A two-plasmid system for stable, selective-pressure-independent expression of multiple extracellular proteins in mycobacteria

Günter Harth, Sasa Maslesa-Galic and Marcus A. Horwitz

Division of Infectious Diseases, Department of Medicine, 37-121 CHS, University of California Los Angeles, 10833 Le Conte Avenue, Los Angeles, CA 90095-1688, USA

Correspondence
Marcus A. Horwitz
mhorwitz{at}mednet.ucla.edu


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Recombinant mycobacteria expressing Mycobacterium tuberculosis extracellular proteins are leading candidates for new vaccines against tuberculosis and other mycobacterial diseases, and important tools both in antimycobacterial drug development and basic research in mycobacterial pathogenesis. Recombinant mycobacteria that stably overexpress and secrete major extracellular proteins of M. tuberculosis in native form on plasmids pSMT3 and pNBV1 were previously constructed by the authors. To enhance the versatility of this plasmid-based approach for mycobacterial protein expression, the Escherichia coli/mycobacteria shuttle plasmid pGB9 was modified to accommodate mycobacterial genes expressed from their endogenous promoters. Previous studies showed that the modified plasmid, designated pGB9.2, derived from the cryptic Mycobacterium fortuitum plasmid pMF1, was present at a low copy number in both E. coli and mycobacteria, and expression of recombinant M. tuberculosis proteins was found to be at levels paralleling its copy number, that is, approximating their endogenous levels. Plasmid pGB9.2 was compatible with the shuttle vectors pSMT3 and pNBV1 and in combination with them it simultaneously expressed the M. tuberculosis 30 kDa extracellular protein FbpB. Plasmid pGB9.2 was stably maintained in the absence of selective pressure in three mycobacterial species: Mycobacterium bovis BCG, M. tuberculosis and M. smegmatis. Plasmid pGB9.2 was found to be self-transmissible between both fast- and slow-growing mycobacteria, but not from mycobacteria to E. coli or between E. coli strains. The combination of two compatible plasmids in one BCG strain allows expression of recombinant mycobacterial proteins at different levels, a potentially important factor in optimizing vaccine potency.


The GenBank accession number for the sequence of plasmid pGB9.2 is AY266291.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Tuberculosis (TB), the first disease declared a global emergency by the World Health Organization (WHO), is a leading cause of death, especially in the developing world (Pablos-Mendez et al., 1998). Exacerbating the problem is the worldwide emergence of multi-drug-resistant strains of Mycobacterium tuberculosis, the primary causative agent of TB. This situation has created an urgent need for a vaccine against tuberculosis that is more efficacious than the current vaccine, and for new drugs to combat the disease (Cohn et al., 1997).

Extracellular proteins of M. tuberculosis figure prominently in both new vaccine and drug development (Harth et al., 1997). Extracellular proteins of M. tuberculosis are proteins that are secreted or otherwise released by the bacterium into its extracellular milieu. In the human host, M. tuberculosis principally multiplies in mononuclear phagocytes, in which case extracellular proteins are released into a membrane-bound phagosome (Harth et al., 1994, 1996; Lee & Horwitz, 1995).

In broth culture, M. tuberculosis releases large numbers of proteins into the medium; 12 of these proteins are present in especially large amounts (Horwitz et al., 1995; Jungblut et al., 1999). The major secretory protein of M. tuberculosis, a protein of 30 kDa molecular mass, is a member of a family of four related proteins, FbpA, B, C and D. Three of the four proteins, FbpA (antigen 85A, encoded by fbpA=Rv3804c), FbpB (antigen 85B, encoded by fbpB=Rv1886c) and FbpC (antigen 85C, encoded by fbpC=Rv0129c) form the 30–32 kDa complex of highly homologous mycolyl transferases (Belisle et al., 1997; Cole et al., 1998; Anderson et al., 2001). The fourth protein of ~24 kDa, FbpD (antigen Mpt51, encoded by fbpD=Rv3803c), is related to FbpA, B and C, but it remains unclear whether the protein is an enzymically active mycolyl transferase. The relative activities of the three proteins FbpA, B and C in vivo are also not known.

The 30 kDa protein is of particular interest in new vaccine and drug development. With respect to vaccine development, immunization of guinea pigs with the 30 kDa protein induces substantial protective immunity against aerosol challenge with the highly virulent Erdman strain of M. tuberculosis (Horwitz et al., 1995). Moreover, immunization of guinea pigs with a recombinant BCG overexpressing this protein (rBCG30) induces protective immunity superior to conventional BCG vaccine (Horwitz et al., 2000; Horwitz & Harth, 2003). The rBCG30 vaccine is currently in Phase I human trials. With respect to drug development, the M. tuberculosis 30 kDa protein and the other two highly homologous mycolyl transferases of 32 kDa molecular mass are leading drug targets. Targeting the proteins' gene transcripts (mRNAs of fbpA, fbpB and fbpC), we have previously shown that antisense phosphorothioate oligonucleotides strongly inhibit M. tuberculosis growth in broth culture (Harth et al., 2002). An analysis of the three-dimensional structure of the 30 kDa protein suggested a class of chemical inhibitors (Anderson et al., 2001), and certain trehalose analogues aimed at the mycolyl transferase complex have been shown to inhibit M. tuberculosis growth (Rose et al., 2002; G. Harth, B. Smith, M. Jung & M. A. Horwitz, unpublished results).

A key feature of the rBCG30 vaccine is the use of a plasmid (pMTB30) to overexpress the M. tuberculosis 30 kDa protein. Indeed, it is the overexpression of this protein which is essential to the enhanced efficacy of this vaccine (Horwitz et al., 2000). The overexpression of multiple M. tuberculosis extracellular proteins in BCG or other mycobacterial hosts may result in even more potent vaccines. This may be accomplished by using a single plasmid containing multiple genes cloned in various orientations or by using two or more compatible plasmids. An advantage of the latter approach is that the relative expression of the proteins from different plasmids can be modulated according to plasmid copy number.

In this report, using the 30 kDa protein as an example, we describe the use of plasmid pGB9.2 to overexpress M. tuberculosis extracellular proteins in mycobacteria. Plasmid pGB9.2 is within the same compatibility group as pJAZ plasmids (Bachrach et al., 2000). We here present the entire DNA sequence of the plasmid and demonstrate that: (a) pGB9.2 stably expresses the M. tuberculosis 30 kDa protein in mycobacteria in the absence of selective pressure; (b) pGB9.2 and a second compatible plasmid simultaneously express the 30 kDa protein in mycobacteria; (c) pGB9.2 is present and stably maintained at a low copy number allowing the intentional expression of selected proteins at a lower level than proteins expressed on a high-copy-number plasmid; (d) a 1·3 kb region of the plasmid is necessary for plasmid stability in mycobacteria; and (e) the plasmid is self-transmissible between fast- and slow-growing mycobacteria, but not from mycobacteria to Escherichia coli.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacterial strains, antibiotics and plasmids.
The strains, antibiotics and plasmids used in this study are listed in Table 1. E. coli strains were incubated for 24 h, Mycobacterium smegmatis strains for 3–5 days, and Mycobacterium bovis and M. tuberculosis strains for 2–3 weeks, all at 37 °C, and in the case of the mycobacteria in a 5 % CO2/95 % air atmosphere.


View this table:
[in this window]
[in a new window]
 
Table 1. Bacterial strains, antibiotics and plasmids

 
Introduction of modifications in plasmid pGB9.
The shuttle plasmid pGB9 was modified in its E. coli portion by eliminating 2·8 kb of non-essential regions, including the camr gene, a partial tetr gene and part of the intergenic region of pACYC184 between the tetr gene and the replication origin, ori p15A. The altered version of pGB9 was designated pGB9.2 (Fig. 1). Another deletion eliminated the two PlacZ and lacZ{alpha}-carrying AseI–AseI fragments, which were replaced by a multi-cloning site.



View larger version (34K):
[in this window]
[in a new window]
 
Fig. 1. Maps of plasmid pGB9.2 and its derivative pGB9.2-30. The shuttle vector pGB9.2 is composed of an E. coli portion, containing the kanamycin resistance gene from Tn5 (Tn5aph) and an origin of replication region from plasmid pACYC184 (p15A ori), and a mycobacterial portion, containing the entire M. fortuitum plasmid pMF1 with its mycobacterial replication region. The map also shows the DNA sequence of the multi-cloning site flanked by AseI and HindIII sites. The recombinant shuttle plasmid pGB9.2-30 contains the M. tuberculosis Erdman fbpB gene inserted into the multi-cloning cassette's PacI site. For convenience, the fbpB gene including its relevant functional features is shown in a left-to-right orientation, although the analysed clones were oriented in the opposite direction. The leader peptide spans the first 40 amino acids. The 6 bp stretch at the 5' end of the fbpB cassette is a HindIII site and the 45 bp stretch at the 3' end is from the fbpB genomic locus immediately downstream of the gene's stop codon.

 
Construction of plasmid pGB9.2 expressing the M. tuberculosis 30 kDa protein.
The modified pGB9 plasmid pGB9.2 was assayed for its capacity to express recombinant proteins by inserting the M. tuberculosis fbpB gene into the plasmid's multi-cloning site and driving expression of the protein by its endogenous promoter. The 30 kDa protein gene cassette, consisting of the fbpB coding region preceded by ~500 bp of upstream DNA sequence, was inserted into the PacI site of pGB9.2 (Fig. 1).

Modifications of plasmid pNBV1 for conjugation assays.
The mycobacterial shuttle vector pNBV1(Hygr) was modified by inserting the gene encoding apramycin resistance between the plasmid's two DraI sites to yield plasmid pNBV1(AprrHygr).

Expression of recombinant M. tuberculosis 30 kDa protein.
Supernates from late-exponential-phase cultures (4 days for M. smegmatis 1-2c; 14 days for M. bovis BCG Tice and M. tuberculosis Erdman) were filtered through 0·45 and 0·22 µm acetate filters, then concentrated in Amicon Diaflo filtration units to a final concentration of 1x108 cell equivalents µl–1 and analysed for secreted 30 kDa protein on 12·5 % denaturing polyacrylamide gels. Protein patterns were assessed densitometrically either directly or after immunoblotting using polyvalent rabbit anti-30 kDa protein antibodies and an enhanced chemiluminescent detection kit. The scanned gels and immunoblots were digitized with Adobe Photoshop software and the amounts of 30 kDa protein were expressed in arbitrary units using the NIH Image 1.62 software program. Baseline expression was the amount of endogenous 30 kDa protein expressed by the wild-type mycobacterial strain.

Assessment of plasmid stability.
Stability of the plasmids pGB9.2 and pGB9.2-30 was assayed in M. bovis BCG Tice and M. smegmatis 1-2c by culturing a total of three independent plasmid-harbouring clones in 7H9 broth without kanamycin for 30 generations (30 days for BCG and 5 days for M. smegmatis) and then enumerating c.f.u. of both strains after plating on 7H11 agar medium containing or lacking kanamycin (20 µg ml–1). Genomic and plasmid DNA from strains harbouring plasmid pGB9.2 were quantified densitometrically (Jacobs et al., 1991; Pushnova et al., 2000). In addition, total DNA (various amounts) was transformed into DH5{alpha}, and kanamycin-resistant E. coli bacteria were enumerated to compare the transformation efficiency of pGB9.2 with that of pMTB30 in E. coli.

Assessment of self-transmissibility of pGB9.
For bacterial mating experiments involving various E. coli and/or mycobacterial strains, donor and recipient bacteria were mixed on an Amicon YM3-43 filter at ratios of 1 : 1 or 1 : 10 and incubated for either 4 h or 16 h at 37 °C. Bacteria were washed with 1–2 ml of medium (Luria Bertani or 7H9) and exconjugants were evaluated after plating and incubation on selective medium.


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
DNA sequence features and coding capacity of pGB9 and generation of pGB9.2
The original plasmid pGB9 (Bachrach et al., 2000) was composed of a 9·2 kb mycobacterial portion derived from the M. fortuitum plasmid pMF1 and a 5·6 kb portion derived from various E. coli plasmids for a total size of 14 841 bp. The mycobacterial portion remains mainly unmapped and very little is known about its coding capacity. One of the HindIII restriction fragments within the 9·2 kb mycobacterial DNA sequence contains the origin of replication and encodes a protein of the Rep family of replication proteins required for plasmid replication (Bachrach et al., 2000).

To make pGB9 more amenable to genetic manipulations, we trimmed the E. coli portion of the plasmid to just the gene conferring kanamycin resistance and the replication origin ori p15A, and introduced into the now unique AseI site a multi-cloning site containing the following unique restriction enzyme sites: PacI, SpeI, SwaI, EcoRV and NdeI. The final version of the modified plasmid was ~11·441 kb in size and was designated pGB9.2 (Fig. 1). Based on the known DNA sequences of the IS50L element and the newly introduced multi-cloning site, we established partial DNA sequences for both the E. coli and the mycobacterial portion of pGB9.2; the remaining portions were determined commercially (Sequetech). Of note was the difference in G+C content between the E. coli (~50 mol%) and the mycobacterial portion (~66 mol%). The p15A ori region places the plasmid in the low-copy-number category (~10–15 copies in E. coli); based on total DNA preparations and E. coli transformations, this is also true for the pGB9.2 copy number in mycobacteria (approximately one to two copies). Although transformation of mycobacteria with plasmids carrying the kanamycin resistance gene is sometimes unsuccessful (Garbe et al., 1994), we did not have difficulty transforming M. bovis BCG, M. smegmatis or M. tuberculosis with pGB9.2. The mycobacterial portion (pMF1) is cryptic in all three mycobacterial species as well as in Mycobacterium fortuitum, from which it originated.

Expression of a major M. tuberculosis extracellular protein by pGB9.2
We first assessed expression of recombinant 30 kDa protein in the fast-growing mycobacterium M. smegmatis. The recombinant mycobacterial strain harbouring one to two copies of the plasmid expressed only about two- to threefold the amount of 30 kDa protein that the wild-type strain expressed (30 kDa protein homologue in M. smegmatis is ~27 kDa). In contrast, recombinant M. smegmatis carrying the plasmid pMTB30, which is present at a higher copy number, expressed five- to sixfold more 30 kDa protein than the wild-type strain (Horwitz et al., 2000). Thus, the magnitude of expression is highly correlated with copy number. Next, we generated recombinant slow-growing mycobacteria, BCG Tice and M. tuberculosis Erdman, to determine if expression of recombinant 30 kDa protein is different from that observed in M. smegmatis. Again, the plasmid copy number was relatively low, one to two copies, and the recombinant strains expressed only about two- to threefold the amount of 30 kDa protein that the wild-type strains expressed, while pMTB30 carrying BCG Tice and M. tuberculosis expressed five- to sixfold more 30 kDa protein than the wild-type strains (Horwitz et al., 2000; this study). Most of the expressed 30 kDa protein was secreted in all mycobacterial strains, and protein expression was quantified by scanning gels and immunoblots (Fig. 2, Table 2). As a control, we assessed expression of recombinant 30 kDa protein in E. coli, where we could not detect any expression of this protein, presumably due to the incompatibility of mycobacterial promoters and the E. coli transcription machinery.



View larger version (26K):
[in this window]
[in a new window]
 
Fig. 2. Secretion of M. tuberculosis 30 kDa protein expressed from pGB9.2-30 in various mycobacteria. Concentrated culture filtrates were subjected to gel electrophoresis (using 5x109 c.f.u.) or to gel electrophoresis and immunoblotting (using 5x108 c.f.u.). Coomassie-stained gels (C) and immunoblots (I) were digitized and the images were analysed to yield the indicated amounts of secreted 30 kDa protein. The ‘control’ lane contains 10 µg of recombinant M. tuberculosis 30 kDa protein purified from M. smegmatis 1-2c.

 

View this table:
[in this window]
[in a new window]
 
Table 2. Secretion of endogenous and recombinant M. tuberculosis 30 kDa protein in various mycobacteria

Protein levels, shown in mg l–1, were measured from scanned gels and immunoblots digitized by Adobe Photoshop 5.5 and analysed using NIH Image 1.62 software programs.

 
Expression of a major M. tuberculosis extracellular protein by pGB9.2 and a compatible second plasmid
One approach to generating recombinant BCG strains expressing various M. tuberculosis proteins involves placing two or more compatible plasmids in BCG. We investigated the two-plasmid approach for the pMTB30-carrying strains M. bovis BCG Tice and M. tuberculosis Erdman, both of which expressed five- to sixfold more 30 kDa protein than the wild-type strains (Horwitz et al., 2000; this study). After introduction of the recombinant plasmid pGB9.2-30 into BCG Tice and M. tuberculosis Erdman, expression of 30 kDa protein was additive in both strains (Fig. 3, Table 2), underscoring the usefulness of this two-plasmid expression approach. For M. bovis BCG, 1 l of bacterial culture (~5x1011 c.f.u.) yielded approximately 0·5 mg protein, increasing about five- to sixfold in the presence of pMTB30 and eightfold in the presence of the two plasmids pMTB30 and pGB9.2-30. The same result was observed in M. tuberculosis, which typically showed the highest level of 30 kDa protein expression. While the relative increase in expression of recombinant 30 kDa protein in M. tuberculosis was similar to the increase observed in BCG, the absolute amount of expressed and secreted 30 kDa protein was always higher in M. tuberculosis than in BCG, because wild-type M. tuberculosis cells express about twofold the amount of endogenous 30 kDa protein that BCG cells express.



View larger version (24K):
[in this window]
[in a new window]
 
Fig. 3. Expression and secretion of M. tuberculosis 30 kDa protein in recombinant BCG carrying two plasmids. Concentrated culture filtrates were subjected to gel electrophoresis (using 5x109 c.f.u.) or to gel electrophoresis and immunoblotting (using 5x108 c.f.u.). Stained gels (C) and immunoblots (I) were digitized and the images analysed to yield the indicated amounts of secreted 30 kDa protein. The ‘control’ lane contains 10 µg of recombinant M. tuberculosis 30 kDa protein purified from M. smegmatis 1-2c.

 
Stability of pGB9.2 in mycobacteria
For pGB9.2 to be useful in our vaccine protocols, it was mandatory that it be retained in recombinant BCG bacteria so as to provide continuously expressed recombinant M. tuberculosis 30 kDa protein. Stability of plasmid pGB9.2-30 was assayed in M. bovis BCG Tice and M. smegmatis 1-2c by culturing the bacteria in medium without kanamycin for 30 generations and then evaluating c.f.u. on medium containing or lacking kanamycin. The enumeration of the plated serial dilutions of three independent assays revealed that the c.f.u. counts were very similar: the mean count of three assays for BCG on kanamycin-containing medium was (7·8±0·03)x107 c.f.u. ml–1 and on kanamycin-lacking medium (7·8±0·2)x107 c.f.u. ml–1; the corresponding numbers for M. smegmatis were (5·7±0·1)x108 c.f.u. ml–1 and (5·8±0·2)x108 c.f.u. ml–1. These results clearly demonstrate that pGB9.2-30 is stably maintained by >95 % of all bacteria in a culture over many generations.

To prove that the plasmid was recoverable, we quantified the two plasmids, pGB9.2 and pGB9.2-30, by agarose gel assays and bacterial transformations. From a total of ~5x109 c.f.u., we isolated ~20 µg of DNA which contained ~70 ng of pGB9.2, as determined by gel scans. Since one pGB9.2 plasmid equals ~1·2x10–8 ng DNA, we expected ~60 ng pGB9.2 DNA in 5x109 bacteria if each bacterium carried one copy of pGB9.2. Transformations of purified plasmid DNA and total bacterial DNA from rBCG Tice-pGB9.2 into E. coli DH5{alpha} yielded kanamycin-resistant E. coli bacteria at mean frequencies of ~106 transformants per µg pGB9.2 plasmid DNA and 5000 transformants per µg total bacterial DNA. Using the values mentioned above, we expected 1 µg total DNA to contain ~3 ng pGB9.2 DNA, which would result in ~3000 transformants. This result shows that the recombinant BCG strain most likely harbours one to two copies of pGB9.2. Values for the plasmid pGB9.2-30 were also consistent with this copy number. The same analyses were performed for pMTB30 (~10 kb), the results of which demonstrated that recombinant BCG bacteria harbouring pMTB30 contain ~13–14 copies of pMTB30. The integrity of the plasmids pGB9.2, pGB9.2-30 and pMTB30 was demonstrated by recovering the plasmids from transformed E. coli clones and digesting them with several restriction endonucleases.

Based on previously published data, we knew that a 4·2 kb HindIII restriction fragment of pMF1 contains the mycobacterial replication region (Bachrach et al., 2000). Hence, we considered introducing changes in a location of pMF1 as far removed from the ori region as possible by deleting a 1·3 kb fragment flanked by HindIII and XbaI restriction sites. This deletion mutant grew well in E. coli, but poorly in M. smegmatis where the plasmid was lost by apparently all cells over a 30-generation growth period, since no kanamycin-resistant clones were obtained upon shifting bacteria from medium lacking kanamycin to medium containing the antibiotic. Hence, the 1·3 kb region was necessary for plasmid stability in mycobacteria in the absence of selective pressure.

Bacterial conjugations to assess self-transmissibility of pGB9.2
In a previous report, we showed that the plasmid pSMT3 is neither self-transmissible nor mobilizable, regardless of the bacterial species in which it resides (Horwitz & Harth, 2003). However, we expected this not to be the case for pGB9.2 because it has a much greater coding capacity and belongs to a different compatibility group. To assess the mobility of plasmid pGB9.2(Kanr), we performed a series of bacterial conjugations and compared its mobility with that of pNBV1(AprrHygr) and pSMT3(Hygr) (Table 3). Transmissibility was analysed by investigating the transfer of pGB9.2 from M. smegmatis 1-2c to the wild-type mycobacterium M. tuberculosis Erdman. Exconjugants, plated on medium containing 2-thiopene carboxylic acid hydrazide and kanamycin, arose at a frequency of 5·1x10–7, demonstrating that pGB9.2 is self-transmissible between mycobacterial strains. The genetic background of the recipient mycobacterium does not influence the mobility of pGB9.2 since mixing of M. smegmatis 1-2c carrying pGB9.2 with either M. bovis BCG Tice[pSMT3(Hygr)] or M. tuberculosis Erdman[pNBV1(AprrHygr)] yielded two-plasmid-carrying BCG and M. tuberculosis clones at frequencies of 2·7 and 1·8x10–7, respectively. In contrast, the small shuttle vectors pSMT3 and pNBV1 were not self-transmissible and could not be mobilized by plasmid pGB9.2.


View this table:
[in this window]
[in a new window]
 
Table 3. Frequencies of conjugational plasmid transfers in various bacterial species

 
As a control, we assessed the mobility of the three plasmids, pGB9.2, pNBV1 and pSMT3, between E. coli strains as well as between E. coli and mycobacterial strains. These assays demonstrated that the three plasmids cannot be introduced into E. coli strains by conjugational transfer. Even the presence in an E. coli donor strain of the self-transmissible plasmid R6K did not result in the transfer of any of the three mycobacterial shuttle vectors. However, the E. coli/mycobacteria shuttle vector pVK173T, which carries a functional origin of transfer, can be mobilized from E. coli to mycobacteria with the help of plasmid R6K.


   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
This laboratory's approach to the design of vaccines against tuberculosis and other important mycobacterial infections such as bovine tuberculosis and leprosy is centred on the expression of major mycobacterial extracellular proteins in native form in a heterologous mycobacterial vaccine host. Our demonstration that a recombinant BCG vaccine overexpressing the M. tuberculosis 30 kDa protein is more protective than conventional BCG in the highly relevant guinea pig model of pulmonary tuberculosis has provided impetus for this approach (Horwitz et al., 2000; Horwitz & Harth, 2003). With this study, we are exploring the possibility that expression of multiple mycobacterial proteins in a recombinant BCG strain will yield a vaccine that is even more potent against tuberculosis than a vaccine expressing a single protein. Our approach was to add a second E. coli/mycobacteria shuttle vector, plasmid pGB9.2, to the established vectors pSMT3 and pNBV1, because these plasmids are compatible and this strategy may overcome certain limitations of a one-plasmid approach.

Used as single plasmids, the high-copy-number plasmids pNBV1 and pSMT3 express mycobacterial proteins at many multiples of their endogenous level in the wild-type host. In contrast, pGB9.2 is a low-copy-number plasmid, allowing expression of heterologous proteins at low levels. Such low-level expression may more closely resemble the level of the protein in the native mycobacterial host, where the gene encoding the extracellular protein is present as a single copy in the bacterial chromosome.

In a two-plasmid approach, pGB9.2 can be added as a second plasmid to a recombinant BCG bacterium carrying either pNBV1 or pSMT3. This strategy may have several advantages. Firstly, where overexpression of multiple proteins is involved, it decreases the stress on the primary plasmid, which otherwise would be forced to maintain two, three, four or more M. tuberculosis genes. Adding additional genes to the same plasmid typically increases the likelihood of a recombination event and the likelihood that plasmid copy number will decrease (G. Harth & M. A. Horwitz, unpublished results). Secondly, the availability of a system involving a low-copy-number plasmid in combination with a high-copy-number plasmid provides the flexibility of expressing some recombinant proteins at low levels and others at high levels. As live recombinant mycobacterial vaccines become more complex and involve the expression of multiple heterologous proteins, optimizing their immunoprotective capacity may require modulating the expression of certain proteins relative to others. Too high an expression of all recombinant proteins may interfere with the immune response to one of the proteins, including endogenous proteins of the recombinant BCG host, or result in other untoward effects, such as the induction of tolerance or a Th2 type of immune response. Thirdly, transcriptional regulation of gene expression is more readily accomplished if the copy number of a target sequence such as a promoter element is low, because the target can be more readily saturated with binding factors, which otherwise might be diluted out were there too many target copies present in a cell.

The utility of pGB9.2 might be enhanced by further investigation of several aspects of the plasmid. Firstly, even though we trimmed down the plasmid to 11·4 kb from its original size of 14·8 kb, the mycobacterial portion remains largely unmapped. It would be helpful to know if further deletions could be introduced without compromising useful features such as self-transmissibility and plasmid stability in the absence of selective pressure. Secondly, while plasmid stability in the absence of selective pressure was demonstrated in vitro, it would be important to confirm plasmid stability in vivo in an infection model, such as guinea pigs, were this two-plasmid approach ever found to be superior to our current one-plasmid recombinant BCG approach, and thus worth consideration as an improved version of BCG. Thirdly, although intended for the study of regulatory proteins, several groups of investigators have also developed two-plasmid systems or used mycobacterial plasmids of different compatibility groups to develop additional E. coli–mycobacteria shuttle vectors (Picardeau & Vincent, 1997; Gavigan et al., 1997; Kaps et al., 2001; Pashley et al., 2003). Presumably, our system would also be appropriate for such studies and merits investigation in this regard.


   ACKNOWLEDGEMENTS
 
This work was supported by NIH grant AI 31338.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Anderson, D. H., Harth, G., Horwitz, M. A. & Eisenberg, D. (2001). An interfacial mechanism and a class of inhibitors inferred from two crystal structures of the Mycobacterium tuberculosis 30 kDa major secretory protein (antigen 85B), a mycolyl transferase. J Mol Biol 307, 671–681.[CrossRef][Medline]

Bachrach, G., Colston, M. J., Bercovier, H., Bar-Nir, D., Anderson, C. & Papavinasasundaram, K. G. (2000). A new single-copy mycobacterial plasmid, pMF1, from Mycobacterium fortuitum which is compatible with the pAL5000 replicon. Microbiology 146, 297–303.[Abstract/Free Full Text]

Belisle, J. T., Vissa, V. D., Sievert, T., Takayama, K., Brennan, P. J. & Besra, G. S. (1997). Role of the major antigen of Mycobacterium tuberculosis in cell wall biogenesis. Science 276, 1420–1422.[Abstract/Free Full Text]

Cheng, L. W., Anderson, D. M. & Schneewind, O. (1997). Two independent type III secretion mechanisms for YopE in Yersinia enterocolitica. Mol Microbiol 24, 757–765.[Medline]

Cohn, D. L., Bustreo, F. & Raviglione, M. C. (1997). Drug-resistant tuberculosis: review of the worldwide situation and the WHO/IUATLD Global Surveillance Project. International Union against Tuberculosis and Lung Disease. Clin Infect Dis 24 (Suppl 1), S121–S130.[Medline]

Cole, S. T., Brosch, R., Parkhill, J. & 39 other authors (1998). Deciphering the biology of Mycobacterium tuberculosis from the complete genome sequence. Nature 393, 537–544.[CrossRef][Medline]

Garbe, T. R., Barathi, J., Barnini, S., Zhang, Y., Abou-Zeid, C., Tang, D., Mukherjee, R. & Young, D. B. (1994). Transformation of mycobacterial species using hygromycin resistance as selectable marker. Microbiology 140, 133–138.[Abstract]

Gavigan, J. A., Ainsa, J. A., Perez, E., Otal, I. & Martin, C. (1997). Isolation by genetic labeling of a new mycobacterial plasmid, pJAZ38, from Mycobacterium fortuitum. J Bacteriol 179, 4115–4122.[Abstract]

Harth, G., Clemens, D. L. & Horwitz, M. A. (1994). Glutamine synthetase of Mycobacterium tuberculosis: extracellular release and characterization of its enzymatic activity. Proc Natl Acad Sci U S A 91, 9342–9346.[Abstract/Free Full Text]

Harth, G., Lee, B.-Y., Wang, J., Clemens, D. L. & Horwitz, M. A. (1996). Novel insights into the genetics, biochemistry, and immunocytochemistry of the 30-kilodalton major extracellular protein of Mycobacterium tuberculosis. Infect Immun 64, 3038–3047.[Abstract]

Harth, G., Lee, B.-Y. & Horwitz, M. A. (1997). High-level heterologous expression and secretion in rapidly growing nonpathogenic mycobacteria of four major Mycobacterium tuberculosis extracellular proteins considered to be leading vaccine candidates and drug targets. Infect Immun 65, 2321–2328.[Abstract]

Harth, G., Horwitz, M. A., Tabatadze, D. & Zamecnik, P. C. (2002). Targeting the Mycobacterium tuberculosis 30/32-kDa mycolyl transferase complex as a therapeutic strategy against tuberculosis: proof of principle by using antisense technology. Proc Natl Acad Sci U S A 99, 15614–15619.[Abstract/Free Full Text]

Herrmann, J. L., O'Gaora, P., Gallagher, A., Thole, J. E. R. & Young, D. B. (1996). Bacterial glycoproteins: a link between glycosylation and proteolytic cleavage of a 19 kDa antigen from Mycobacterium tuberculosis. EMBO J 15, 3547–3554.[Abstract]

Horwitz, M. A. & Harth, G. (2003). A new vaccine against tuberculosis affords greater survival after challenge than the current vaccine in the guinea pig model of pulmonary tuberculosis. Infect Immun 71, 1672–1679.[Abstract/Free Full Text]

Horwitz, M. A., Lee, B.-W. E., Dillon, B. J. & Harth, G. (1995). Protective immunity against tuberculosis induced by vaccination with major extracellular proteins of Mycobacterium tuberculosis. Proc Natl Acad Sci U S A 92, 1530–1534.[Abstract]

Horwitz, M. A., Harth, G., Dillon, B. J. & Maslesa-Galic, S. (2000). Recombinant BCG vaccines expressing the Mycobacterium tuberculosis 30 kDa major secretory protein induce greater protective immunity against tuberculosis than conventional BCG vaccines in a highly susceptible animal model. Proc Natl Acad Sci U S A 97, 13853–13858.[Abstract/Free Full Text]

Howard, N. S., Gomez, J. E., Ko, C. & Bishai, W. R. (1995). Color selection with a hygromycin-resistance-based Escherichia coli-mycobacterial shuttle vector. Gene 166, 181–182.[CrossRef][Medline]

Jacobs, W. R., Jr, Kalpana, G. V., Cirillo, J. D., Pascopella, L., Snapper, S. B., Udani, R. A., Jones, W., Barletta, R. G. & Bloom, B. R. (1991). Genetic systems for mycobacteria. Methods Enzymol 204, 537–555.[Medline]

Jungblut, P. R., Schaible, U. E., Mollenkopf, H. J. & 7 other authors (1999). Comparative proteome analysis of Mycobacterium tuberculosis and Mycobacterium bovis BCG strains: towards functional genomics of microbial pathogens. Mol Microbiol 33, 1103–1117.[CrossRef][Medline]

Kaps, I., Ehrt, S., Seeber, S., Schnappinger, D., Martin, C., Riley, L. W. & Niederweis, M. (2001). Energy transfer between fluorescent proteins using a co-expression system in Mycobacterium smegmatis. Gene 278, 115–124.[CrossRef][Medline]

Lee, B.-Y. & Horwitz, M. A. (1995). Identification of macrophage and stress-induced proteins of Mycobacterium tuberculosis. J Clin Invest 96, 245–249.[Medline]

Pablos-Mendez, A., Raviglione, M. C., Laszlo, A. & 8 other authors (1998). Global surveillance for antituberculosis-drug resistance, 1994–1997. World Health Organization–International Union against Tuberculosis and Lung Disease Working Group on Anti-Tuberculosis Drug Resistance Surveillance. N Engl J Med 338, 1641–1649.[Abstract/Free Full Text]

Paget, E. & Davies, J. (1996). Apramycin resistance as a selective marker for gene transfer in mycobacteria. J Bacteriol 178, 6357–6360.[Abstract]

Pashley, C. A., Parish, T., McAdam, R. A., Duncan, K. & Stoker, N. G. (2003). Gene replacement in mycobacteria by using incompatible plasmids. Appl Environ Microbiol 69, 517–523.[Abstract/Free Full Text]

Picardeau, M. & Vincent, V. (1997). Characterization of large linear plasmids in mycobacteria. J Bacteriol 179, 2753–2756.[Abstract]

Pushnova, E. A., Geier, M. & Zhu, Y. S. (2000). An easy and accurate agarose gel assay for quantitation of bacterial plasmid copy numbers. Anal Biochem 284, 70–76.[CrossRef][Medline]

Rose, J. D., Maddry, J. A., Comber, R. N., Suling, W. J., Wilson, L. N. & Reynolds, R. C. (2002). Synthesis and biological evaluation of trehalose analogs as potential inhibitors of mycobacterial cell wall biosynthesis. Carbohydr Res 337, 105–120.[CrossRef][Medline]

Tullius, M. V., Harth, G. & Horwitz, M. A. (2003). Glutamine synthetase GlnA1 is essential for growth of Mycobacterium tuberculosis in human THP-1 macrophages and guinea pigs. Infect Immun 71, 3927–3936.[Abstract/Free Full Text]

Received 24 February 2004; revised 20 April 2004; accepted 26 April 2004.



This Article
Abstract
Full Text (PDF)
Alert me when this article is cited
Alert me if a correction is posted
Citation Map
Services
Email this article to a friend
Similar articles in this journal
Similar articles in PubMed
Alert me to new issues of the journal
Download to citation manager
Google Scholar
Articles by Harth, G.
Articles by Horwitz, M. A.
Articles citing this Article
PubMed
PubMed Citation
Articles by Harth, G.
Articles by Horwitz, M. A.
Agricola
Articles by Harth, G.
Articles by Horwitz, M. A.


HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
INT J SYST EVOL MICROBIOL MICROBIOLOGY J GEN VIROL
J MED MICROBIOL ALL SGM JOURNALS
Copyright © 2004 Society for General Microbiology.