, M. tuberculosis, in contrast to M. avium, is not tolerant of the acid pH of the stomach (14). Therefore, M. tuberculosis may use similar machinery to invade respiratory epithelial cells (18). Another possible explanation is that the M. avium homologue to Rv3723 is in fact a different protein (only 34% identical to the M. tuberculosis protein) and evolved a different function. Similar modification of existing proteins for novel functions has been previously observed in many other microorganisms (26).
The Rv3802c homologue is a probable membrane protein that contains an N-terminal signal sequence followed by a proline-rich region. These are characteristics encountered in internalin (inlA), a Listeria monocytogenes protein associated with the invasion of epithelial cells (13). Our data indicate that M. avium Rv3802c also has a role in invasion. If this protein has a similar function to that of internalin, in terms of interaction with host cells, its identification would represent an important breakthrough in our understanding of M. avium invasion, but definitive demonstration of this role will require additional more detailed analysis of the role of this specific gene during interactions with host cells.
The application of a system for exogenously controlled expression of genes offers a useful strategy to screen for virulence genes that are only expressed well or at all under conditions found in the host. The previous observation that M. avium exposed to high osmolarity or anaerobiosis had an increased ability to invade intestinal epithelial cells (5) suggests that, under conditions encountered in the intestinal tract, M. avium genes that participate in the invasion process are upregulated. Due to these findings, we hypothesized that invasion-related proteins or lipids would be expressed at a higher level in vivo. Our results support our hypothesis, and only one among the originally identified genes in M. smegmatis was not associated with increased invasion when expressed constitutively. The fact that a couple of the genes that display increased expression when bacteria were exposed to either low oxygen tension or high osmolarity are suppressed when they are exposed simultaneously to both conditions suggests that the bacterium may not be exposed to both of these environmental conditions at the same time in vivo.
Several other genes have been associated with Mycobacterium invasion of epithelial cells, such as fibronectin attachment proteins (to bladder epithelial cells), mce-1 (M. tuberculosis binding to HeLa cells), the M. avium 27-kDa protein (8), and M. tuberculosis hemagglutinin (29) in the binding to alveolar epithelial cells. M. avium binding and invasion of epithelial intestinal cells in vivo appear to require an array of moieties to interact with mucin, microvilli, and membranes of the host cell, triggering cytoskeletal reorganization (9), activation of RhoA (35), and protein phosphorylation (9).
The methods employed in this paper have limitations, although they are the best that can be achieved with M. avium at this point. We have developed genetic systems for M. avium, and future studies will address the importance of the identified genes to invasion.
ACKNOWLEDGMENTS
This work was supported by the grant AI-43199 from the National Institute of Allergy and Infectious Diseases.
We thank Karen Allen and Denny Weber for help in the preparation of the manuscript.
REFERENCES
1. Arruda, S., G. Bomfim, R. Knights, T. Huima-Byron, and L. W. Riley. 1993. Cloning of an M. tuberculosis DNA fragment associated with entry and survival inside cells. Science 261:1454-1457.
2. Bannantine, J. P., J. F. J. Huntley, E. Miltner, J. R. Stabel, and L. E. Bermudez. 2003. Mycobacterium avium subsp paratuberculosis 35 KDa protein plays a role in invasion of bovine epithelial cells. Microbiology 149:2061-2069.
2. Barletta, R. G., D. D. Kim, S. B. Snapper, B. R. Bloom, and J. W. R. Jacobs. 1992. Identification of expression signals of the mycobacteriophages Bxb1, L1 and TM4 using Escherichia-Mycobacterium shuttle plasmids pYUB75 and pYUB76 designed to create translational fusions to the lacZ gene. J. Gen. Microbiol. 138:23-30.
3. Bermudez, L. E., and J. Goodman. 1996. Mycobacterium tuberculosis invades and replicates within type II alveolar cells. Infect. Immun. 64:1400-1406.
4. Reference deleted.
5. Bermudez, L. E., A. Parker, and J. R. Goodman. 1997. Growth within macrophages increases the efficiency of Mycobacterium avium in invading other macrophages by a complement receptor-independent pathway. Infect. Immun. 65:1916-1925.
6. Bermudez, L. E., M. Petrofsky, and J. Goodman. 1997. Exposure to low oxygen tension and increased osmolarity enhance the ability of Mycobacterium avium to enter intestinal epithelial (HT-29) cells. Infect Immun. 65:3768-3773.
7. Bermudez, L. E., M. Petrofsky, P. Kolonoski, and L. S. Young. 1992. An animal model of Mycobacterium avium complex disseminated infection after colonization of the intestinal tract. J. Infect. Dis. 165:75-79.
8. Bermudez, L. E., K. Shelton, and L. S. Young. 1995. Comparison of the ability of Mycobacterium avium, M. smegmatis and M. tuberculosis to invade and replicate within HEp-2 epithelial cells. Tuber. Lung Dis. 76:240-247.
9. Bermudez, L. E., and L. S. Young. 1994. Factors affecting invasion of HT-29 and HEp-2 epithelial cells by organisms of the Mycobacterium avium complex. Infect. Immun. 62:2021-2026.
10. Bodmer, T., E. Miltner, and L. E. Bermudez. 2000. Mycobacterium avium resists exposure to the acidic conditions of the stomach. FEMS Microbiol. Lett. 182:45-49.
11. Camacho, L. R., D. Ensergueix, E. Perez, B. Gicquel, and C. Guilhot. 1999. Identification of a virulence gene cluster of Mycobacterium tuberculosis by signature-tagged transposon mutagenesis. Mol. Microbiol. 34:257-267.
12. Cole, S. T., R. Brosch, J. Parkhill, T. Garnier, C. Churcher, D. Harris, S. V. Gordon, K. Eiglmeier, S. Gas, C. E. Barry III, F. Tekaia, K. Badcock, D. Basham, D. Brown, T. Chillingworth, R. Connor, R. Davies, K. Devlin, T. Feltwell, S. Gentles, N. Hamlin, S. Holroyd, T. Hornsby, K. Jagels, B. G. Barrell et al. 1998. Deciphering the biology of Mycobacterium tuberculosis from the complete genome sequence. Nature. 393:537-544.
13. Cossart, P., and M. Lecuit. 1998. Interactions of Listeria monocytogenes with mammalian cells during entry and actin-based movement: bacterial factors, cellular ligands and signaling. EMBO J. 17:3797-3806.
14. Curtiss, K. J., and M. H. Sleisenger. 1978. Infectious and parasitic diseases, p. 348. In M. H. Sleisenger and J. S. Fordtran (ed.), Gastrointestinal disease, Philadelphia, Pa.
15. Damsker, B., and E. J. Bottone. 1985. Mycobacterium avium-Mycobacterium intracellulare from the intestinal tracts of patients with the acquired immunodeficiency syndrome: concepts regarding acquisition and pathogenesis. J. Infect. Dis. 151:179-181.
16. Reference deleted.
16. Ernst, J. D. 1998. Macrophage receptor for Mycobacterium tuberculosis. Infect. Immun. 66:1277-1279.
17. Falkinham, J. O., III. 1996. Epidemiology of infection by nontuberculous mycobacteria. Clin. Microbiol. Rev. 9:177-215.
18. Finlay, B. B., and S. Falkow. 1997. Common themes in microbial pathogenicity revisited. Microbiol. Mol. Biol. Rev. 61:136-169.
19. Fitzmaurice, A. M., and P. E. Kolattukudy. 1997. Open reading frame 3, which is adjacent to the mycocerosic acid synthase gene, is expressed as an acyl coenzyme A synthase in Mycobacterium bovis BCG. J. Bacteriol. 179:2608-2615.
20. Fujimura, Y. 1986. Functional morphology of microfold cells (M cells) in Peyer's patches—phagocytosis and transport of BCG by M cells into rabbit Peyer's patches. Gastroenterol. Jpn. 21:325-335.
21. Galan, J. E., and D. Zhou. 2000. Striking a balance: modulation of the actin cytoskeleton by Salmonella. Proc. Natl. Acad. Sci. USA 97:8754-8761.
22. Glickman, M. S., J. S. Cox, and W. R. Jacobs, Jr. 2000. A novel mycolic acid cyclopropane synthetase is required for cording, persistence, and virulence of Mycobacterium tuberculosis. Mol. Cell 5:717-727.
23. Gray, J. R., and L. Rabeneck. 1989. Atypical mycobacterial infection of the gastrointestinal tract in AIDS patients. Am. J. Gastroenterol. 84:1521-1524.
24. Inderlied, C. B., C. A. Kemper, and L. E. Bermudez. 1993. The Mycobacterium avium complex. Clin. Microbiol. Rev. 6:266-310.
25. Jacobs, W. R., Jr., G. V. Kalpana, J. D. Cirillo, L. Pascopella, S. B. Snapper, R. A. Udani, W. Jones, R. G. Barletta, and B. R. Bloom. 1991. Genetic systems for mycobacteria. Methods Enzymol. 204:537-555.
26. Jungblut, P. R., D. Bumann, G. Haas, U. Zimny-Arndt, P. Holland, S. Lamer, F. Siejak, A. Aebischer, and T. F. Meyer. 2000. Comparative proteome analysis of Helicobacter pylori. Mol. Microbiol. 36:710-725.
27. Lucas, R. L., C. P. Lostroh, C. C. DiRusso, M. P. Spector, B. L. Wanner, and C. A. Lee. 2000. Multiple factors independently regulate hilA and invasion gene expression in Salmonella enterica serovar Typhimurium. J. Bacteriol. 182:1872-1882.
28. McGarvey, J. A., and L. E. Bermudez. 2001. Phenotypic and genomic analyses of the Mycobacterium avium complex reveal differences in gastrointestinal invasion and genomic composition. Infect. Immun. 69:7242-7249.
29. Menozzi, F. D., R. Bischoff, E. Fort, M. J. Brennan, and C. Locht. 1998. Molecular characterization of the mycobacterial heparin-binding hemagglutinin, a mycobacterial adhesin. Proc. Natl. Acad. Sci. USA 95:12625-12630.
30. Momotani, E., D. L. Whipple, A. B. Thiermann, and N. F. Cheville. 1988. Role of M cells and macrophages in the entrance of Mycobacterium paratuberculosis into domes of ileal Peyer's patches in calves. Vet. Pathol. 25:131-137.
31. Neutra, R. R., and J. P. Kraehenbuhl. 1992. Transepithelial transport and mucosal defence. I. The role of M cells. Trends Cell Biol. 2:134-138.
32. Orme, I. M., S. K. Furney, and A. D. Roberts. 1992. Dissemination of enteric Mycobacterium avium infections in mice rendered immunodeficient by thymectomy and CD4 depletion or by prior infection with murine AIDS retroviruses. Infect. Immun. 60:4747-4753.
33. Parker, A. E., and L. E. Bermudez. 1997. Expression of the green fluorescent protein (GFP) in Mycobacterium avium as a tool to study the interaction between mycobacteria and host cells. Microb. Pathog. 22:193-198.
33. Pepe, J., J. L. Badger, and V. L. Miller. 1994. Growth phase and low pH affect the thermal regulation of the Yersinia enterocolitica inv gene. Mol. Microbiology 11:123-135.
34. Reddy, V. M., and B. Kumar. 2000. Interaction of Mycobacterium avium complex with human respiratory epithelial cells. J. Infect. Dis. 181:1189-1193.
35. Sangari, F. J., J. Goodman, and L. E. Bermudez. 2000. Mycobacterium avium enters intestinal epithelial cells through the apical membrane, but not by the basolateral surface, activates small GTPase Rho and, once within epithelial cells, expresses an invasive phenotype. Cell Microbiol. 2:561-568.
36. Sangari, F. J., J. Goodman, M. Petrofsky, P. Kolonoski, and L. E. Bermudez. 2001. Mycobacterium avium invades the intestinal mucosa primarily by interacting with enterocytes. Infect. Immun. 69:1515-1520.
37. Schorey, J. S., M. A. Holsti, T. L. Ratliff, P. M. Allen, and E. J. Brown. 1996. Characterization of the fibronectin-attachment protein of Mycobacterium avium reveals a fibronectin-binding motif conserved among mycobacteria. Mol. Microbiol. 21:321-329.
38. Teitelbaum, R., W. Schubert, L. Gunther, Y. Kress, F. Macaluso, J. W. Pollard, D. N. McMurray, and B. R. Bloom. 1999. The M cell as a portal of entry to the lung for the bacterial pathogen Mycobacterium tuberculosis. Immunity 10:641-650.
39. Tran Van Nhieu, G., R. Bourdet-Sicard, G. Dumenil, A. Blocker, and P. J. Sansonetti. 2000. Bacterial signals and cell responses during Shigella entry into epithelial cells. Cell Microbiol. 2:187-193.
40. Utley, M., D. P. Franklin, K. A. Krogfelt, D. C. Laux, and P. S. Cohen. 1998. A Salmonella typhimurium mutant unable to utilize fatty acids and citrate is avirulent and immunogenic in mice. FEMS Microbiol. Lett. 163:129-134.
41. Wilson, M., J. DeRisi, H. H. Kristensen, P. Imboden, S. Rane, P. O. Brown, and G. K. Schoolnik. 1999. Exploring drug-induced alterations in gene expression in Mycobacterium tuberculosis by microarray hybridization. Proc. Natl. Acad. Sci. USA 96:12833-12838.
42. Yuan, Y., R. E. Lee, G. S. Besra, J. T. Belisle, and C. E. Barry III. 1995. Identification of a gene involved in the biosynthesis of cyclopropanated mycolic acids in Mycobacterium tuberculosis. Proc. Natl. Acad. Sci. USA 92:6630-6634.(Elizabeth Miltner, Kooros)