OUP user menu

Identification of Mycobacterium avium genes up-regulated in cultured macrophages and in mice

Lia Danelishvili , Melanie J. Poort , Luiz E. Bermudez
DOI: http://dx.doi.org/10.1016/j.femsle.2004.08.014 41-49 First published online: 1 October 2004

Abstract

To investigate Mycobacterium avium gene expression upon infection of macrophages, we created a M. avium-promoter library upstream of a promoter-less gene encoding the green fluorescent protein (GFP) in Mycobacterium smegmatis. Clones were evaluated for increased expression of GFP after infection of U937 macrophages. A number of M. avium genes were up-regulated more than 3-fold after 24 and 48 h following macrophage infection. M. avium genes expressed by M. smegmatis during growth in macrophages include genes encoding transport/binding proteins, synthesis, modification and degradation of macromolecules, and a great majority of genes for which no function is currently known. For some of the unknown genes, homologues were identified in bacteria such as Mycobacterium leprae, Salmonella typhimurium and Agrobacterium tumefaciens.

In order to investigate if these genes were also expressed in M. avium during macrophage infection in vitro and in vivo, transcripts of selected genes were quantified using real time RT-PCR. Evaluation of most expressed genes in M. smegmatis confirmed their up-regulation in M. avium after 24 h infection of macrophages in vitro and mice.

Keywords
  • Mycobacterium avium
  • GFP library
  • Macrophage
  • Gene expression
  • Real time RT-PCR

1 Introduction

Mycobacterium avium is an environmental organism encountered in water and soil. The bacterium has been isolated from birds, swine, cattle, and non-human primates. M. avium enters the host through the gastrointestinal and respiratory tracts. It is a major opportunistic bacterial pathogen, causing bacteremia and disseminated disease in patients in the advanced stages of AIDS. In non-AIDS cases, M. avium causes pulmonary infection in individuals with predisposing lung conditions such as pneumococcosis, silicosis, cured tuberculosis, and chronic obstructive lung disease [1,2].

M. avium is phagocytosed by macrophages following entry into the host. Recently, it has been shown that M. avium uptake by monocytes and macrophages is associated with the presence of integrins, such as complement receptors (CR3, and CR4) [3,4], the vitronectin receptor [5], as well as with the mannose receptor [3]. Mycobacteria also bind fibronectin by using bacterial molecules, such as the antigen 85 complex (30–31 KDa proteins) [6]. Once within macrophages, M. avium resides in a membrane-bound vacuole that does not acidify and does not express proton pump ATPase [7,8]. Intracellular bacteria manipulate signaling pathways of the host cell in order to create a suitable microenvironment for survival and replication within the macrophage. The bacterial genes encoding proteins that play key roles in the macrophage phase of infection are mostly unknown.

A few studies have addressed M. avium genes and proteins expressed in macrophages. The Mig protein, for instance, was shown to be involved in the metabolism of fatty acid during macrophage infection [9]. Using cDNA subtractive hybridization technique, it was shown that the mig gene was uniquely expressed by M. avium during growth in macrophages [10]. The M. avium oxyR gene is a regulator responsible for inducing the expression of several genes associated with oxidative stresses [11]. More recently, Houand colleagues, using selective capture of transcribed sequences (SCOTS), identified 46 genes that are expressed by M. avium during growth in human macrophages [12].

To expand knowledge and gain new insights into genes that are important for mycobacterial survival and growth within macrophages, we have screened a M. avium-GFP library created in Mycobacterium smegmatis for genes up-regulated in the intracellular environment. Our work shows that this strategy was successful in identifying several M. avium genes regulated primarily in macrophages.

2 Materials and methods

2.1 Bacterial strains, plasmids, and growth conditions

M. avium 104 strain was isolated from the blood of an AIDS patient and used as a source of genomic DNA for library construction. M. smegmatis mc2 155 was a gift from Dr. William Jacobs Jr. (Albert Einstein School of Medicine, NY) and was used for preparing mycobacterial competent cells for M. avium GFP-promoter library screening upon infection of macrophages. Escherichia coli strain XL1-blue MRF’ (Stratagene, La Jolla, CA) was used as the host strain for construction and maintenance of the M. avium-promoter library prepared in the pEMC1 vector. Plasmid DNA was isolated from E. coli with the QIAprep Miniprep Kit (QIAGEN, Valencia, CA) according to the manufacturer's instruction. M. smegmatis were harvested, re-suspended in 50 mM glucose, 25 mM Tris pH 8, 10 mM EDTA (GTE) solution with 10 mg/ml lysozyme and lysed overnight. The next day, plasmid DNAs were isolated using QIAprep Miniprep kit (QIAGEN, Valencia, CA) according to the method recommended by the manufacturer.

M. avium was cultured on Middlebrook 7H11 agar supplemented with oleic acid, albumin, dextrose, and catalase (OADC, Difco Laboratories, Detroit, Mich) for 10 days, and the isolated transparent colonies were washed and re-suspended in Middlebrook 7H9 broth for an additional 5 days. M. smegmatis mc2 155 was cultured as described for M. avium, except that it was harvested after 3 days of growth. E. coli XL1-blue MRF’ transformants were selected on LB agar containing 50 μg/ml kanamycin. M. smegmatis transformants were selected on 7H11 agar with 0.05% Tween 80, OADC and 50 μg/ml kanamycin.

2.2 GFP-vector construction

We used a green fluorescent protein reporter system to identify M. avium genes expressed intracellularly. The E. coli-Mycobacterium shuttle vector pEMC1 (Fig. 1) was constructed using the plasmid pMV261 as a base, by removing the hps60 promoter and inserting the promoterless GFPmut2 gene (obtained from Rafael Valdivia and Stanley Falkow, Sanford University [13]), into XbaI and PstI restriction sites. The trpA transcriptional terminator was added upstream of the GFP gene between EcoRI and HindIII sites (Fig. 1). The trpA terminator placed before the gene ensures that GFP expression is the result of the activation of bacterial promoters.

1

Plasmid map of pEMC1. This plasmid has been modified to contain a random library of 300 to 1000 bp Sau3AI DNA fragments of M. avium 104 upstream of the inserted GFP for transcriptional fusion of promoterless GFP gene. Abbreviations: oriE, E. coli origin of replication; oriM, mycobacterial origin of replication; GFPmut2, gene encoding Green Fluorescent Protein; trpA, transcriptional terminator.

2.3 M. avium promoter GFP-library construction

M. avium 104 genomic DNA was extracted as previously described [14,15] and digested partially with Sau3A. After electrophoresis on a 1% agarose gel, fragments with 300–1000 bp size were extracted from the gel and cloned into the dephosphorylated BamH1 site in the pEMC1 plasmid. The resulting library was expanded in E. coli XL1-blue MRF’ competent cells (Strategene, La Jolla, CA) by electroporation [16] and plated on Luria Bertani (LB) agar containing 50 μg/ml kanamycin. Screening of E. coli transformants by restriction analysis of pEMC1 showed that approximately 80% of the library contained DNA inserts. This library (containing 20,000 clones in E. coli) was used as the source of plasmids for GFP-promoter library construction in M. smegmatis. Mycobacterial competent cells were prepared and transformed by electroporation as previously described [16,17]. Transformants were plated on Middlebrook 7H11 agar containing 50 μg/ml kanamycin. Over 10,000 individual M. smegmatis clones representing 2.5-times coverage of M. avium genome were stored in pools of five in 96-well plates containing Middlebrook 7H9 broth with 50% glycerol. The pools in replicate 96-well plates were used for macrophage experiments.

2.4 Infection of macrophages

U937 macrophages (ATCC CRL-1593.2) were grown in RPMI 1640 supplemented with 10% fetal bovine serum as previously described [18]. U937 cells were seeded (2 × 105 cells per well) in 96-well flat-bottomed tissue culture plates at 80–100% confluence. The macrophages were treated with 500 ng/ml of Phorbol 12-Myristate 13-Acetate (PMA, Sigma, St. Louis, MO) overnight. Adherent monolayers were infected with bacteria (MOI 10:1) and incubated for 2 h at 37 °C and 5% CO2. The supernatant was removed and the wells were washed three times with Hank's Balanced Salt solution (HBSS, GIBCO, Grand Island, NY) to remove extracellular bacteria. A baseline level of GFP was recorded after infection using a Cytofluroimeter II (Biosearch, Bedford MA). The level of GFP expression was then quantified after 4, 24 and 48 h of incubation at 37 °C. Wells associated with at least a 2-fold increase in GFP expression were selected for further evaluation. Pools from wells associated with increased expression of GFP were plated on 7H11 agar plates with 50 μg/ml of kanamycin to separate individual clones. Individual clones were then propagated in 7H9 broth with 50 μg/ml of kanamycin, and used to infect macrophages as described above. Expression of GFP was then measured following the same time points of 4, 24, and 48 h. Clones that resulted in 2.5-fold or more increase in GFP production over baseline were selected for sequencing.

2.5 Nucleotide sequence/data analysis

Clones associated with increased GFP expression were sequenced at the Central Service Laboratory (CSL), Center for Gene Research and Biotechnology (CGRB), Oregon State University, Corvallis by using specific primers for GFP and kanamycin sequences. Database search and sequence comparisons were performed with these fragments at the National Center for Biotechnology Information (http://www.ncbi.nih.gov), using BLAST network service, and at the database at the TIGR Institute (http://www.tigr.org). Sequence homology between M. avium and M tuberculosis was used to determine the M. avium gene corresponding to the promoter sequence obtained from each clone. Gene sequences reported in this paper have been deposited in GenBank under Accession Numbers AY632749AY632773.

2.6 RNA source, total RNA extraction and RT

Intracellular M. avium from infected U937 macrophages and M. avium growing in 7H9 broth at exponential phase were used as sources of total experimental and control RNA, respectively. Macrophage monolayers were infected with M. avium 104 as described above for M. smegmatis. After 24 h, monolayers were lysed with 0.1% SDS and centrifuged at 300× rpm for 5 min to remove lysed cells from the suspension. After, the supernatant was centrifuged at 3500× rpm for 10 min and the pellet was used for bacterial RNA extraction. Total bacterial RNA extraction was based on the combination of a guanidine thiocyanate-based buffer (Trisol) (Invitrogen, Carlsbad, CA) and rapid mechanical cell lysis of M. avium in a bead-beater, as previously described [19]. RNA was cleaned up with RNA clean kit (QIAGEN, Valencia, CA) and treated with DNase I from the RNA extraction kit prior to RT experiments. RNA integrity was electrophoretically verified by ethidium bromide staining and by OD260/280 nm absorption. Mycobacterial total RNA (1 g) was reverse transcribed with 100U of Superscript II Plus RNase H Reverse Transcriptase (Invitrogen, Carlsbad, CA), using RT primers according to the manufacturer's instruction.

2.7 LightCycler real time PCR

Real time PCR was performed using SYBR Green technology with the following LightCycler experimental run protocol: denaturation program 95 °C for 10 min; amplification and quantification program repeated for 40 times: 95 °C for 30 s, 60 °C for 30 s, and 72 °C for 2 min with a single fluorescence measurement. Target cDNAs from experimental and control samples were amplified in separate PCR reactions.

The threshold cycle (Ct), which is defined as the fractional cycle number at which the fluorescence reaches 10× the standard deviation of the baseline, was quantitated as described in User Bulletin #2 for the ABI PRISM 7700 sequence detection system [20]. To verify the fold change in gene expression we used an amplification-based strategy. The calculated threshold cycle (Ct) for each gene amplification was normalized to the Ct of the 16S rRNA gene amplified from the corresponding sample before calculating fold change using the following formula: Embedded Image where Embedded Image Embedded Image

Quantitative analysis was performed using iCycler iQ software (BIORAD, Hercules, CA). We used a relative quantification in which the expression levels of M. avium target genes were compared to data from a standard curve generated by amplifying several dilutions of a known quantity of amplicons.

2.8 Quantitation of bacterial gene expression in infected mice by real time PCR

Expression of M. avium-selected genes within the animal tissues was examined directly by quantitative real time PCR. C57 BL/6 black mice, 8–10 weeks of age, were infected intravenously through the tail vein with M. avium 104. After 24 h of infection, animals were sacrificed and spleens were harvested. The spleen tissue suspensions were lysed with 0.3% SDS. Lysed cells from suspension were removed by differential centrifugation at 300 rpm for 5 min and the supernatant was removed and centrifuged at 3500 rpm for 10 min to recover intact bacterial cells. The resulting bacterial pellets were re-suspended in Trisol (Invitrogen, Carlsbad, CA). RNA extraction was performed as previously described [19]. Samples were cleaned up with RNA clean kit (QIAGEN, Valencia, CA) and treated with DNase I. The RNA preparation was considered acceptable for further use after verification of RNA concentration and quality by ethidium bromide staining and the A260/280 ratio. Reverse transcription was performed with the Invitrogen kit, as described above. Quantitation of the expression of M. avium genes was carried out with SYBR Green I assay by real time PCR detection system using gene-specific primers (Table 1). Fluorescence was monitored in real time during the 60 °C annealing steps.

View this table:
1

Sense and antisense primers for real time RT PCR amplified genes

3 Results

3.1 M. avium gene expression in macrophages identified by GFP-promoter library

To analyze changes in M. avium gene expression upon macrophage infection, we screened a M. avium promoter library in M. smegmatis for significantly increased GFP expression. M. smegmatis is capable of infecting U937 cells and surviving for approximately 72 h (data not shown). Therefore, genes expressed early in the infection would be detected by GFP expression. Screening of the library identified 29 clones displaying differential induction of GFP over the baseline (Table 2). Monitoring of GFP expression indicated that some promoters were activated after 24 h, others were activated after 48 h. A number of the identified promoters were activated at both time points. There was no significant GFP expression at 4 h postinfection compared to the expression at baseline. Among the identified sequences are promoters for genes associated with bacterial metabolic pathways, nucleotide synthesis (guaB3 and corA), DNA replication (dna ZX), polyketide synthesis (pks7), transport/binding (dppD, glnQ), protein translation and modification (infB, greA), energy metabolism/respiration (nirB), degradation of macromolecules (lipL), and genes with unknown function. The majority of the genes have homologues with either Mycobacterium tuberculosis and/or other bacterial genes. Three of the clones contained sequences within genes. Because the M. avium genome has not been annotated, it is not certain that these ORFs have been correctly chosen by our analysis. In addition these sequences could indicate the potential for a cryptic promoter, and were not further analyzed in this study.

View this table:
2

Identity of M. avium promoters associated with infection of macrophages

3.2 M. avium gene expression analysis by real time quantitative PCR

To confirm that the changes in M. avium gene expression level upon macrophage infection were similar to the ones seen using M. smegmatis as a surrogate host, six genes were selected for real time RT-PCR analysis. These six genes showed the greatest levels of GFP induction compared to other identified clones. The expression of M. avium gene homologues to the following M. tuberculosis H37Rv genes, Rv2573 (Ma2573), Rv3884c (Ma3884c), pks7 (Ma-pks7), dppD (Ma-dppD),lprC (Ma-lprC) and nirB (Ma-nirB), within macrophages was examined directly by quantitative real time PCR. Rv1787 (Ma1787), a PPE gene, was selected as a positive control that was shown to be expressed only upon infection of macrophages by GFP expression and RT-PCR (Li et al., submitted). Rv0359 (Ma0359), a gene encoding a probable conserved integral membrane protein, was sequenced from a GFP promoter library clone that did not have expression in macrophages, and used as a negative control. To determine the sensitivity of the quantitative PCR assay and relative changes in expression for the detection of the M. avium-selected genes, we used a dilution series of cDNA template with a fixed concentration of the primers. real time PCR efficiencies were calculated from the given slopes in LightCycler software. The corresponding real time PCR efficiency (E) of one cycle in the exponential phase was calculated according to the equation: E= 10(−1/slope)[21]. Investigated transcripts from the range 0.5–40 ng cDNA input (n= 2) showed high real time efficiency with high linearity.

The expression levels of target genes from intracellular bacteria and from broth-grown bacteria were normalized to the expression level of the endogenous reference 16S rRNA in each sample. The expression of 16S rRNA was constant, independent of conditions. M. avium genes with greater GFP expression in the original screen were also highly expressed following uptake of the bacteria by macrophages compared with Ma0359 control, as shown in real time PCR data (Fig. 2). Expression of the M. avium gene, Ma1787 (homologue to M. tuberculosis Rv1787, member of the PPE family), was up-regulated 12.2-fold compared with control, while expression of Ma-dppD gene (homologue to Rv3663c, or dppD, hypothetical ABC-transporter), was up-regulated 14.9-fold. In addition, there was significant induction of Ma2573, Ma3884c, Ma-pks7, Ma-lprC and Ma-nirB genes by intracellular M. avium, confirming that the genes detected by using M. smegmatis system were expressed intracellularly by M. avium.

2

M. avium gene expression following 24 h of macrophage infection. Total RNA from broth-grown bacteria as well as from intracellular M. avium-infected macrophages was used as the source for relative quantitation of M. avium gene expression. The fold change in M. avium target genes relative to the 16S rRNA endogenous control gene was determined as described in Section 2.

3.3 Quantitation of bacterial gene expression in infected mouse spleens by real time PCR

To examine how M. avium expression of the genes identified in macrophages is modulated during the host infection, we used a mouse model of infection. Mice were challenged intravenously with M. avium. After 24 h of infection, the animals were harvested to determine the levels of expression in M. avium genes. Changes in transcription of the six selected M. avium genes associated with the highest GFP expression in the initial screening and the control Ma1787 and Ma0359 genes were examined. RNA isolated from mouse spleens was used for real time PCR quantitation of M. avium target gene expression in vivo. For each selected gene, the ratio between the cDNA levels of bacteria from mouse spleens and the cDNA level of M. avium genes in culture during exponential phase of growth was calculated. The quantitative amount of the target gene transcripts was standardized with 16S rRNA expression. Figs. 2 and 3 show that all genes up-regulated during infection of macrophages in vitro were also up-regulated upon mice infection. In general, gene expression in vivo was greater than in macrophages in vitro. The increase of cDNA copy numbers in infected mice was 26-fold for Ma1787 gene, 24.4-fold for Ma-pks7, and greater than 34-fold for MA-dppD.

3

Real time PCR quantification of selected M. avium genes in the spleens of C57 BL/6 after 24 h of post-infection. Spleens were harvested and bacterial RNA was purified, as described in materials and methods. RNA was used for determine the number of copies of cDNAs for target and reference genes. Data were analyzed on the basis of Ct values of each sample and normalized with an internal housekeeping gene control, 16S rRNA.

4 Discussion

Several classes of genes were found to be up-regulated in the first 24 h of infection in macrophages by the GFP screening, including genes involved in anaerobiosis (nirB), nucleotide synthesis, polyketides, ribosomal protein synthesis, DNA replication, macromolecules degradation and transport. A large number of the identified genes have unknown function; although among them, significant homology was identified with the apbA gene in Salmonella and with the dihydrolipoamide dehydrogenase genes in Mycobacterium leprae.

The investigation of whether six of the genes identified in the M. smegmatis GFP system would be up-regulated in M. avium upon macrophage uptake showed that, in fact, all of the genes had their expression up-regulated inside phagocytic cells compared to the expression observed in broth. The dppD gene, for example, was not expressed when the bacterium was cultured in broth, but was up-regulated 14.9-fold in the intracellular environment of macrophages. The level of expression of the dppD gene in macrophages was as high as the expression of the PPE gene homologous to Rv1787. This gene was used as control because of the characteristic of being expressed by the bacterium only inside macrophages (Li et al., submitted). The DppD protein is an ABC-transporter similar to the agaA gene of Agrobacterium tumefaciens[23]. The agaA gene has been described in the Agrobacterium genome, but its function is not presently known [23]. The mycobacterium protein (DppD) similar to agaA is also up-regulated by greater than 34-fold in vivo, in mice spleens. These findings suggest that this protein probably has a role in the bacterial survival in the intracellular environment, perhaps by transporting virulence-associated factors.

The over-expression of nirB in the intracellular environment has been previously shown by Hou and colleagues [12]. The nirB gene encodes for a nitrite reductase, part of the nitrogen metabolism, and has been identified during the genome sequences of Salmonella typhimurium and Bacillus subtilis. Transport of nitrite by bacteria has been associated with virulence, and proteins involved in such a function have been detected in the pathogen Shigella flexnerii, but not in the laboratory strain E. coli K-12 [24].

The observation that a polyketide synthase is up-regulated in vitro in macrophages confirms previous findings that polyketides are involved in virulence of mycobacteria [2527]. The up-regulation is likely to mean that M. avium modifies its envelope during the intracellular phase of infection. Curiously, pks7 is up-regulated approximately 3 to 4-fold in vitro, but 23-fold in vivo. The difference in the level of expression has no current explanation, but could represent either a different environment between mouse and human macrophages or increased levels in host cell expression of some polyketide genes over time. Other genes involved in polyketide synthesis such as pks11 and pks12 have been demonstrated to participate in M. tuberculosis virulence [25].

Lipoproteins have been associated with a number of functions, including bacteria binding to host proteins. The fact that lprC, an esterase, has increased expression when M. avium infects macrophages in vitro and in vivo (∼8-fold) also can be suggestive of envelope remodeling. Up-regulation of the expression of lipoproteins, or enzymes involved in its synthesis or degradation, have been described in previous studies [12,22].

Many genes are expressed secondary to a stress response. Some of the genes identified in our study would fit the definition. For example, the gene homologous to Rv0148 has significant homology to a steroid dehydrogenase. Other genes such as dihydrolipoamide dehydrogenase may also belong to the category. It is an interesting observation that a gene with homology to a sporulation gene in Bacillus has been identified, however, its function in mycobacteria is unknown.

We have used a surrogate system to identify M. avium genes up-regulated in macrophages. Although this system has limitations (for example, it does not provide information about genes up-regulated at later time points), the similarity of findings shared with a few other papers describing screening strategies increases our confidence in the system. Furthermore, the fact that we could confirm our initial findings with M. smegmatis by using M avium in macrophages and in mice makes this system a feasible option for a number of studies. Inactivation of some of the identified genes is an on-going project in the laboratory and will define their role in M. avium virulence.

Acknowledgements

This work was supported by the grant #AI-43199 from the National Institute of Allergy and Infectious diseases.

Footnotes

  • Editor: R.Y.C. Lo

References

  1. [1].
  2. [2].
  3. [3].
  4. [4].
  5. [5].
  6. [6].
  7. [7].
  8. [8].
  9. [9].
  10. [10].
  11. [11].
  12. [12].
  13. [13].
  14. [14].
  15. [15].
  16. [16].
  17. [17].
  18. [18].
  19. [19].
  20. [20].
  21. [21].
  22. [22].
  23. [23].
  24. [24].
  25. [25].
  26. [26].
  27. [27].
View Abstract