Devices & Hardware

Two-dimensional bacterial genome display: a method for the genomic analysis of mycobacteria

Description
Two-dimensional bacterial genome display: a method for the genomic analysis of mycobacteria
Published
of 7
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
     D  o  w  n   l  o  a   d  e   d   f  r  o  m   w  w  w .  m   i  c  r  o   b   i  o   l  o  g  y  r  e  s  e  a  r  c   h .  o  r  g   b  y   I   P  :   5   4 .   1   5   9 .   4   9 .   1   1   3   O  n  :   S  a   t ,   1   0   S  e  p   2   0   1   6   0   7  :   5   9  :   2   5 Microbiology   (2002),  148 , 3111–3117  Printed in Great Britain Two-dimensional bacterial genome display: amethod for the genomic analysis ofmycobacteria Edith M. Dullaghan, 1,2 Chad A. Malloff, 3,4 Alice H. Li, 1,3 Wan L. Lam 3,4 and Richard W. Stokes 1,2,3 Author for correspondence:  Edith M. Dullaghan. Tel:   1 604 875 2491. Fax:   1 604 875 2226.e-mail: dullagha  interchange.ubc.ca 1,2,3 The Division ofInfectious andImmunological Diseases,British Columbia’sChildren’s Hospital 1 , andDepartments ofPaediatrics 2 andPathology & LaboratoryMedicine 3 , University ofBritish Columbia,Vancouver, BC, Canada 4 British Columbia CancerResearch Center,601 West 10th Avenue,Vancouver, BC, V5Z 1L3,Canada Annually,  Mycobacterium tuberculosis  is the cause of approximately threemillion deaths worldwide. It would appear that currently available therapiesfor this disease are inadequate. The identification of genes involved inmycobacterial virulence will facilitate the design of new prophylactic andtherapeutic interventions. A method for high-resolution comparison ofbacterial genomes has been developed to facilitate the identification of genespossibly involved in the virulence of clinically relevant mycobacteria. This‘two-dimensional bacterial genome display’ (2DBGD) method utilizes two-dimensional DNA electrophoresis to separate, on the basis of size and G M Ccontent, genomic fragments generated with different restrictionendonucleases. The use of this method to identify genomic differencesbetween species, strains and, most importantly, isogenic mutants ofmycobacteria is reported. That 2DBGD can be used to identify differencesresulting from either insertional mutagenesis using a gentamicin-resistancegene or from a frameshift mutation is demonstrated. Keywords:  Mycobacterium avium  complex, tuberculosis, strain differentiation,fingerprint, two-dimensional DNA electrophoresis INTRODUCTION A number of species of mycobacteria are pathogenic inman. The causative agent of tuberculosis,  Mycobac - terium tuberculosis , is the leading cause of death due toa single infectious species, killing an estimated threemillion people annually. Mycobacteria belonging to the Mycobacterium avium  complex (MAC), i.e.  M .  avium and  Mycobacterium intracellulare , can also establishpulmonary infections in humans and are on the increasemainly as a result of the HIV epidemic (Bermudez  et al .,2000; Havlir & Barnes, 1999).  M .  tuberculosis  andmembers of the MAC complex are facultative intra-cellular pathogens that can survive and replicate withinmacrophages. It is well established that the virulence of separate strains or isolates of   M .  tuberculosis  or MACcan vary (Steenken  et al ., 1934; Schaefer  et al ., ................................................................................................................................................. Abbreviations:  2DBGD, two-dimensional bacterial genome display;2DDE, two-dimensional DNA electrophoresis; DGGE, denaturing gradientgel electrophoresis; MAC,  Mycobacterium avium  complex. 1970; Collins & Smith, 1969; Collins & Stokes, 1987;North & Izzo, 1993). However, little is known, at agenetic level, of the virulence mechanisms employed bymycobacteria to evade the host immune system. Wepropose the use of a novel approach that utilizes two-dimensional DNA electrophoresis (2DDE) technologyto compare genomes from different bacterial isolatesthatcouldbeappliedtoidentifymycobacterialvirulencegenes.It has become feasible to electrophoretically separateanddisplay,intwo dimensions,DNAfragments derivedfrom genomic digests. Separation in the first dimensionis by fragment size and in the second dimensionseparation is by mobility in denaturing gradients.Through the appropriate choice of restriction enzymes,changes as small as single base point mutations can bevisualized in 2D gels. This technique has already beensuccessfully used to display microsatellite polymor-phisms in the human genome for use in genetic mappingand in studying genomic alterations in animal modelsand human cancers (Lam  et al ., 1996; Hughes  et al .,1998; Marcinek etal ., 1997). Inaddition,2DDEanalysis 0002-5649  2002 SGM  3111     D  o  w  n   l  o  a   d  e   d   f  r  o  m   w  w  w .  m   i  c  r  o   b   i  o   l  o  g  y  r  e  s  e  a  r  c   h .  o  r  g   b  y   I   P  :   5   4 .   1   5   9 .   4   9 .   1   1   3   O  n  :   S  a   t ,   1   0   S  e  p   2   0   1   6   0   7  :   5   9  :   2   5 E. M. Dullaghan and others D673 D673104 104 ..................................................................................................... Fig. 1.  2DBGD displays of  Alu I-digestedgenomic DNA from  M. intracellulare  D673and  M. avium  104 (large displays on theleft). Regions of high spot density (the darkareas at the leading edge) can be resolvedby altering electrophoresis conditions andusing alternative restriction enzymes. Close-ups of a corresponding area from each2DBGD display are shown for  M.intracellulare  D673 and  M. avium  104. Thetwo samples were run in parallel in separatedenaturing gradient gels. The majority ofspots do not align upon comparison of thesetwo displays. has been used to distinguish different strains of   Borde - tella pertussis  (Malloff   et al ., 2002).We describe a method, two-dimensional bacterial ge-nome display (2DBGD), for producing displays of mycobacterialgenomesusing2DDEtoseparategenomicsegmentscutwithvariousrestrictionendonucleases.Wedemonstrate the utility of this method by detectinggenomic differences at the species and strain level andbetween isogenic mutants. METHODS BacterialstrainsandDNA. M . tuberculosis H37RvandH37Ra(Steenken  et al ., 1934),  M .  intracellulare  D673 (Dunbar  et al .,1968), D673-Katg (Marklund  et al ., 1998), D673-19KDa(Mahenthiralingam  et al ., 1998) and 1403 (ATCC 35761), and M .  avium  104 (currently being sequenced by TIGR, www.tigr.org  tdb  )weregrownin7H9broth(Difco)supplementedwitholeicacid-albumin-glucosecomplex(OADC)plus0  05%Tween80.OADCwaspreparedbydissolving8  1 gNaCl,50 gBSA and 20 g   -glucose in 950 ml dH  O. The solution wasthen adjusted to pH 7  0 using NaOH. To this was added a30 ml solution of 0  6 ml sodium oleate plus 0  6 ml 6 M NaOHin dH  O. The OADC was warmed to 56   C to clear andwas filter-sterilized.Isolationofgenomic DNA from mycobacteria was carriedoutusing the method of Belisle & Sonnenburg (1998). GenomicDNA was digested with a variety of restriction enzymes thathad been initially tested using the genome restriction digesttool of the Comprehensive Microbial Resource (Peterson  etal ., 2001) to select ones that would produce an even dis-tribution of fragments ranging from 200 to 2000 bp. 2D DNA electrophoresis.  Five hundred nanograms of digestedDNA fragments was treated with calf intestinal alkalinephosphatase (New England Biolabs) prior to radiolabellingwith35kBq[ γ -  P]ATP(6000 Cimmol −  ;AmershamPharma-cia Biotech) using T4 polynucleotide kinase (New EnglandBiolabs). The resulting fragments were size-fractionated in5% non-denaturing acrylamide gels in electrophoresis buffer(40 mMTris,20 mMsodiumacetate,1 mMNa  EDTA,0  2%,v  v,glacialaceticacid,pH 7  4)for 1600volt-hours. Followingthis, each gel lane was cut and placed on top of a large format(25  20 cm) 6% polyacrylamide denaturing gradient gel thatcontainedanascendinggradientofformamide(10–40 %,v  v)and urea (1  8–7 M) in electrophoresis buffer. In the seconddimension, parallel denaturing gradient gel electrophoresis(DGGE) was performed using an ISO-DALT apparatus(Amersham Pharmacia Biotech) for 1700 volt-hours and aconstant temperature of 68  5   C. DGGE gels were run inparallel in the same buffer chamber to ensure uniformity of electrophoretic conditions. A maximum of 10 gels can be run 3112     D  o  w  n   l  o  a   d  e   d   f  r  o  m   w  w  w .  m   i  c  r  o   b   i  o   l  o  g  y  r  e  s  e  a  r  c   h .  o  r  g   b  y   I   P  :   5   4 .   1   5   9 .   4   9 .   1   1   3   O  n  :   S  a   t ,   1   0   S  e  p   2   0   1   6   0   7  :   5   9  :   2   5 Two-dimensional DNA analysis of mycobacteria D673 D6731403 1403 ..................................................................................................... Fig. 2.  Strain differentiation using 2DBGD. M. intracellulare  D673 and 1403 were bothdigested with  Sau 3AI and run in parallel inseparate denaturing gradient gels. Acorresponding area (5  4 cm) from each gel(boxes in large display) shows that many ofthe spots do not align. simultaneously in the ISO-DALT apparatus. The gels weredried prior to exposure to film. Alternatively, a gel could beleft hydrated prior to electroblotting for Southern analysis.2D gel comparisons were carried out by visual inspection.Spot constellations were easily aligned when comparing localareas of approximately 4 cm  . Commercially available soft-ware for comparing 2D protein gels are suitable for suchimage analysis and comparison and we tested Malanie II (Bio-Rad) and NIH Image. However, in our experience, 2DBGDimages were sufficiently reproducible that spot differencescould be detected by simple visual inspection. DNA probes and Southern hybridizations.  Hybridizationswere performed using positively charged nylon membranes(Roche). For Southern transfer of the DNA, 2D gels wereelectroblotted using a DALT blotting kit in the ISO-DALTelectrophoresis tank. DNA probes for hybridization weregenerated using PCR amplification of DNA from  M .  intra - cellulare  D673,  M .  intracellulare  D673-19KDa and  M .  intra - cellulare  D673-Katg. PCR amplification was performed understandard conditions with a programme of 30 cycles of 94   Cfor1min,63   Cfor1 minand72   Cfor1 minandafinalcycleof 72   C for 10 min. Digoxigenin-labelled PCR products weregeneratedforuseasprobesusingoligonucleotides5  -CACCT-ACCGCATCCACGAC-3  katG  –   ,5  -GGTCTCCTCGT-CGTTCAT-3   katG  –   , 5  -GTTCGGGTGGTAACAAG-TCG-3   Kdaf  and 5  -GCCGCTGATCTTGTAGCTGT-3   Kda rev . Prehybridization and hybridization was carried outaccording to the manufacturer’s instructions (Roche). RESULTS We have used a wide range of mycobacterial species andstrains in the development of this methodology. Ourmain purpose in developing the use of 2DBGD is for theidentificationofvirulencedeterminantsinmycobacteria.Thistechniqueenablesamoresubtleexaminationofthemycobacterial genome than other techniques currentlyavailable,whichcanbelimitedtodetectingagainorlossof DNA. In addition to detecting a gain or loss of DNA,2DBGD can be used to identify small changes in DNAsequence as well as changes in intergenic regions. Species differentiation Various species of mycobacteria were resolved using2DBGD. The conditions for resolution of each specieswere determined empirically. On each occasion, no twospecies of mycobacteria produced the same display,regardless of the enzymes used. The choice of enzymesfor use in this work were selected using the  M . tuberculosis  H37Rv genome sequence (Cole  et al ., 1998)and the genome restriction digest tool of the Com-prehensive Microbial Resource (Peterson  et al ., 2001)and included  Hin fI,  Alu I,  Sau 3AI,  Sau 96AI,  Afl III and Nco I. Fig. 1 shows  Alu I displays of the genomes of   M . intracellulare  D673 and  M .  avium  104. The result 3113     D  o  w  n   l  o  a   d  e   d   f  r  o  m   w  w  w .  m   i  c  r  o   b   i  o   l  o  g  y  r  e  s  e  a  r  c   h .  o  r  g   b  y   I   P  :   5   4 .   1   5   9 .   4   9 .   1   1   3   O  n  :   S  a   t ,   1   0   S  e  p   2   0   1   6   0   7  :   5   9  :   2   5 E. M. Dullaghan and others H37RvH37RaH37RvH37Ra 65 1 23465 1 234 ..................................................................................................... Fig. 3.  Identification of differences in avirulent and avirulent laboratory strain of M. tuberculosis .  M. tuberculosis  H37Ra andH37Rv were subjected to 2DBGD afterdigestion with the restriction endonuclease Hin fI. The two samples were run in parallelin separate denaturing gradient gels. Acorresponding area (2  5  3  5 cm) fromH37Rv and H37Ra (boxes in large display)shows numerous spots that are identicalbetween the two strains. However, one spot(4) that is present in H37Rv is absent fromH37Ra while another spot (6) appears tomigrate differently in the two strains. Otherspots of interest are numbered 1, 2, 3 and 5. obtained clearly demonstrates that the resolving powerof this method can distinguish between different speciesof mycobacteria. Strain differentiation To investigate the ability of this technique to resolvestrainsofmycobacteria,twostrainsof  M . intracellulare ,D673 and 1403, were compared using 2DBGD. Thesestrains srcinate from two independent clinical isolatesobtained approximately 50 years ago. Interestingly,analysis of these strains by 2D-PAGE reveals fewer than10 obvious differences over the whole proteome (L. A.Brooks, personal communication). However, Fig. 2demonstrates that no area of   Sau 3AI displays of thesetwo strains can be overlaid. Similar results were ob-tained with  Sau 96I displays (not shown).The application of 2DBGD to compare closely relatedstrainsisdemonstratedwith M . tuberculosis H37Rvandits avirulent counterpart H37Ra, commonly used ref-erence strains that are derived from the same parentstrain, H37 (Steenken  et al ., 1934). These strains areideal candidates for the identification of virulencedeterminants. However, despite the many attempts toidentify the genetic differences between H37Rv andH37Ra, which may explain their different virulence, therestoration of virulence to H37Ra using genes fromH37Rv has not been demonstrated (Brosch  et al .,1999; Pascopella  et al ., 1994; Rindi  et al ., 1999, 2001;Schmidt  et al ., 1998). The comparison of   Hin fI-digestedgenomic DNA enabled the identification of severalfragments (spots) of interest that highlighted differencesbetween these two strains, of which one is shown in Fig.3. These differences are currently being cloned andsequenced. Any novel virulence gene candidates willthen be validated by site-directed mutagenesis. Differentiation of isogenic mutants To test the application of this method for detectingmutations, we subjected two isogenic mutants of   M . intracellulare  D673, which had been constructed in ourlaboratory, to 2DBGD.The first mutant, D673-19KDa, was produced byinsertional mutagenesis using a cassette containing the19 kDa antigen (19Ag) gene of   M .  intracellulare  D673disrupted by the gentamicin-resistance (Gm r ) gene of pUC-GM (Mahenthiralingam  et al ., 1998). The myco-bacterial 19Ag is a highly expressed glycolipoproteinknown to be immunodominant in infected patients andconsidered a candidate virulence factor (Young &Garbe, 1991). Southern analysis of   Afl III-digested geno-mic DNA from  M .  intracellulare  D673 and the 19Agmutant was performed using a digoxigenin-labelled 3114     D  o  w  n   l  o  a   d  e   d   f  r  o  m   w  w  w .  m   i  c  r  o   b   i  o   l  o  g  y  r  e  s  e  a  r  c   h .  o  r  g   b  y   I   P  :   5   4 .   1   5   9 .   4   9 .   1   1   3   O  n  :   S  a   t ,   1   0   S  e  p   2   0   1   6   0   7  :   5   9  :   2   5 Two-dimensional DNA analysis of mycobacteria D673 D67319Ag 1234 19Ag 1234 ..................................................................................................... Fig. 4.  Use of 2DBGD to identify a mutantproduced by insertional mutagenesis using agentamicin-resistance gene. Full displays of M. intracellulare  D673 and the 19Ag mutant(19Ag) following digestion with  Alu I areshown. Corresponding sections (2  2 cm)from 2DBGDs of D673 and the 19Ag mutant(boxes in large display) are shown forcomparison. Spot 3 migrates differently inthe two strains. Spots 1 and 4 are used fororientation purposes. Gm r  cassette as probe. As expected, this produced twohybridization signals in the mutant but none in theparental D673 (data not shown). Therefore, this enzymewas a logical choice for resolution of this mutant,although digests using other restriction enzymes werealsousedtoidentifydifferences.Interestingly,regardlessof which restriction enzyme was used with this mutant,between five and ten visible differences could always beidentified, of which one is illustrated in Fig. 4.The katG geneencodesaninducible,heat-labilecatalaseperoxidase suggested to protect mycobacteria fromreactive oxidative metabolites produced by host phago-cytes (Zhang  et al ., 1992). Directed mutagenesis of the katG  gene of   M .  intracellulare  D673 was previouslyundertaken by generating a frameshift mutation atposition 691 in the coding sequence, resulting in the lossofan Nco I site(Marklund etal ., 1998) toproduceD673-Katg. Genomic DNA from this mutant was comparedwith  M .  intracellulare  D673 using 2DBGD to test theability of this method to identify very minor genomicdifferences. The fusion of the  Nco I fragments at thispositionproduceda12  5 kbfragment,whichistoolargetoberesolvedby2DBGD.However,comparisonof  Alu Idisplays of D673 and the KatG mutant revealed twoshifted spots of approximately 450 bp (Fig. 5). Southernanalysis confirmed the presence of the  katG  gene in thearea of these spots. The KatG display was transferred topositivelychargednylonmembraneandhybridizedwitha digoxigenin-labelled  katG  probe. This enabled orien-tation of the mutated DNA with the hybridized probe.Cloning and sequencing of the DNA represented inthe spots is needed to confirm the identity of thedifferences. DISCUSSION The genetic basis of mycobacterial pathogenesis ispoorly understood due in part to the difficulties asso-ciated with working with the slow-growing pathogenicspecies. Only in recent years have general recombinantDNA techniques become applicable to mycobacterialsystems. Although genome sequence information isavailable for  M .  tuberculosis  and  Mycobacteriumleprae ,itwouldbeunrealistictoexpectthatallstrainsof pathogenic mycobacteria will be sequenced in the nearfuture and therefore we need to develop methodologiesthat will allow us to compare the genomes of individualisolates.Theestablishmentof2DBGD,ahigh-resolutiongenomic fingerprinting technique, will add to the rep-ertoire of tools for identifying genetic alterations asso-ciated with pathogenicity.A number of factors determine the resolution limits of 2DBGD; however, 500–1000 spots can be typicallyresolved. Electrophoresis conditions (time, voltage andtemperature), steepness of the denaturation gradientand acrylamide concentration can all be adjusted to bestresolve any specific size range and G  C content. Thereare, however, areas of poor resolution present in the 3115
Search
Similar documents
View more...
Tags
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks