UBC Faculty Research and Publications

Inositol monophosphate phosphatase genes of Mycobacterium tuberculosis Movahedzadeh, Farahnaz; Wheeler, Paul R; Dinadayala, Premkumar; Av-Gay, Yossef; Parish, Tanya; Daffé, Mamadou; Stoker, Neil G Feb 18, 2010

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-12866_2009_Article_1006.pdf [ 1.49MB ]
JSON: 52383-1.0220557.json
JSON-LD: 52383-1.0220557-ld.json
RDF/XML (Pretty): 52383-1.0220557-rdf.xml
RDF/JSON: 52383-1.0220557-rdf.json
Turtle: 52383-1.0220557-turtle.txt
N-Triples: 52383-1.0220557-rdf-ntriples.txt
Original Record: 52383-1.0220557-source.json
Full Text

Full Text

RESEARCH ARTICLE Open AccessInositol monophosphate phosphatase genes ofMycobacterium tuberculosisFarahnaz Movahedzadeh1,6*, Paul R Wheeler2, Premkumar Dinadayala3, Yossef Av-Gay4, Tanya Parish5,Mamadou Daffé3, Neil G Stoker1AbstractBackground: Mycobacteria use inositol in phosphatidylinositol, for anchoring lipoarabinomannan (LAM),lipomannan (LM) and phosphatidylinosotol mannosides (PIMs) in the cell envelope, and for the production ofmycothiol, which maintains the redox balance of the cell. Inositol is synthesized by conversion of glucose-6-phosphate to inositol-1-phosphate, followed by dephosphorylation by inositol monophosphate phosphatases(IMPases) to form myo-inositol. To gain insight into how Mycobacterium tuberculosis synthesises inositol we carriedout genetic analysis of the four IMPase homologues that are present in the Mycobacterium tuberculosis genome.Results: Mutants lacking either impA (Rv1604) or suhB (Rv2701c) were isolated in the absence of exogenousinositol, and no differences in levels of PIMs, LM, LAM or mycothiol were observed. Mutagenesis of cysQ (Rv2131c)was initially unsuccessful, but was possible when a porin-like gene of Mycobacterium smegmatis was expressed,and also by gene switching in the merodiploid strain. In contrast, we could only obtain mutations in impC (Rv3137)when a second functional copy was provided in trans, even when exogenous inositol was provided. Experimentsto obtain a mutant in the presence of a second copy of impC containing an active-site mutation, in the presenceof porin-like gene of M. smegmatis, or in the absence of inositol 1-phosphate synthase activity, were alsounsuccessful. We showed that all four genes are expressed, although at different levels, and levels of inositolphosphatase activity did not fall significantly in any of the mutants obtained.Conclusions: We have shown that neither impA, suhB nor cysQ is solely responsible for inositol synthesis. Incontrast, we show that impC is essential for mycobacterial growth under the conditions we used, and suggest itmay be required in the early stages of mycothiol synthesis.BackgroundMycobacterium tuberculosis is a major global pathogen.In 2007, approximately 1.7 million deaths were causedby tuberculosis (TB) and an estimated 9.3 million peopleacquired the infection [1]. Patients can usually be curedthrough a six month course of a multiple drug regimen[2]. The efficacy of chemotherapy has however beencompromised by the appearance of multi- and exten-sively drug resistant strains [3,4]. The search for poten-tial novel drug targets and the subsequent developmentof new antibiotics is therefore urgent. Ideal candidateswould be mycobacterial-specific and include pathwaysinvolved in the biosynthesis of the unusual cell envelope[5,6]; the target of some existing antibiotics, includingisoniazid, ethionamide, ethambutol and pyrazinamide[7].Inositol is a polyol that is not synthesized in most bac-terial species. However, in the mycobacteria, inositol isfound in lipoarabinomannan (LAM), a lipoglycan that ispresent in high levels in the cell envelope. LAM is com-posed of a mannan backbone with branched arabinosylchains. It is anchored in the cell envelope by means of aphosphatidylinositol (PI) moiety. Other lipoglycansfound in the cell envelope include lipomannan (LM)and PI mannosides (PIMs). PI-containing moleculeshave been demonstrated as essential for growth in thefast-growing species Mycobacterium smegmatis, asmutants lacking PI synthase are not viable [8].The function of LAM in cell envelope integrity isunknown, but evidence suggests that it has profoundeffects on the host., for example, it stimulates* Correspondence: movahed@uic.edu1Department of Pathology and Infectious Diseases, Royal Veterinary College,Royal College Street, London NW1 0TU, UKMovahedzadeh et al. BMC Microbiology 2010, 10:50http://www.biomedcentral.com/1471-2180/10/50© 2010 Movahedzadeh et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the CreativeCommons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, andreproduction in any medium, provided the original work is properly cited.macrophages to produce TNFa [9], nitric oxide [10],and matrix metalloproteinases [11]. LAM may thereforeplay a major role in the stimulation of an inappropriatehost immune response, leading to the pathology that ischaracteristic of TB. LAM also induces transcriptionalactivation of HIV-1 [12,13] and may play a role in thesynergy seen between HIV and TB. In addition to theseeffects, LAM is a major antigen [14,15]. While somePIMs are probable precursors of LAM, they may alsohave important functions of their own. PI dimannoside(PIM2), for example, has been implicated as a receptorfor interacting with mammalian cells [16], as a secretedactivator of Toll-like receptor 2 in macrophages leadingto TNFa induction [17], and as an inducer of granu-loma formation [18].Inositol is also a constituent of the major mycobacter-ial thiol, mycothiol (1-D-myo-inosityl-2- [N-acetyl-L-cysteinyl] amido-2-deoxy-a-D-glucopyranoside) [19,20],which helps maintain the redox state of the cell anddetoxifies harmful molecules. A mutant of M. smegmatisthat essentially fails to produce mycothiol is viable, butgrows poorly, and is sensitive to H2O2 [20] However, inM. tuberculosis the mshA and mshC genes, required formycothiol biosynthesis, are essential genes [21,22].Mycothiol may be more important in pathogenic myco-bacteria as during infection they would be exposed toreactive oxygen intermediates within the macrophage.The biosynthesis of inositol normally occurs in twosteps. In the first, glucose-6-phospate is converted toinositol-1-phosphate (I-1-P) by inositol phosphatesynthase (Ino1). We have shown previously that an ino1(Rv0046c) mutant of M. tuberculosis is an inositol auxo-troph, and is severely attenuated in vivo [23]. In the sec-ond step, the I-1-P is dephosphorylated by an inositolmonophosphate phosphatase (IMPase) to form inositol.Previously, we identified the M. smegmatis impA gene,which is predicted to encode an IMPase, and showedthat inactivation of this gene resulted in an altered col-ony morphology, reduced levels of PI dimannoside(PIM2), and altered permeability of the cell wall. Thisdata suggests that impA is partly responsible for inositolsynthesis in this species, presumably compensated bythe presence of other imp genes [24]. In this paper, wedescribe the genetic analysis of four IMPase homologuesof M. tuberculosis. We demonstrate that three, impA,suhB and cysQ are dispensible, while impC is essential,even in the presence of exogenous inositol.MethodsBacterial strains, plasmids and mediaBacterial strains and plasmids used are shown in Table1. M. tuberculosis H37Rv (ATCC 25618) was culturedon Middlebrook 7H10 agar plus 10% (vol/vol) oleicacid-albumin-dextrose-catalse (OADC) supplement(Becton Dickinson). Middlebrook 7H9 broth (Difco)plus 10% (vol/vol) OADC supplement and 0.05% (wt/vol) Tween 80 was used to grow liquid cultures. Hygro-mycin (100 μg ml-1), kanamycin (20 μg ml-1), gentami-cin (10 μg ml-1) and X-Gal (5-bromo-4-chloro-3-indolyl-b-D-galactopyranoside) at 50 μg ml-1, wereadded where appropriate. For supplementation with ino-sitol, a 14% stock (w/v) (0.77 M) of myo-inositol (Sigma)was prepared and filter-sterilised. E. coli DH5a was usedfor all plasmid constructions.BioinformaticsHomology searches were carried out using BLASTP ver2.2.13 [25] The four homologs identified all had e-values<10-3, and no other protein match approached signifi-cance. Prosite database information was obtained athttp://www.expasy.ch/prosite/, using Release 20.56 datedNovember 4th, 2009.Construction of M. tuberculosis mutantsTargeted mutagenesis was carried out using a two-stepstrategy [26] in order to introduce an unmarked muta-tion without any potential polar effects.ImpAPrimers tbimpA1 (CTCGACGTACAGGTTGAGC-TATCC) and tbimpA2 (CTTCACCTGACCGATCGT-CAGCTC) were used to amplify the impA gene andflanking regions (2,108 bp) from M. tuberculosis H37Rvusing PCR. The resulting 2.1 kb fragment was clonedinto the EcoRV site of pGEM5, producing pIMP50. A200 bp SphI fragment within impA was removed follow-ing partial digestion and religated to make pIMP51. The2,348 bp PvuII fragment of pIMP51 was cloned intop2NIL, producing pIMP57.To create a deletion where the majority of impAwas deleted (769 bp deleted from 813 bp), inversePCR was performed on pIMP57. Primers tbimpAinv1(TCGTGCCAGCTGACCAACGAATCCAAGTGCAT)and tbimpAinv2 (TCGTGCCAGCTGATAGGGGAAC-CAGAGGACTA) were used, simultaneously creating adeletion and introducing a PvuII site in the deleted con-struct. Following the PCR reaction the DNA was digestedwith DpnI for 1 h at 37C to destroy the template, thendigested with PvuII and religated to produce pFM74.Insertion of a PacI gene cassette from pGOAL19 wascloned at the PacI site of pFM74 producing the finaldelivery plasmid pFM75. The PacI cassette carries lacZand sacB, which can be used for positive and negativeselection of unmarked mutant colonies, respectively.SuhBA 3,534 bp XhoI fragment of cosmid Y5ab was clonedinto the SalI site of plasmid p2NIL to produce pFM33.Movahedzadeh et al. BMC Microbiology 2010, 10:50http://www.biomedcentral.com/1471-2180/10/50Page 2 of 15Table 1 M. tuberculosis strains and plasmidsStrains/plasmids Characteristics SourceE. coli DH5a InvitrogenM. tuberculosis H37Rv wild-type laboratory strain ATCC 25618FAME1 M. tuberculosis suhBΔ This studyFAME2 M. tuberculosis impAΔ This studyFAME4 M. tuberculosis impCΔ::pFM96 This studyFAME7 M. tuberculosis::pFM54 (impCΔ SCO) This studyFAME9 FAME7 ::pFM96 This studyFAME11 FAME7::pFM123 This studyFAME63 FAME7::FM203 This studyFAME5 M. tuberculosis ino1Δ [23]FAME12 M. tuberculosis ino1Δ::pFM54 (SCO) This studyFAME35 M. tuberculosis::pFM151 (cysQΔ SCO) This studyFAME43 FAME35::FM164 This studyFAME53 cysQΔ::FM164 This studyFAME87 FAME35::FM203 This studyFAME93 cysQΔ::FM203 This studyFAME 120 M. tuberculosiscysQΔ:: pUC-Hyg-intThis studypBluescript II SK+ StratagenepGEM5 PromegapUC-Gm-int pUC-based plasmid with HindIII cassette carrying gm and L5 int [54]pUC-Hyg-int pUC-based plasmid with HindIII cassette carrying hyg and L5 int [54]p2NIL gene manipulation vector, kan [26]pGOAL19 hyg pAg85-lacZ sacB PacI cassette vector [26]pIMP50 pGEM5::impA This studypIMP51 pGEM5::impAΔ (SphI 200 bp) This studypIMP57 p2NIL::impAΔ (SphI 200 bp) This studypFM74 p2NIL::impAΔ (769 bp) This studypFM75 pFM74 with PacI cassette of pGOAL19 This studypFM33 p2NIL::suhB This studypFM48 pFM33::suhBΔ This studypFM52 pFM48 with PacI cassette of pGOAL19 This studypFM31 p2NIL::impC This studypFM53 pFM31::impCΔ This studypFM54 pFM53 with PacI cassette of pGOAL19 This studypFM94 pBluescript SK+::impC (+288 bp upstream) This studypFM96 pFM94::int gm This studypFM123 pFM96::impC D86N This studyPMN013 plasmid carrying the M. smegmatis porin gene mspA [44]pFM203 pMN013::int gm This studypFM145 p2NIL::cysQ This studypFM148 pFM145::cysQΔ This studypFM151 pFM148 with PacI cassette of pGOAL19 This studypFM160 pBluescript SK+::cysQ (+352 bp upstream) This studypFM164 pFM160::int gm This studyMovahedzadeh et al. BMC Microbiology 2010, 10:50http://www.biomedcentral.com/1471-2180/10/50Page 3 of 15A fragment of 817 bp was deleted from the 874 suhBgene by inverse PCR on pFM33 using primers tbsuhBΔ1(TCAGCATGCGTTCGTTGTCAGGTCGTGTC) &tbsuhBΔ2 (TCAGCATGCGATTCAACGGCCTAGAGC); this introduced a SphI site in the deleted con-struct. Following treatment with DpnI and SphI, thiswas religated to produce pFM48. Insertion of the genedelivery cassette from pGOAL19 produced the finaldelivery plasmid pFM52.ImpCA 2,503 bp StuI fragment of cosmid Y3A2 was clonedinto the PmlI site of p2NIL, producing pFM31. A 731bp deletion was generated in the 783 bp gene byinverse PCR on pFM31 using primers tbimpCΔ1(TGCCAGCTGCATTAGATCGTCGTGGCTCA) &tbimpCΔ2 (TGCCAGCTGGAGGTGCTGACACGGCTC) to introduce a PvuII site in the deleted construct.Following treatment with DpnI and PvuII, this was reli-gated to produce pFM53. Insertion of the delivery genecassette from pGOAL19 produced the final deliveryplasmid pFM54.CysQPrimers tbcysQ1 (CCTGGTCGACCTGTTTCC) andtbcysQ2 (GCGGCTCTTTGACATCTTGT) were used toamplify the cysQ gene and flanking regions (2,748 bp)from M. tuberculosis H37Rv DNA. The product wascloned into the PmlI site of p2NIL, producing pFM145.Primers tbcysQΔ1 (AGTCAGGTCGTCCGTCAGATC)& tbcysQΔ2 (TACAACCAACTGGACCCCTAC) wereused to generate a 666 bp deletion in the 804 cysQ geneby inverse PCR on pFM145. Following treatment withKlenow polymerase and T4 polynucleotide kinase (Pro-mega), this product was religated to produce pFM148.Insertion of the gene delivery cassette from pGOAL19produced the final delivery plasmid pFM151.MutagenesisDeletion plasmids were constructed as described above.The delivery plasmids were introduced into M. tubercu-losis H37Rv or M. tuberculosis H37Rv ino1, and singlecrossovers (SCOs) isolated by selection for blue hygRkanR colonies. One SCO colony was plated onto 2%(wt/vol) sucrose-50 μg ml-1 X-Gal to isolate bacteriawith a second crossover; this will lead to mutant orwild-type cells depending on the location of the recom-bination event. In order to screen for impC mutant,DNA was extracted from sucroseS kanS white colonies(obtained from plating M. tuberculosis FAME9 ontosucrose medium) and analysed by PCR using primersthat flank the impC gene (TBC1: GGACCGCGATCAGTATGAGT and TBC2: TCGACACAGAATCCGCTAGA). Strains carrying the impC wild-type allelewould produce a band of 1148 bp whereas strains carry-ing an impC mutation would carry the deletion band of417 bp. Mutant candidates and a wild-type control weredigested with PvuII and subjected to Southern blot ana-lysis using a 2.5 kb impC probe (impC plus flankingregion). The wild-type strain showed a 4 kb band whilstthe mutant showed a 3.2 kb deletion band along with a2.5 kb band for the integrated impC copyComplementationA construct expressing the impC gene was made byPCR amplification of the impC gene, together with 288bp of upstream sequence using chromosomal M. tuber-culosis H37Rv as template DNA. The primers tbimpC-BamP (CGCGGATCCGGCGATGGTGACAT) andtbimpCBam (CGCGGATCCTTACCCGGCGTTGAGC)were used. The product was digested with BamHI andcloned into the BamHI site of pBluescript-SK+ to pro-duce pFM94. The HindIII cassette of pUC-Gm-int, car-rying the int and gm genes was cloned into the HindIIIsite of pFM94 to produce pFM96.A construct expressing the cysQ gene was made byPCR amplification of the cysQ gene including 352 bp ofupstream sequence using M. tuberculosis H37Rv; chro-mosomal template DNA; primers tbcysup (GCATAGAGCAGGAGGTTTGC) and tbcysend (GCGCCACGCGTCGGCGAT) were used. The PCR product wastreated with T4 polynucleotide kinase and cloned intothe SmaI site of pBluescript-SK+ to produce pFM160.The HindIII cassette of pUC-Gm-int, carrying the intand gm genes was cloned into the HindIII site ofpFM160 to produce pFM164.Site-directed mutagenesisSite-directed mutagenesis was carried out using thenon-PCR-based Quickchange kit (Stratagene). Oligonu-cleotides D86N-forward (GGATCGTAGACCCGATCAACGGCACCAAAAACTTTGTGC) & D86N-reverse(GCACAAAGTTTTTGGTGCCGTTGATCGGGTCTACGATCC) were used to prime DNA synthesis withpFM96. Sequencing confirmed the presence of therequired mutation.Real-time quantitative PCRRNA was prepared from an exponential (7-day) rollingculture of M. tuberculosis H37Rv [27] and cDNA synth-esis was carried out using Superscript II (Invitrogen)according to the manufacturer’s protocol. Primers weredesigned for Real-time quantitative PCR (RTq-PCR) forsigA (endogenous control), impA suhB, impC and cysQ)using the Primer3 software, ensuring products would beless than 500 bp (Table 2). RTq-PCR reactions were setup using the DyNAmo SYBR Green qPCR kit (MJResearch). RT-qPCR was performed using the DNAMovahedzadeh et al. BMC Microbiology 2010, 10:50http://www.biomedcentral.com/1471-2180/10/50Page 4 of 15Engine Opticon 2 System (Genetic Research Instrumen-tation) using a standard 1 × DNA master SYBR Green Imix, 1 μl cDNA product and 0.3 mM of each primer in20 μl on ice. The primer concentrations had first beenoptimised. Samples were heated to 95°C for 10 minbefore cycling for 35 cycles of 95°C for 30 s, 60°C(impA, suhB, impC and cysQ) or 62°C (sigA) for 20 s,and 72°C for 20 s. Fluorescence was captured at the endof each cycle after heating to 80°C to ensure the dena-turation of primer dimers. In order to measure relativegene expression levels, standard curves for each primerset were generated by performing PCR with SYBR greendetection on serial dilutions of quantified genomicDNA. CT values were converted into the equivalent ofcopy number by comparison to the standard curve.Control reactions where RNA had not been reversetranscribed were used to confirm that there was no sig-nificant contaminating genomic DNA present. In orderto control for any differences in reverse transcriptaseefficiencies each value was standardized to sigA to gen-erate unit-less values. SigA is a major housekeeping geneand levels of sigA mRNA remains constant under a widerange of conditions [28]. Two independent RNA sam-ples were assayed in triplicate for each gene.Cell wall analysisExtraction and analysis of PIMsCells (0.2 g) were delipidated with chloroform/methanol(1/1, v/v) for 48 h at room temperature with continuousstirring. Lipids were separated from the delipidated cellsby centrifugation (3000 rpm, 15 min, 2600 xg) and ana-lysed by TLC on silica gel-coated plates developed withchloroform/methanol/water (60:35:8, v/v/v). The variousPIMs were identified by their mobilities on TLC andtheir positive reactivity compared to authentic standards;these included a sugar and phospholipid-specific reagent(0.2% anthrone in concentrated H2SO4 followed by heat-ing) and the Dittmer-Lester reagent that specificallydetects phosphorous-containing lipids, respectively [29].Production and analysis of LAM and LMDelipidated cells were washed and disrupted using a Celldisrupter (2 kbars, Constant System Ltd; one shotmodel). The resulting material was extracted with 40mL ethanol/water (1/1, v/v) for 8 h at 65°C; the bacterialresidues were discarded and the supernatant was dried.Six ml hot phenol/water (1/1, v/v) were added and themixture was heated for 1 h at 70°C under continuousstirring, followed by a two-phase partition. The phenolphase was discarded and the upper phase extensivelywashed and dried. The extract was solubilised in waterand Triton X114 (2% wt/v) was added to the cooled sus-pension. The mixture was stirred for 10 min and thenheated at 50°C until two phases formed. The detergentphase was recovered, diluted by adding 1 ml water andwashed three times with CHCl3. The resulting aqueousphase was dried to evaporate the chloroform and resus-pended in water (0.2 ml). This portion was analysed bySDS-PAGE with a 5% stacking gel and a 15% runninggel. Samples were denatured in the presence of 2% SDSin 50 mM Tris-HCl (pH 6.8). After electrophoresis, gelswere treated with periodate/ethanol/acetic acid (0.7/40/5, w/v/v), and silver-stained. Authentic samples ofmycobacterial LAM and LM from Mycobacterium bovisBCG were used as standard.Sugar compositional analysisThe sugar constituents of the various materials weredetermined after acid hydrolysis with 2 M CF3COOH at110°C for 1 h; the mixture of hydrolysed products wasdried, treated with trimethylsilyl reagents [30] to deriva-tise monosaccharides and analysed by gas chromatogra-phy (GC) for their sugars.Gas chromatography and mass spectrometryGC was performed using a Hewlett Packard HP4890Aequipped with a fused silica capillary column (25 mlength × 0.22 mm i.d.) containing WCOT OV-1 (0.3 mmfilm thickness, Spiral). A temperature gradient of100-290°C at 5°C min-1, followed by a 10-min isothermplateau at 290°C, was used.Mycothiol assayLabelling of cell extracts with monobromobimane(mBBr) to determine thiol content was performed withmodifications to previously published protocols [31,32].Cell pellets from 3 ml culture were resuspended in 0.5ml of warm 50% acetonitrile-water containing 2 mMmBBr (Cal Biochem), and 20 mM HEPES-HCl, pH 8.0.The suspension was incubated for 15 min in a 60°Cwater bath and then cooled on ice. A final acidic pHwas produced by adding 2-5 μl 5 M HCl or 5 M tri-fluoracetic acid.The control samples were extracted with 0.5 ml ofwarm 50% acetonitrile-water containing 5 mM N-ethyl-malemide and 20 mM HEPES-HCl, pH 8.0. The suspen-sion was incubated for 15 min in a 60°C water bath andTable 2 Primers used in Real time quantitative PCRGene Primer pair Primer sequencesigA SIGAF ATCTGCTGGAAGCCAACCTSIGAR GATCACCTCGACCATGTGCimpA IMPAF: CGATCTCGTCTTCGTCGCIMPAR: CCCTATGCTGCCAAGAATCTCsuhB IMPBF: GCGAGAAGCAGGCAGAATTIMPBR: CTCTCGGCGTTGACAACAAimpC IMPCF: GCTGCTTGAAGATGGCGTCIMPCR: CCACCAGGCAGTAAGACAGAAcysQ CYSQF: ATCTGACGGACGACCTGACTCYSQR: CCAACGGGTCAATAATCCACMovahedzadeh et al. BMC Microbiology 2010, 10:50http://www.biomedcentral.com/1471-2180/10/50Page 5 of 15then cooled on ice. 2 mM mBBR were added to thesolution followed by a second incubation for 15 min ina 60°C. The control sample was cooled but not acidified.Cell debris was pelleted in each sample by centrifugation(5 min 14,000 × g).HPLC analysis of thiols was carried out by injecting 25μl of 1:4 dilution of samples in 10 mM HCl on to aBeckman Ultrasphere IP 5 μ(250 mm × 4.6 mm) col-umn using 0.25% glacial acetic acid pH 3.6 (buffer A)and 95% methanol (buffer B). The gradient was: 0 min,10% B; 15 min, 18% B; 30 min, 27% B; 32 min, 100% B;34 min, 10% B; and 60 min, 10% B (reinjection). Theflow rate was 1 ml min-1, and the fluorescence detectionwas accomplished on a Varian Fluorichrom model430020 with a 370 nm excitation filter and a 418-700nm emission filter. Data collection and analysis was per-formed on Dynamax Mac Integrator (RaininInstruments).Impase activityBacteria were grown to mid-log phase, and collected bycentrifugation. Each pellet was washed once in distilledwater followed by resuspension in 2.5 ml 2 mM dithio-threitol in 50 mM Tris-Cl, pH8. The suspended bacteriawere disrupted in a FastPrep220A at 4 m/sec for 3 cyclesof 20 sec in Lysing Matrix B (0.1 mm silica beads), withcooling on ice between cycles. The resulting cell-extractswere then clarified at 4000 g for 4 min using a benchcentrifuge and filter-sterilised through 0.2 μm pore cellu-lose acetate filters (Sartorius Minisart). Each clarified cellextract was desalted through Pharmacia PD-10 columnsaccording to the manufacturer’s instructions; with theexception that 3.2 ml (not 3.5 ml) protein fraction wascollected. For equilibrating, desalting and eluting usingPD-10, 50 mM Tris-Cl, pH8 was used.Phosphatase assays were conducted using 0.4 mMsubstrates at 37°C, as described previously [33] althoughthe reaction volume used was 120 μl and was stoppedwith 30 μl malachite green reagent. No precipitates wereformed so the entire assay was performed in ELISAplate wells. Inorganic phosphate present in each wellwas calculated by reading the OD against a standardcurve. Enzyme activity was then calculated by subtract-ing the phosphate formed in wells with cell extract andsubstrate, from phosphate formed in correspondingwells with cell extract but without substrate.ResultsBioinformatics analysisThere are four genes in the M. tuberculosis genome thatencode proteins with significant homology to IMPases.All four M. tuberculosis proteins are equally distantfrom the human IMPase (PDB structure 1IMA; 22-30%identity, 37-46% similarity) [34] and the aligned aminoacid sequences are shown in Figure 1A. The fourproteins are only as similar to each other, as to thehuman protein (27-32% identity, 36-44% similarity).These four genes are generally conserved in other acti-nomycete genomes, with for example, apparent ortho-logs in Mycobacterium avium, Mycobacteriumsmegmatis, and Corynebacterium glutamicum (data notshown). M. leprae, which has many pseudogenes, has nofunctional impA. Other genomes do also have a smallnumber of other IMPase genes (thus, M. avium has afifth paralog that is similar to cysQ). While levels ofhomology between the different M. tuberculosis IMPaseparalogs are moderate (22-30% amino acid identity),similarities between orthologs are much higher (forexample, 75-79% identity between M. tuberculosis andM. leprae, and 51-67% identity between M. tuberculosisand M. smegmatis).The genomic contexts of these genes are shown inFigure 2. As with M. smegmatis [24], the impA gene(Rv1604) lies in the middle of the main his operonbetween hisA and hisF. The stop codon of hisA overlapswith the putative start codon of impA, and the stopcodon of impA overlaps with the putative start codon ofhisF. These impA genes are 70% identical.The suhB gene (Rv2701c) was named in the originalgenome annotation [35], because it is the gene mostsimilar to the Escherichia coli suhB gene. The E. colisuhB gene was so-named because deletion of the generesulted in a cold-sensitive phenotype, and suppressionof a thermosensitive rpoH mutation [36]. It has alsobeen shown to suppress secY [37], dnaB [38], and era[39] mutations. However, these phenotypes are notrelated to the enzymatic properties of the protein, asthey are unaffected by a null point mutation in theactive site [40] (Figure 1B). Furthermore, inositol pro-duction is not believed to occur in E. coli, so the biolo-gical context is very different from that in mycobacteria.Recombinant SuhB from M. tuberculosis has been con-firmed to have IMPase activity [41]. SuhB is monocistro-nic in M. tuberculosis (Figure 2).The third homologous gene is Rv3137, which we havecalled impC. It appears to be the first gene in a two-gene operon; a 457 bp intergenic gap upstream of impCsuggests it has its own promoter., and a second gene,pflA, is predicted to start only 14 bp downstream, so isprobably co-transcribed. PflA shows homology to pyru-vate formate lyase-activating proteins. Beyond this is acluster of fad genes (fadE24-fadE23-fadB4), but the gapbeyond pflA and fadE24 is 79 bp, so is less likely to bepart of the same operon.The fourth homologous gene is cysQ (Rv2131c), so-named because it is most similar to the E. coli cysQgene. E. coli cysQ mutants are cysteine auxotrophs dur-ing aerobic growth [42]. Interestingly M. smegmatis con-tains two paralogs of this gene.Movahedzadeh et al. BMC Microbiology 2010, 10:50http://www.biomedcentral.com/1471-2180/10/50Page 6 of 15Figure 1 Alignment of IMPases. The M. tuberculosis H37RvIMPases were aligned using ClustalW. (A) Complete sequences. Motifs shown in bold;(B) Prosite motifs: ‘*’ identical residues in all sequences; ‘:’ conserved substitutions; ‘.’ semi-conserved substitutions. Sequences were obtained fromhttp://genolist.pasteur.fr/TubercuList/. Reported Prosite motifs are 1 (N-terminal; PS00629): [FWV]-x(0,1)- [LIVM]-D-P- [LIVM]-D- [SG]- [ST]-x(2)- [FY]-x-[HKRNSTY]; and 2 (C-terminal; PS00630): [WYV]-D-x- [AC]- [GSA]- [GSAPV]-x- [LIVFACP]- [LIVM]- [LIVAC]-x(3)- [GH]- [GA]. Residues that are notencompassed by these motifs are in bold italics. Arrows indicate putative metal binding aspartate and isoleucine residues reported for humanIMPase [55]. The underlined residue shows the aspartate mutated in this study, which is equivalent to mutations introduced into the E. coli andhuman proteins (see main text).Movahedzadeh et al. BMC Microbiology 2010, 10:50http://www.biomedcentral.com/1471-2180/10/50Page 7 of 15Two sequence motifs have been described for IMPasesin the Prosite database [43] (see legend to Figure 1B).One motif, near the N-terminus contains the metal-binding aspartate residues of the active site, and theother lies near the C-terminus. All of the gene productsexcept SuhB had small differences from at least one ofthe two IMPase motifs (Figure 1B). However, they allcontain the important metal-binding residues in bothmotifs.The M. tuberculosis impA and suhB genes are dispensableThe impA gene has previously been shown to play anindirect role in inositol synthesis in M. smegmatis [24],and a knockout plasmid construct was therefore pre-pared to isolate an M. tuberculosis impA mutant. As thegene lies within the his operon (Figure 2), this plasmidcarried an unmarked deletion that would not have polareffects. The mutant was generated using a two-stepmethod [26], and grew well on solid medium. Unlikethe M. smegmatis impA mutant which had altered col-ony morphology, there were no obvious differences incolony morphology between the wild-type and mutantstrains.We carried out a similar experiment to determinewhether suhB plays a role in inositol metabolism. Again,a deletion construct was prepared, and an unmarkedmutant isolated, with no obvious differences in colonymorphology.Inactivation of CysQWe constructed a plasmid to delete the cysQ gene. Initi-ally, we were unable to obtain a mutant; of 97 doublecrossovers (DCOs) screened in the presence of inositol,all were wild-type. We therefore made a merodiploidstrain by integrating a second copy of cysQ into the sin-gle crossover (SCO) strain, and repeating the selectionfor DCOs on sucrose. Using this method, 24 out of 30colonies were found to be mutants. The ability to isolatea mutant only in the presence of a functional copy ofthe gene indicates that this gene was essential under theconditions tested. It could be inferred that cysQ synthe-sizes all the inositol in the cell, or all the inositol for aspecific essential molecule. However, this hypothesis isimprobable, as, if true, we would predict thatmutantswould be inositol auxotrophs, yet no mutants were iso-lated even in the presence of high levels of inositol. Onepossibility is that inositol does not penetrate the cellwall, which is known to be highly impermeable. How-ever, as we had successfully isolated a mutant lackinginositol-1-phosphate synthase (an inositol auxotroph),only when the media was supplemented with extremelyhigh levels of exogenous inositol (50-77 mM) [23], itseems that inositol does enter the cell in sufficient quan-tities but permeability to this molecule is poor. Thissuggests that even a slight increase in the requirementfor inositol might make mutant isolation impossible,since we had reached the limits of inositol solubility.Figure 2 Genomic context of M. tuberculosis IMPase genes. White arrows: imp genes; black arrows: other genes; open rectangles deletedregions in knock out plasmids.Movahedzadeh et al. BMC Microbiology 2010, 10:50http://www.biomedcentral.com/1471-2180/10/50Page 8 of 15We reasoned that an increase in the availability of inosi-tol by introduction of a porin might allow a mutant tobe isolated. We therefore electroporated an integratingplasmid (pMN013) carrying the M. smegmatis poringene mspA [44,45] (for which M. tuberculosis has noorthologue) into the SCO strains, and repeated thesucrose selection. Using this method, we successfullyisolated a cysQ mutant in the presence of 77 mM inosi-tol. We screened 16 DCO colonies and two weremutants.We then plated the mutant on inositol-free medium,and were surprised to observe normal growth, indicatingthat once the mutant has been isolated, it does notrequire inositol. To determine whether the mutant couldsurvive in the absence of the mspA gene, the integratingplasmid containing this gene was switched with an emptyintegrative plasmid carrying the hygromycin resistancegene (pUC-Hyg-int). This would result in the replace-ment of the cysQ-carrying plasmid, leaving a stain withno functional cysQ. Surprisingly, we were able to obtaincysQ mutants using this approach although we had failedto isolate a mutant by our standard mutagenesis proce-dure. We therefore conclude that cysQ is also dispensible,and a cysQ mutant does not require inositol for growth.The impC gene is essentialWe attempted to construct an unmarked impC deletionmutant. The first step of the mutagenesis to produceSCOs worked well, however, when cells carrying a sec-ond crossover were isolated, only wild-type bacteriawere obtained. In theory, the second crossover couldtake place on either side of the deletion, resulting ineither mutant or wild-type cells. The fact that weobtained only wild-type cells (n = 48) suggested that themutants are not viable.These initial mutagenesis experiments were carriedout in the absence of exogenous inositol. We thereforerepeated the mutagenesis, including different levels ofinositol in the media at all times. Again, only wild-typebacteria were isolated following the second cross-over(n= 97; 16 on 15 mM inositol, 8 on 30 mM, 16 on 46mMl, and 57 on 77 mM).The inability to obtain a mutant may be due to otherfactors, such as a low frequency of recombination onone side of the gene, even though the length of flankingDNA should be sufficient (847 and 874 bp). Thereforewe constructed a merodiploid strain by inserting a sec-ond functional copy of impC into the SCO strain. Thisextra copy was present on an L5-based integrating vec-tor, and contained 288 bp upstream of impC, which waslikely to carry its promoter. When this strain (FAME9)was plated onto sucrose to isolate DCOs, three out ofeight colonies isolated had lost the original copy ofimpC. The fact that this gene could only be lost when asecond copy of the gene is present suggests that impC isessential for survival, even in the presence of high levelsof exogenous inositol (Fisher’s exact test, p < 0.01, com-paring only the experiments with 77 mM inositol andthe complemented strain). To further investigate theessentiality of the impC gene, and in view of what wasobserved with cysQ, we introduced the mspA gene intothe impC SCO strain; this time we were not successfulin obtaining a mutant, indicating that the difficulty weencountered making an impC mutant differed fromcysQ.A difference between an IMPase mutant and an ino1mutant may be that inositol-1-phosphate accumulates inthe IMPase mutant, which might somehow be detrimen-tal to the cell. We therefore carried out the essentialityexperiment in an ino1 mutant background. The impCmutant construct was introduced into M. tuberculosisino1, and a SCO strain isolated. On plating for DCOs,only wild-type colonies were isolated (n = 38), againsuggesting that this was not the explanation for theessentiality.Site directed mutagenesis of impCOur results suggest that impC does not have a criticalrole in inositol production and hence our inability toobtain an impC mutant may indicate that impC has adifferent or secondary function that prevents isolation ofa mutant. For example, the enzyme might form part ofan enzyme complex, and play a vital structural role inmaintaining the integrity of that complex. Deletion ofthe gene would then have both enzymatic and structuraleffects. An analogous situation was found with the E.coli SuhB protein; where phenotypes in suhB mutantswere not related to IMPase activity, as a point mutationin the active site did not produce the suppressing phe-notype [40]. We therefore used the same approach totry to separate enzymatic activity from a structural role.A D93N change in E. coli SuhB and an equivalentD90N change in the human IMPase suppress activity[40,46] (Figure 1B). Site-directed mutagenesis was usedto introduce a corresponding mutation (D86N) in theM. tuberculosis impC gene using the integrating plasmidpFM96 previously used for complementation. This plas-mid (pFM123) was introduced into the SCO strainFAME7, and the resultant strain (FAME11) was streakedonto sucrose/inositol plates. DCO colonies were ana-lysed, and, in contrast to the situation with pFM96, allwere shown to be wild-type (n = 52). The fact that thefunctional impC gene could not be replaced by thismutated gene, even in the presence of inositol (p <0.01), shows that the mutation did inactivate enzymaticactivity, and (assuming that the structure was notaffected) that it is this enzymatic activity that is essen-tial, rather than an additional structural role.Movahedzadeh et al. BMC Microbiology 2010, 10:50http://www.biomedcentral.com/1471-2180/10/50Page 9 of 15Enzyme activitiesIn order to gain a greater understanding of the functionof these IMPases, we expressed impC as a recombinantprotein. However, despite using different plasmid con-structs and strategies, we were unable to obtain a solu-ble protein (not shown). As an alternative to directlyassaying enzyme activity, we assayed IMPase activity incell extracts of the mutant strains to obtain informationabout their relative contributions to inositol synthesis.We compared enzyme activities in whole cell extractsfrom the wild-type and mutant strains (Tables 3 and 4).Of the seven substrates tested, phosphate release as aresult of adding the enzyme source was significantlyhigher than controls for fructose bisphosphate (FBP),the inositol phosphates, 5’ AMP and p-nitrophenyl-phosphate. Deletion of the impA, suhB, or cysQ genesmade no significant difference to IMPase activity. ThecysQ mutants had significantly less FBPase than the par-ent strain, (P < 0.05; t-test). However, the fructoseFBPase activity in the H37Rv control for the cysQmutants (Table 4) is significantly less than in H37Rvcontrol used for impA and suhB mutants (P < 0.05; t-test) (Table 3) suggesting that the small but significantdifferences reported in this study may be due to batch-to-batch variation rather than in relation to anymutations.PIM, LAM and mycothiol levels are normal in the impA,suhB and cysQ mutantsCell extracts of the mutant strains were prepared forthe assay of inositol-containing molecules (cell envelopeglycolipids and mycothiol). TLC analyses showed thatPIMs were normal in the mutant strains (Figure 3A),whilst polyacrylamide gel electrophoresis (Figure 3B)and sugar compositional analysis (not shown) demon-strated normal levels of LAM and LM. Mycothiol levelswere assayed by HPLC analysis; levels in the impA,suhB and cysQ mutants were similar to wild-type (seeFigure 4).Gene expression levels of imp genes in M. tuberculosisThe relative contributions of the IMPase homologues willbe proportional to their activity, and their level of expres-sion. We therefore carried out RTq-PCR experiments todetermine the levels of expression of impA, suhB, cysQand impC mRNA in exponential cultures of M. tubercu-losis. Expression levels were normalized to those of sigAmRNA which remains constant. The level of cysQ wasthe highest, almost equal to sigA (Table 5). impA andimpC were expressed at approximately 40% of this level,while suhB was lowest, at 12% of the cysQ level.DiscussionTo investigate how M. tuberculosis synthesises inositol,we carried out a genetic analysis of four IMPase homolo-gues in M. tuberculosis. The impA and suhB genes wereshown to be dispensable, with no phenotype detected interms of the levels of mycothiol, PIMs, LM or LAM.CysQ is also dispensible, although isolating the mutantproved more difficult, requiring introduction of the M.smegmatis mspA porin gene for mutant isolation, but notfor subsequent survival. It cannot be excluded, however,that the cysQ mutants that were eventually obtained hadacquired a suppressor mutation, which had allowed dele-tion of cysQ or had allowed growth of the mutant onmedia lacking inositol and preventing cell death. In con-trast to these three genes, we were only able to inactivateimpC by introducing a second copy of the gene. TheTraSH mutagenesis protocol which provides a genome-wide indication of essentiality [47] supports our data,with only impC of these four genes being reported asputatively required for optimal growth in vitro.Inositol production is likely to be essential for myco-bacterial growth, because of the essentiality of bothclasses of mycobacterial inositol-containing molecules,namely phospholipids [8] and mycothiol Our previouswork showing that a PI synthase mutant is an inositolauxotroph [23] is consistent with this. Both SuhB andCysQ have been shown to have IMPase activity [41,48]Table 3 Phosphatases in cell extracts of impA, suhBmutantsSubstrate H37Rv ΔimpA ΔsuhBFructose-1,6-bisP 26.04 ± 1.85 (5) 28.18 ± 0.92 (5) 32.70 ± 0.44 (5)Inositol-1-P 0.63 ± 0.13 (6) 0.79 ± 0.12 (5) 0.63 ± 0.25 (6)Inositol-2-P 1.20 ± 0.15 (4) 1.33 ± 0.22 (5) 1.03 ± 0.15 (6)Glycerol-2-P 0.08 ± 0.06 (12) -0.02 ± 0.03 (2) 0.39 ± 0.03 (2)Glycerol-3-P -0.13 ± 0.12 (12) -0.08 ± 0.03 (2) 0 ± 0.21 (2)5’ AMP 4.22 ± 0.36 (8) 4.13 ± 0.40 (2) 5.74 ± 0.04 (2)p-nitrophenyl-P 3.00 ± 0.35 (12) 3.55 ± 0.14 (2) 4.38 ± 0.36 (2)Values: nmol/min/mg protein, mean ± SEM (n). Differences between levels inmutants and the parent strain were not significant (P > 0.05; t-test).Table 4 Phosphatases in cell extracts of the cysQ mutantsSubstrate H37Rv ΔcysQ 203/12 ΔcysQ203/16Fructose-1,6-bisP 18.94 ± 1.00 (6) 13.09 ± 1.24 (6) 12.41 ± 0.54 (7)Inositol-1-P 0.40 ± 0.09 (8) 0.49 ± 0.17 (9) 0.57 ± 0.16 (9)Inositol-2-P 0.84 ± 0.12 (8) 0.90 ± 0.27 (10) 0.70 ± 0.23 (10)Glycerol-2-P 0.75 ± 0.32 (8) 1.02 ± 0.27 (10) 0.55 ± 0.15 (10)Glycerol-3-P -0.37 ± 0.28 (3) -0.35 ± 0.14 (3) 0.27 ± 0.45 (3)5’ AMP 1.42 ± 0.31 (3) 1.69 ± 0.14 (3) 1.39 ± 0.03 (3)p-nitrophenyl-P 5.51 ± 0.36 (2) 3.64 ± 1.92 (2) 2.83 ± 0.25 (3)Values: nmol/min/mg protein, mean ± SEM (n). Level of FBPase in cysQmutants relative to parent strain is significantly different (P < 0.05; t-test).Level of FBPase in H37Rv parent strain reported in table 4 is significantlydifferent (P < 0.05; t-test) to that reported in Table 3.Movahedzadeh et al. BMC Microbiology 2010, 10:50http://www.biomedcentral.com/1471-2180/10/50Page 10 of 15and we have shown that the M. smegmatis ImpA hasIMPase activity (unpublished data). However, none ofthe three mutants constructed are auxotrophic for inosi-tol, indicating that there is potential redundancy offunction between the homologs and deletion of three orfour genes might be required to see sufficient loss ofactivity to cause auxotrophy. A recent report suggeststhat CysQ is likely to play a role in sulphur metabolism,as its activity as 3’-phosphoadenosine-5’-phosphatase isseveral orders of magnitude higher than as an inositolphosphatase [49]. However, it may still contribute to theredundancy in inositol phosphatase activity.To determine the potential relative contribution ofeach of these genes to inositol synthesis we carried outtwo experiments. One experiment looked at the relativeamounts of mRNA using real-time RTq-PCR. AllmRNA species were detectable, with cysQ being mostabundant (approximately the same level as sigA, theFigure 3 Analyses of cell wall major constituents of some representative mutants; the other strains exhibited profiles similar to thoseshown. (A) TLC analysis of extractable lipids. (B) SDS-PAGE of lipopolysaccharides. WT: M. tuberculosis H37Rv; ΔA: impA mutant; ΔB: suhB mutant;S: authentic standard of mycobacterial LAM and M. bovis BCG LM; TMM: trehalose monomycolate; PE: phosphatidylglycerol; PG:phosphatidylethanolamine; LAM: lipoarabinomannan; LM: lipomannan; PIM: phosphatidylinositol mannoside.Movahedzadeh et al. BMC Microbiology 2010, 10:50http://www.biomedcentral.com/1471-2180/10/50Page 11 of 15major housekeeping sigma factor), and impA being theleast abundant, with a level only one-tenth that of cysQ.We also assayed the level of IMPase activity in thewhole cell extracts of each mutant, reasoning that wemight see a decrease in activity when one of the geneswas deleted. However, no decrease in activity wasobserved in any of the three mutants compared to thewild-type strain. This could be a reflection on the sensi-tivity of our assay, or could indicate that the activity isregulated (either at the transcriptional or post-transcrip-tional level) such that a constant level is maintained.We also have preliminary data that expression of theimpC gene is regulatable. We grew a Δino1 mutant ofM. tuberculosis (which needs >50 mM inositol for itsnormal growth [23]) and looked at the effect of removalof the inositol on gene expression. The only IMPasegene with changed expression was impC, which was 3-fold increased. We cannot link this change directly tothe inositol, because it could also be caused by thechange in osmolarity, but at the very least indicates thisindicates this gene is regulatable (unpublished results).The situation with impC is complicated in that wecould neither obtain a mutant, nor do we have biochem-ical evidence that it functions as an IMPase (despitemany attempts to achieve both). The essentiality cannotbe a simple case of impC producing the majority of theinositol in the cell, as we added inositol exogenously. Itis true that the ino1 mutant we made previously, whichis an inositol auxotroph, required levels of inositolapproaching the maximum solubility limit, so a require-ment for a slightly increased level of inositol mightexplain our findings. However, this is unlikely because(i) we also introduced a porin gene to increase inositoluptake, with no effect, (ii) we would also have to explainwhy the other three IMPase genes are not sufficient,and (iii) the level of impC mRNA is only 21% of thetotal IMPase mRNA (41% if cysQ is excluded). The onlypieces of evidence we have, therefore, that link impC toinositol production are (i) its clear homology toIMPases, and (ii) the circumstantial evidence that levelsof impC increased in a microarray experiment whereinositol was removed from an ino1 auxotroph, whereasthe expression level of the other IMPase genes was notsignificantly changed. We recognise the difficulty of car-rying out the latter experiment in a controlled way sinceremoving such a high level of inositol from the mediumcould have other effects. Interestingly, impC was alsoupregulated in the Wayne low oxygen model, particu-larly when M. tuberculosis cultures entered a microaero-philic state known as nonreplicating persistent stage 1(NRP1), where there is cessation of bacterial replication,strong induction of respiratory nitrate reductase activity,and a change in energy metabolism (3.3-fold induced)[50].Figure 4 HPLC analysis of mycothiol (marked with an arrow) inrepresentative mutants; the other strains exhibited profiles similarto those shown. WT: M. tuberculosis H37Rv; ΔA: impA mutant; ΔB:suhB mutant. Free thiol peaks are marked as standard (lower panel)MS-mB marks the mycothiol peak and and Ac-Cys-mb represent anAcetyl-Cysteine thiol.Table 5 mRNA levelsGene mRNA level normalised to sigA*impA 0.41 (0.3- 0.5)suhB 0.11 (0.096- 0.13)impC 0.36 (0.27- 0.46)cysQ 0.95 (0.76- 1.18)*To ensure equal amounts of cDNA were used each value was standardizedto sig A to generate unit-less values(95% confidence interval)Movahedzadeh et al. BMC Microbiology 2010, 10:50http://www.biomedcentral.com/1471-2180/10/50Page 12 of 15We tested the hypothesis that the essentiality of impCis unrelated to its enzymatic activity by constructing asite-directed mutation. The mutation introducedchanges at an active-site of glutamate to glutamine; theanalogous mutation has been shown to abrogate activityin the human protein [40,46]. Our inability to isolatemutants, strongly suggests that (i) the point mutationdoes indeed affect the activity of the enzyme and (ii)impC carrying this point mutation cannot complementa null mutant even in the presence of inositol. Thesefindings oppose our hypothesis of a structural role forImpC, and support an enzymatic role, as an explanationof its essentiality. There still remains a possibility thatthe mutation also affects the structure as we have notshown that folded protein is still produced, but webelieve this is unlikely given the subtle nature of thechange introduced. Another possible explanation for theinositol-independent essentiality is that removal ofImpC results in a build up of inositol-1-phosphate,which is somehow deleterious to the cell. However, wewere unable to obtain an impC mutant in an ino1 back-ground. It is feasible that ImpC uses a substrate otherthan inositol i.e. one involved in mycothiol production.The elegant work of Fahey and co-workers has definedmost of the mycothiol biosynthesis pathway, but is miss-ing a predicted phosphatase., which dephosphorylatesN-acetyl glucosamine-(a1,3)-1L-inositol-1-phosphate.We carried out preliminary experiments attempting tomake an impC mutant using this substrate (kindly pro-vided by R. Fahey and G. Newton), without success (notshown). However, we have no evidence that it wouldpenetrate the cell, so we feel we cannot draw anyconclusions.The impC gene lies upstream of the pflA gene andmay be co-transcribed, as the intergenic gap is only 19bp. PflA shows homology to pyruvate formate lyase-acti-vating proteins; oxygen-sensitive iron-sulfur proteinsthat activate an anaerobic ribonucleotide reductase insome bacteria [51], although there does not appear tobe a homologue to E. coli pyruvate formate lyase in theM. tuberculosis genome. We designed an unmarkeddeletion of impC, in order to prevent polar effects. Inaddition, complementation with impC alone was suffi-cient to allow mutants to be isolated. We have thereforeexcluded polar effects on pflA as an explanation for theessentiality.The Mycobacterium leprae genome contains manypseudogenes therefore genomic comparisons may givean indication as to which mycobacterial genes are essen-tial. In M. leprae, the impA orthologous gene is a pseu-dogene, with several frameshifts in the distal half of thegene, whereas the other three orthologous IMPase genesare retained. The suhB orthologous gene (ML1024)appears to be functional, and has a similar genomiccontext to the M. tuberculosis gene. The orthologousimpC gene (ML0662) appears to be monocistronic inthis species, and the orthologous cysQ gene (ML1301) isalso present. The lack of phenotype in an M. tuberculo-sis impA mutant contrasts with the situation seen in M.smegmatis, where an impA mutant had altered colonymorphology, slower growth, and reduced levels of PIM2[24]. The fact that the M. smegmatis mutant is viablesupports the idea of some redundancy of function, andwe suggest that the differences in phenotype are causedby different levels of ImpA compared to other IMPasesin the two species.Given that inositol monophosphatase and fructose-bisphosphatase activities were detected in cell extractsfrom impA, suhB and cysQ mutants, none of thesegenes can encode the major enzyme for these activities.The cysQ gene product does in fact act as a phosphatasewith fructose-1,6- bisphosphate and inositol-1-phosphate[48], but enzyme activity in assays does not alwaysequate to functionality in living bacteria. An example isfound in Thermococcus kodakarensis where knockingout the fbp gene encoding a fructose bisphosphatasewith high substrate specificity resulted in a strain unableto grow on gluconeogenic substrates whilst knockingout its imp gene encoding a member of the carbohy-drate phosphate superfamily with substrate specificityincluding fructose-1,6- bisphosphate did not affect itsgrowth on any carbon sources [52]. In M. tuberculosis,the effect of knocking out the glpX gene that encodesfructose bisphosphatase is so drastic it is difficult toenvisage that impA, suhB or cysQ can compensate forits loss [53].ConclusionsWe have demonstrated that the M. tuberculosis impA,suhB and cysQ genes are dispensable, but that impC isessential under the growth conditions used. The reasonfor the essentiality is unclear in terms of inositol synth-esis; at present the most attractive hypothesis is thatimpC is required for mycothiol synthesis.AcknowledgementsWe thank Jane Turner for excellent technical assistance; Bob Cox for thesuggestion to use mspA, Gerry Newton, Bob Fahey, Anne Lemassu, PhilipDraper and Del Besra for helpful discussions, and Michael Niederweis andClaudia Mailaender for plasmid pMN013. FM was funded by the WellcomeTrust (project grant 051880) and the European Union TB vaccine clusterContract no. QLK2-1999-01093 and Wellcome Trust grant 073237. PRW wasfunded by the Department for Environment, Food & Rural Affairs (UK), and(DEFRA). M. tuberculosis cosmids were kindly provided by Carol Churcher atthe Sanger Centre.Author details1Department of Pathology and Infectious Diseases, Royal Veterinary College,Royal College Street, London NW1 0TU, UK. 2Tuberculosis Research,Veterinary Laboratories Agency, New Haw, Addlestone, Surrey KT15 3NB, UK.3CNRS, Institut de Pharmacologie et Biologie Structurale, UMR CNRS-Movahedzadeh et al. BMC Microbiology 2010, 10:50http://www.biomedcentral.com/1471-2180/10/50Page 13 of 15Université Paul Sabatier (UMR 5089), 205, route de Narbonne, 31077Toulouse cedex 04, France. 4Department of Medicine, Division of InfectiousDiseases, University of British Columbia, 2733 Heather St., Vancouver, BritishColumbia, Canada V5Z 3J5. 5Queen Mary University of London, Barts and theLondon School of Medicine and Dentistry, 4 Newark Street, London E1 2AT,UK. 6Current address: Institute for Tuberculosis Research (M/C 964), Collegeof Pharmacy, Rm 412, University of Illinois at Chicago, 833 S. Wood St.Chicago, Illinois USA 60612-7231.Authors’ contributionsFM carried out the molecular genetic studies, participated in the design andcoordination of the study and drafted the manuscript. PW conceived of thestudy, carried out the enzyme assays and wrote the corresponding sectionof the manuscript. PD performed cell wall analysis. MD designed the cellwall analysis and aided in drafting the manuscript. YA conceived of thestudy, designed and carried out the Mycothiol assay. TP conceived of thestudy, participated in the design and coordination, and aided in drafting themanuscript. NS conceived of the study, participated in its design andcoordination, performed the bioinformatics and participated in drafting themanuscript. All authors read and approved the final manuscript.Received: 8 April 2009Accepted: 18 February 2010 Published: 18 February 2010References1. WHO:http://www.who.int/tb/publications/global_report/2009/pdf/full_report.pdf.2. Dye C, Garnett GP, Sleeman K, Williams BG: Prospects for worldwidetuberculosis control under the WHO DOTS strategy. Directly observedshort-course therapy. Lancet 1998, 352(9144):1886-1891.3. Espinal MA, Laszlo A, Simonsen L, Boulahbal F, Kim SJ, Reniero A, Hoffner S,Rieder HL, Binkin N, Dye C, et al: Global trends in resistance toantituberculosis drugs. World Health Organization-International Unionagainst Tuberculosis and Lung Disease Working Group on Anti-Tuberculosis Drug Resistance Surveillance. N Engl J Med 2001,344(17):1294-1303.4. Van Rie A, Enarson D: XDR tuberculosis: an indicator of public-healthnegligence. Lancet 2006, 368(9547):1554-1556.5. Daffe M, Draper P: The envelope layers of mycobacteria with referenceto their pathogenicity. Adv Microb Physiol 1998, 39:131-203.6. Lee RE, Brennan PJ, Besra GS: Mycobacteriumtuberculosis cell envelope.Curr Top Microbiol Immunol 1996, 215:1-27.7. Zhang Y, Telenti A: Genetics of drug resistance in Mycobacteriumtuberculosis. Molecular genetics of mycobacteria Washington, D.C.: ASMPressHatfull GF, Jacobs WR Jr 2000, 235-254.8. Jackson M, Crick DC, Brennan PJ: Phosphatidylinositol is an essentialphospholipid of mycobacteria. J Biol Chem 2000, 275(39):30092-30099.9. Moreno C, Taverne J, Mehlert A, Bate CA, Brealey RJ, Meager A, Rook GA,Playfair JH: Lipoarabinomannan from Mycobacterium tuberculosis inducesthe production of tumour necrosis factor from human and murinemacrophages. Clin Exp Immunol 1989, 76(2):240-245.10. Chan ED, Morris KR, Belisle JT, Hill P, Remigio LK, Brennan PJ, Riches DW:Induction of inducible nitric oxide synthase-NO* by lipoarabinomannanof Mycobacterium tuberculosis is mediated by MEK1-ERK, MKK7-JNK, andNF-kappaB signaling pathways. Infect Immun 2001, 69(4):2001-2010.11. Chang JC, Wysocki A, Tchou-Wong KM, Moskowitz N, Zhang Y, Rom WN:Effect of Mycobacterium tuberculosis and its components onmacrophages and the release of matrix metalloproteinases. Thorax 1996,51(3):306-311.12. Zhang Y, Nakata K, Weiden M, Rom WN: Mycobacterium tuberculosisenhances human immunodeficiency virus-1 replication bytranscriptional activation at the long terminal repeat. J Clin Invest 1995,95(5):2324-2331.13. Bernier R, Barbeau B, Olivier M, Tremblay MJ: Mycobacterium tuberculosismannose-capped lipoarabinomannan can induce NF-kappaB-dependentactivation of human immunodeficiency virus type 1 long terminalrepeat in T cells. J Gen Virol 1998, 79(Pt 6):1353-1361.14. Da Costa CT, Khanolkar-Young S, Elliott AM, Wasunna KM, McAdam KP:Immunoglobulin G subclass responses to mycobacteriallipoarabinomannan in HIV-infected and non-infected patients withtuberculosis. Clin Exp Immunol 1993, 91(1):25-29.15. Del Prete R, Picca V, Mosca A, D’Alagni M, Miragliotta G: Detection of anti-lipoarabinomannan antibodies for the diagnosis of active tuberculosis.Int J Tuberc Lung Dis 1998, 2(2):160-163.16. Hoppe HC, de Wet BJ, Cywes C, Daffe M, Ehlers MR: Identification ofphosphatidylinositol mannoside as a mycobacterial adhesin mediatingboth direct and opsonic binding to nonphagocytic mammalian cells.Infect Immun 1997, 65(9):3896-3905.17. Jones BW, Means TK, Heldwein KA, Keen MA, Hill PJ, Belisle JT, Fenton MJ:Different Toll-like receptor agonists induce distinct macrophageresponses. J Leukoc Biol 2001, 69(6):1036-1044.18. Gilleron M, Ronet C, Mempel M, Monsarrat B, Gachelin G, Puzo G: Acylationstate of the phosphatidylinositol mannosides from Mycobacterium bovisbacillus Calmette Guerin and ability to induce granuloma and recruitnatural killer T cells. J Biol Chem 2001, 276(37):34896-34904.19. Spies HS, Steenkamp DJ: Thiols of intracellularpathogens. Identification ofovothiol A in Leishmaniadonovani and structural analysis of a novel thiolfrom Mycobacterium bovis. Eur J Biochem 1994, 224(1):203-213.20. Newton GL, Bewley CA, Dwyer TJ, Horn R, Aharonowitz Y, Cohen G,Davies J, Faulkner DJ, Fahey RC: The structure of U17 isolated fromStreptomyces clavuligerus and its properties as an antioxidant thiol. Eur JBiochem 1995, 230(2):821-825.21. Buchmeier N, Fahey RC: The mshA gene encoding the glycosyltransferaseof mycothiol biosynthesis is essential in Mycobacterium tuberculosisErdman. FEMS Microbiol Lett 2006, 264(1):74-79.22. Sareen D, Newton GL, Fahey RC, Buchmeier NA: Mycothiol is essential forgrowth of Mycobacterium tuberculosis Erdman. J Bacteriol 2003,185(22):6736-6740.23. Movahedzadeh F, Smith DA, Norman RA, Dinadayala P, Murray-Rust J,Russell DG, Kendall SL, Rison SC, McAlister MS, Bancroft GJ, et al: TheMycobacterium tuberculosis ino1 gene is essential for growth andvirulence. Mol Microbiol 2004, 51(4):1003-1014.24. Parish T, Liu J, Nikaido H, Stoker NG: A Mycobacterium smegmatis mutantwith a defective inositol monophosphate phosphatase gene homologhas altered cell envelope permeability. J Bacteriol 1997,179(24):7827-7833.25. Altschul SF, Madden TL, Schaffer AA, Zhang J, Zhang Z, Miller W,Lipman DJ: Gapped BLAST and PSI-BLAST: a new generation of proteindatabase search programs. Nucleic Acids Res 1997, 25(17):3389-3402.26. Parish T, Stoker NG: Use of a flexible cassette method to generate adouble unmarked Mycobacterium tuberculosis tlyA plcABC mutant bygene replacement. Microbiology 2000, 146(Pt 8):1969-1975.27. Betts JC, Lukey PT, Robb LC, McAdam RA, Duncan K: Evaluation of anutrient starvation model of Mycobacterium tuberculosis persistence bygene and protein expression profiling. Mol Microbiol 2002, 43(3):717-731.28. Manganelli R, Dubnau E, Tyagi S, Kramer FR, Smith I: Differential expressionof 10 sigma factor genes in Mycobacterium tuberculosis. Mol Microbiol1999, 31(2):715-724.29. Dittmer JCF, Lester RL: A simple specific spray for the detection ofphospholipids on thin layer chromatography. Journal of Lipid Research1964, 5:126-127.30. Sweeley CC, Bentley R, Makita M, Wells WW: Gas-liquid chromatography oftrimethylsilyl derivatives of sugars and related substances. J Am ChemSoc 1963, 85:2497-2507.31. Anderberg SJ, Newton GL, Fahey RC: Mycothiol biosynthesis andmetabolism. Cellular levels of potential intermediates in the biosynthesisand degradation of mycothiol in mycobacterium smegmatis. J Biol Chem1998, 273(46):30391-30397.32. Newton GL, Arnold K, Price MS, Sherrill C, Delcardayre SB, Aharonowitz Y,Cohen G, Davies J, Fahey RC, Davis C: Distribution of thiols inmicroorganisms: mycothiol is a major thiol in most actinomycetes. JBacteriol 1996, 178(7):1990-1995.33. Nigou J, Besra GS: Characterization and regulation of inositolmonophosphatase activity in Mycobacterium smegmatis. Biochem J 2002,361(Pt 2):385-390.34. Bone R, Frank L, Springer JP, Pollack SJ, Osborne SA, Atack JR, Knowles MR,McAllister G, Ragan CI, Broughton HB, et al: Structural analysis of inositolmonophosphatase complexes with substrates. Biochemistry 1994,33(32):9460-9467.35. Cole ST, Brosch R, Parkhill J, Garnier T, Churcher C, Harris D, Gordon SV,Eiglmeier K, Gas S, Barry CE, textitet al: Decipheringthe biology ofMovahedzadeh et al. BMC Microbiology 2010, 10:50http://www.biomedcentral.com/1471-2180/10/50Page 14 of 15Mycobacterium tuberculosis from the complete genome sequence. Nature1998, 393(6685):537-544.36. Yano R, Nagai H, Shiba K, Yura T: A mutation that enhances synthesis ofsigma 32 and suppresses temperature-sensitive growth of the rpoH15mutant of Escherichia coli. J Bacteriol 1990, 172(4):2124-2130.37. Shiba K, Ito K, Yura T: Suppressors of the secY24 mutation: identificationand characterization of additional ssy genes in Escherichia coli. J Bacteriol1986, 166(3):849-856.38. Chang SF, Ng D, Baird L, Georgopoulos C: Analysis of an Escherichia colidnaB temperature-sensitive insertion mutation and its cold-sensitiveextragenic suppressor. J Biol Chem 1991, 266(6):3654-3660.39. Inada T, Nakamura Y: Lethal double-stranded RNA processing activity ofribonuclease III in the absence of suhB protein of Escherichia coli.Biochimie 1995, 77(4):294-302.40. Chen L, Roberts MF: Overexpression, purification, and analysis ofcomplementation behavior of E. coli SuhB protein: comparison withbacterial and archaeal inositol monophosphatases. Biochemistry 2000,39(14):4145-4153.41. Nigou J, Dover LG, Besra GS: Purification and biochemical characterizationof Mycobacterium tuberculosis SuhB, an inositol monophosphataseinvolved in inositol biosynthesis. Biochemistry 2002, 41:4392-4398.42. Neuwald AF, Krishnan BR, Brikun I, Kulakauskas S, Suziedelis K, Tomcsanyi T,Leyh TS, Berg DE: cysQ, a gene neededfor cysteine synthesis inEscherichia coli K-12 only during aerobic growth. J Bacteriol 1992,174(2):415-425.43. Hofmann K, Bucher P, Falquet L, Bairoch A: The PROSITE database, itsstatus in 1999. Nucleic Acids Res 1999, 27(1):215-219.44. Mailaender C, Reiling N, Engelhardt H, Bossmann S, Ehlers S, Niederweis M:The MspA porin promotes growth and increases antibiotic susceptibilityof both Mycobacterium bovis BCG and Mycobacterium tuberculosis.Microbiology 2004, 150(Pt 4):853-864.45. Niederweis M, Ehrt S, Heinz C, Klocker U, Karosi S, Swiderek KM, Riley LW,Benz R: Cloning of the mspA gene encoding a porin fromMycobacterium smegmatis. Mol Microbiol 1999, 33(5):933-945.46. Pollack SJ, Knowles MR, Atack JR, Broughton HB, Ragan CI, Osborne S,McAllister G: Probing the role of metal ions in the mechanism of inositolmonophosphatase by site-directed mutagenesis. Eur J Biochem 1993,217(1):281-287.47. Sassetti CM, Boyd DH, Rubin EJ: Genes required for mycobacterial growthdefined by high density mutagenesis. Mol Microbiol 2003, 48(1):77-84.48. Gu X, Chen M, Shen H, Jiang X, Huang Y, Wang H: Rv2131c gene product:an unconventional enzyme that is both inositol monophosphatase andfructose-1,6-bisphosphatase. Biochem Biophys Res Commun 2006,339(3):897-904.49. Hatzios SK, Iavarone AT, Bertozzi CR: Rv2131c from Mycobacteriumtuberculosis is a CysQ 3’-phosphoadenosine-5’-phosphatase. Biochemistry2008, 47(21):5823-5831.50. Muttucumaru DG, Roberts G, Hinds J, Stabler RA, Parish T: Gene expressionprofile of Mycobacterium tuberculosis in a non-replicating state.Tuberculosis (Edinb) 2004, 84(3-4):239-246.51. Tamarit J, Mulliez E, Meier C, Trautwein A, Fontecave M: The anaerobicribonucleotide reductase from Escherichia coli. The small protein is anactivating enzyme containing a [4fe-4s](2+) center. J Biol Chem 1999,274(44):31291-31296.52. Sato T, Imanaka H, Rashid N, Fukui T, Atomi H, Imanaka T: Geneticevidence identifying the true gluconeogenic fructose-1,6-bisphosphatasein Thermococcus kodakaraensis and other hyperthermophiles. J Bacteriol2004, 186(17):5799-5807.53. Movahedzadeh F, Rison SC, Wheeler PR, Kendall SL, Larson TJ, Stoker NG:The Mycobacterium tuberculosis Rv1099c gene encodes a GlpX-likeclass II fructose 1,6-bisphosphatase. Microbiology 2004, 150(Pt10):3499-3505.54. Mahenthiralingam E, Marklund BI, Brooks LA, Smith DA, Bancroft GJ,Stokes RW: Site-directed mutagenesis of the 19-kilodalton lipoproteinantigen reveals No essential role for the protein in the growth andvirulence of Mycobacterium intracellulare. Infect Immun 1998,66(8):3626-3634.55. Gill R, Mohammed F, Badyal R, Coates L, Erskine P, Thompson D, Cooper J,Gore M, Wood S: High-resolution structure of myo-inositolmonophosphatase, the putative target of lithium therapy. Acta Cryst2005, D61:545-555.doi:10.1186/1471-2180-10-50Cite this article as: Movahedzadeh et al.: Inositol monophosphatephosphatase genes of Mycobacterium tuberculosis. BMC Microbiology2010 10:50.Submit your next manuscript to BioMed Centraland take full advantage of: • Convenient online submission• Thorough peer review• No space constraints or color figure charges• Immediate publication on acceptance• Inclusion in PubMed, CAS, Scopus and Google Scholar• Research which is freely available for redistributionSubmit your manuscript at www.biomedcentral.com/submitMovahedzadeh et al. BMC Microbiology 2010, 10:50http://www.biomedcentral.com/1471-2180/10/50Page 15 of 15


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items