INFECTION AND IMMUNITY, Jan. 1993, p. 260-267
Vol. 61, No. 1
0019-9567/93/010260-08$02.00/0
Copyright X 1993, American Society for Microbiology
Expression of the Mycobacterium tuberculosis 19-Kilodalton
Antigen in Mycobacterium smegmatis: Immunological
Analysis and Evidence of Glycosylation
THOMAS GARBE,t* DAVID HARRIS, MARTIN VORDERMEIER, RAJU LATHIGRA,4
JURAJ IVANYI, AND DOUGLAS YOUNG
Medical Research Council Tuberculosis and Related Infections Unit, Royal Postgraduate
Medical School, Hammersmith Hospital, Ducane Road,
London W12 OHS, United Kingdom
Received 8 July 1992/Accepted 21 October 1992
In addition to the utility of such systems for expressing a
range of heterologous antigens from other unrelated pathogens, the same approach is also convenient for transferring
genes between mycobacterial species. It has been demonstrated, for example, that genes encoding the superoxide
dismutase enzymes of M. leprae and M. tuberculosis are
expressed from their own promoters in Mycobacterium
smegmatis (a rapid-growing species suitable for laboratory
manipulation), while provision of an exogenous promoter
was essential for expression of the same genes in E. coli (26,
33). In addition, it was found that the mycobacterial system
allowed expression of functionally active superoxide dismutase, in contrast to the enzymatically inactive recombinant
product in E. coli (33). In the present study, we have
examined the use of a mycobacterial expression system to
facilitate the immunological characterization of a 19-kDa
antigen from M. tuberculosis. The 19-kDa antigen was
originally identified by using a set of murine monoclonal
antibodies binding to M. tuberculosis and a limited number
of nontuberculous mycobacteria (5) and was subsequently
shown to elicit both humoral and cell-mediated immune
responses in mice and patients with tuberculosis (7, 9, 13).
Results from nucleotide sequence analysis (2) and biochemical characterization of the 19-kDa antigen (31) suggest that
the mature protein is secreted across the cell membrane and
is present as a lipoprotein in M. tuberculosis. Similarly,
posttranslational acylation of the corresponding protein from
Mycobacterium avium-intracellulare has been suggested
(20), and analysis of the 19-kDa antigen purified from M.
bovis has provided evidence of glycosylation (8). The 19-kDa
antigen has no marked sequence homology with other
known proteins, and its biochemical function has not yet
been established.
We report here on the application of an M. smegmatis
expression system to study posttranslational modification
and T-cell recognition of the 19-kDa antigen from M. tuber-
An extensive panel of mycobacterial proteins involved in
recognition by the host immune system has been identified
by biochemical fractionation or by screening of recombinant
DNA expression libraries (reviewed in reference 32). Several of these antigens have been proposed as potential
targets for improved diagnostic tests or for incorporation
into novel subunit vaccines, but difficulties in obtaining
sufficient quantities of the purified reagents have impeded
comparative experimental testing of such suggestions (13,
30, 32). In addition to the requirement for strict containment
facilities, pathogenic mycobacteria grow very slowly (Mycobacterium tuberculosis doubling time, 24 h) or not at all
(Mycobacterium leprae) in laboratory culture, presenting
significant practical obstacles to large-scale growth for biochemical fractionation. For some antigens, members of
conserved heat shock protein families, for example, these
problems have been overcome by high-level expression of
the relevant genes in standard Escherichia coli recombinant
DNA systems (18, 27). Several other antigens have been
expressed as fusion proteins in E. coli but have proved
difficult to overexpress as free proteins (9, 17, 29). This latter
class includes proteins containing signal sequences or other
features indicative of a requirement for posttranslational
modification (6, 8, 17, 30, 31). It is attractive to speculate
that a mycobacterial host may provide the optimal system
for expression of such antigens.
The recent development of techniques and vectors for
transformation of mycobacteria (14, 21, 23, 24) has been
stimulated by the goal of creating a new generation of
recombinant Mycobacterium bovis BCG vaccines (1, 14, 25).
*
Corresponding author.
t Present address: Department of Molecular Genetics, Biochem-
istry and Microbiology, University of Cincinnati College of Medicine, 3110 Medical Sciences Building, 231 Bethesda Avenue, Cincinnati, OH 45267-0524.
: Present address: MedImmune Inc, Gaithersburg, MD 20878.
culosis.
260
Downloaded from http://iai.asm.org/ on December 29, 2014 by guest
The gene encoding a 19-kDa antigen from Mycobacterium tuberculosis was expressed as a recombinant
protein in the rapid-growing species Mycobacterium smegmatis. The recombinant antigen was expressed at a
level approximately ninefold higher than in M. tuberculosis and, like the native antigen, was found in the pellet
fraction after high-speed centrifugation of bacterial extracts. The 19-kDa antigen in crude bacterial extracts,
and the purified recombinant antigen, bound strongly to concanavalin A, indicating the possibility of
posttranslational glycosylation. The recombinant antigen stimulated T-cell proliferation in vitro when added to
assays either in the form of whole recombinant bacteria or as a purified protein. Homologous expression of
mycobacterial antigens in a rapid-growing mycobacterial host may be particularly useful for the immunological
characterization of proteins which are subject to posttranslational modification.
RECOMBINANT ANTIGENS IN M. SMEGMATIS
VOL. 61, 1993
MATERIALS AND METHODS
medium with 50 ,ug of kanamycin sulfate per ml in a rotary
incubator at 150 rpm. Bacteria were pelleted for 10 min at
10,000 x g, resuspended in 50 ml of water, and sonicated for
a total of 10 min (MSE Soniprep 150, 19-mm probe, at
maximum output). Cell debris was removed by centrifugation for 10 min at 16,000 x g, and the supernatant was
centrifuged at 48,000 x g overnight at 20°C. The resulting
pellet was resuspended in 10 ml of water, urea was added at
480 mg/ml, and the suspension was rolled overnight at 4°C.
After centrifugation at 230,000 x g for 3 h at 20°C, a turbid
pellet was obtained on top of a much-larger, translucent,
reddish-brown pellet. The upper, turbid, pellet was resuspended in 8 M urea to a final volume of 25 ml and centrifuged
for a further 3 h at 230,000 x g. The turbid grey pellet was
resuspended in a minimum volume of 8 M urea and then
diluted 1:10 in water and centrifuged for 3 h at 13,000 rpm
(Sorvall, GSA rotor). The resulting pellet was resuspended
overnight at 4°C in an equal volume of 50 mM Tris HCl (pH
8.0) containing 2% (vol/vol) Triton X-100 and 5 mM EDTA.
After centrifugation for 3 h at 27,000 x g, the resulting white
pellet was resuspended in 5 ml of 50 mM Tris HCl (pH 8.0)
containing 5% SDS and 10% 3-mercaptoethanol and held on
a boiling water bath for 5 min. The sample was centrifuged at
10,000 x g for 10 min at 4°C, and the supernatant (with 10%
sucrose added) was loaded onto a Biogel P-100 gel bed (2.6
by 100 cm; Bio-Rad) equilibrated with 50 mM Tris HCl (pH
8.0)-1% SDS-1% 3-mercaptoethanol. The 19-kDa antigen
eluted between 137 and 153 ml.
Antigens and synthetic peptides for T-cell assays. A heatkilled preparation of M. tuberculosis H37Ra was obtained
from Difco. A recombinant protein consisting of the M.
tuberculosis 19-kDa antigen fused to glutathione S-transferase (rGST19) was expressed in E. coli and isolated as
described in detail elsewhere (9). A synthetic peptide, p19.7,
corresponding to residues 61 to 80 of the M. tuberculosis
19-kDa antigen (VTGSVVCITAAGNVNIAIGG), was synthesized by simultaneous solid-phase multiple-peptide technology as previously described (9).
Murine T-celi line. A CD4+ murine T-cell line specific for
the M. tuberculosis 19-kDa antigen was generated as follows. C57BV10 mice were immunized in the hind footpads
with a total of 50 ,ug of rGST19 emulsified in incomplete
Freund's adjuvant. Seven days later, the draining popliteal
lymph node cells were removed, and single-cell suspensions
were prepared in complete tissue culture medium (RPMI1640 medium supplemented with 5% fetal calf serum
[GIBCO, Paisley, Scotland], 5 x 10' M P-mercaptoethanol, 2 mM L-glutamine, 100 U of penicillin per ml, and 100 ,g
of streptomycin sulfate per ml). Primed lymph node cells
were cultured in 24-well plates (Nunc, Roskilde, Denmark)
at a concentration of 4 x 106 cells per well in the presence of
20 ,ug of rGST19 per ml. After 6 days, viable cells were
recovered by centrifugation over Ficoll gradients and recultured at a concentration of 0.5 x 106 cells per well together
with 3 x 106 irradiated syngeneic spleen cells as antigenpresenting cells. After 5 days of rest, 0.5 x 106 cells per well
were restimulated in the presence of irradiated antigenpresenting cells and 20 ,ug of rGST19 per ml. A stable cell
line was maintained by the same cycles of rest and restimulation for more than 6 months.
T-celi proliferation assays. Proliferation assays with the
rGST19 T-cell line were performed at the end of a resting
cycle. T cells (2 x 104 cells per well) were added in triplicate
to 96-well flat-bottom microtiter plates (Nunc) containing
antigen diluted to the appropriate concentration and 3 x 101
irradiated syngeneic spleen cells per well. Microcultures
Downloaded from http://iai.asm.org/ on December 29, 2014 by guest
Bacterial strains and plasmids. M. smegmatis 1-2c is a
derivative of M. smegmatis mc26 (14), which shows high
efficiency of transformation (33). M. smegmatis was grown
in Middlebrook 7H9 medium (Difco Laboratories; Detroit,
Mich.) supplemented with glucose (2%, wt/vol). Kanamycin
sulfate was added at 50 ,ug/ml for culture of strains transformed with shuttle plasmids. M. tuberculosis H37Rv is a
virulent strain originally isolated from a tuberculosis patient
and was supplied by B. W. Allen (Royal Postgraduate
Medical School, Hammersmith Hospital, London, United
Kingdom). M. tuberculosis was grown on Middlebrook 7H11
agar plates, supplemented with 0.05% Tween 80 (BDH) and
OADC (oleic acid, albumin, dextrose, catalase [Difco]).
Protein extracts were prepared from M. tuberculosis harvested from plates and disrupted in distilled water by using
glass beads as described previously (13). The total protein
concentration was estimated by using a protein assay system
supplied by Bio-Rad Laboratories (Richmond, Calif.) with
bovine serum albumin (BSA) as the standard. E. coli JM105
(Pharmacia) and TG1 (22) were grown on Luria-Bertani
medium with 50 Wg of kanamycin sulfate per ml added as
described by Sambrook et al. (22). pBAK-7q is a derivative
of pBAK14, a shuttle plasmid capable of replicating in
mycobacteria and in E. coli (33), containing a recombinant
1.8-kb SmaI fragment which includes the structural gene
encoding the M. tuberculosis 19-kDa antigen (2, 9).
DNA manipulation. Plasmid DNA was prepared by standard procedures and analyzed by restriction enzyme digestion and agarose gel electrophoresis as described by Sambrook et al. (22). Transformation of M. smegmatis with
shuttle plasmids was carried out by electroporation as described previously (33). E. coli was transformed by using
standard methods (22).
Gel electrophoresis and Western blotting (immunoblotting).
Electrophoresis in polyacrylamide gels containing sodium
dodecyl sulfate (SDS-PAGE) and blotting onto nitrocellulose
membranes was carried out by using standard procedures
(15, 28). Samples used for Western blot analysis contained
0.3 to 3 ,ug of total protein. Antigens were stained on
nitrocellulose membranes by using monoclonal antibodies
specific for the 19-kDa antigen, namely, TB23 (3, 5), HYT6
(5), and F29-47 (5), by using techniques described previously
(33). Quantitative analysis of Western blots was carried out
by using a Shimadzu CS-9000 dual-wavelength, flying-spot
scanning densitometer at 550 nm. Results are expressed in
terms of peak area as relative absorbance units.
The procedures for staining of nitrocellulose blots with
peroxidase-conjugated concanavalin A (ConA) were essentially identical to those used for antibody staining. Nonspecific binding was blocked by incubating blots for 1 h with 4%
(wt/vol) BSA in phosphate-buffered saline (PBS) with Triton
X-100 (0.2%, vol/vol). After repeated washes with PBS and
PBS-Triton X-100, the blots were incubated with 20 ml of
ConA-peroxidase conjugate (0.2 purpurogallin units per ml;
Sigma) in 2% (wt/vol) BSA in PBS-Triton X-100 for 1 h.
After further washing, blots were stained for peroxidase
activity by adding 3,3'-diaminobenzidine HCI and hydrogen
peroxide in PBS.
Purification of the recombinant 19-kDa antigen. The recombinant 19-kDa antigen was purified by a novel procedure
based on exploitation of its relative insolubility in urea and
subunit molecular weight. M. smegmatis 1-2c transformed
with pBAK-7q was grown at 37°C for 4 days in six 2-liter
conical flasks containing 500 ml of Middlebrook 7H9-glucose
261
262
GARBE ET AL.
INFECT. IMMUN.
kDa
77
66
43
30 -0
otw ~d
HindllI1'1
XhoI-f
*W.
4t
pBAK-7q
10.0 kB
0
17
BamHI
12
1 2 3 4 5 6 7 8
EcoRV
incubated for 3 days at 37°C in an atmosphere of 5%
CO2 and then radiolabelled with [3H]thymidine (37 kBq per
well; Amersham International, Amersham, United Kingdom). After a further 6 to 8 h, cells were harvested onto
glass-fiber filter paper and radioactive incorporation was
determined by liquid scintillation counting (29).
Proliferation assays with intact mycobacteria. For use in
T-cell proliferation assays, M. smegmatis was harvested
from the logarithmic phase of growth and washed with sterile
were
PBS. The bacterial count was estimated on the basis of A600,
and samples (corresponding to 105 to 107 CFU/ml) were
either added directly to T-cell proliferation assays or first
killed by heating for 20 min at 60°C. Cultures were incubated
for 3 days, and proliferation was assessed by incorporation
of radiolabelled thymidine as described above. The addition
of live M. smegmatis did not result in any significant
increase in thymidine incorporation; it is likely that the
presence of streptomycin sulfate in the medium was sufficient to inhibit bacterial multiplication during the assay.
RESULTS
Expression of the 19-kDa antigen in M. smegmatis. The gene
encoding the 19-kDa antigen of M. tuberculosis was excised
from pRL19k2.8 (2, 9) by digestion with SmaI, generating a
1.8-kb fragment containing the structural gene flanked by
approximately 0.9 (5') and 0.4 (3') kb of additional DNA.
The SmaI fragment was inserted into the ScaI site of the
mycobacterial shuttle vector pBAK14 to prepare pBAK-7q
(Fig. 1). pBAK-7q was introduced into E. coli JM105, and a
small-scale plasmid preparation was used to transform M.
smegmatis 1-2c. Extracts from E. coli and M. smegmatis
recombinants were screened for antigen expression by Western blot analysis (Fig. 2). Very little expression of the 19-kDa
antigen was detected in E. coli transformed with pBAK-7q,
but a prominent band was observed in the recombinant M.
smegmatis extracts. The recombinant antigen expressed in
M. smegmatis, like the native protein in M. tuberculosis,
was found predominantly in the cell wall or membrane
fractions generated by centrifugation of sonicated bacterial
FIG. 2. Expression and solubility of the native and recombinant
19-kDa antigen. (Lanes 1 to 6) Solubility of native and recombinant
19-kDa antigen. Extracts prepared from glass-bead-disrupted M.
tuberculosis and M. smegmatislpBAK-7q were centrifuged in an
MSE MicroCentaur centrifuge at 10,000 rpm for 5 min. Supernatant
fractions (lOks) were further separated into supernatant (SOks) and
pellet (50kp) fractions after centrifugation for 1 h at 230,000 x g.
Extracts were analyzed by SDS-PAGE in gels containing 15%
(wt/vol) acrylamide and by Western blotting with monoclonal antibody HYT6. Lanes: 1 to 3, M. tuberculosis extracts (1, lOks [6 pug of
total protein]; 2, 50ks [4 ,ug of total protein]; 3, 50kp [12 p.g of total
protein]); 4 to 6, M. smegmatis/pBAK-7q extracts (4, lOks [6 p,g of
total protein]; 5, 50ks [4 p,g of total protein]; 6, 50kp [12 pg of total
protein]). (Lanes 7 to 10) Recombinant expression in M. smegmatis
and E. coli. Unfractionated extracts from M. smegmatis and E. coli
were analyzed for expression of the 19-kDa antigen. Lanes: 7,
M. smegmatis/pBAK14; 8, M. smegmatislpBAK-7q; 9, E. colil
pBAK14; 10, E. coli/pBAK-7q.
extracts at 230,000 x g (Fig. 2). To compare the level of
expression of the recombinant 19-kDa antigen in M. smegmatis with that in M. tuberculosis, samples containing
various amounts of total protein were analyzed by Western
blotting by using two different monoclonal antibodies, and
the results were quantitated by use of a scanning densitometer (Fig. 3). From the intensity of Western blot staining, the
level of expression of the 19-kDa antigen in recombinant M.
60000
50000
40000
I-
'19
30000
a
a.
20000
10000
0
800
400
600
Protein [g.g/miJ
FIG. 3. Quantitative analysis of 19-kDa expression in M. tuberculosis and recombinant M. smegmatis. Protein extracts from M.
tuberculosis and M. smegmatis/pBAK-7q were analyzed by Western blotting with monoclonal antibodies HYT6 and F29-47. Antibody binding to the 19-kDa band was quantitated by use of a
scanning densitometer and is shown plotted against protein concentration. The level of expression in M. smegmatislpBAK-7q was
approximately ninefold higher than that in M. tuberculosis.
0
200
Downloaded from http://iai.asm.org/ on December 29, 2014 by guest
FIG. 1. pBAK-7q, shuttle vector expressing the 19-kDa antigen.
A 1.8-kb SmaI fragment containing the gene encoding the 19-kDa
antigen (2) was inserted into the Scal site of pBAK14, a shuttle
plasmid stable in mycobacteria and E. coli (33). Abbreviations:
Thl9kD, SmaI fragment from pRL19k2.8 (2); ALori, origin of
replication from mycobacterial plasmid pAL5000 (21); KanR, kanamycin resistance gene from TnS.
9 10
RECOMBINANT ANTIGENS IN M. SMEGMATIS
VOL. 61, 1993
TABLE 1. Inhibition of ConA binding by a-methyl mannosidea
A
kDa
Sample
77
66
43
30
17
263
M. tuberculosis
M. smegmatis/pBAK-7q
__u'-4b
12 -
_2
of":
100
100
65
50
53
34
0
0
a Nitrocellulose blots
prepared with protein extracts from M. tuberculosis
and M. smegmatis/pBAK-7q were screened for ConA binding as described in
the legend to Fig. 4, except that different concentrations of a-methyl mannoside were added during incubation with peroxidase-conjugated ConA. ConA
binding to the 19-kDa band on blots was quantitated by use of a scanning
a-
1
% Absorbance at a-methyl mannoside
concn
10 mM
100 mM
0
IM
3
B
densitometer.
b Results are expressed as a percentage of the absorbance in control lanes
without a-methyl mannoside.
I
66
43
30
--
17
12
1 2 3
FIG. 4. Carbohydrate association with the 19-kDa antigen. (A)
Neat extract from E. coli/pBAK-7q (lane 1) and 10-2 diluted extract
from M. smegmatislpBAK-7q (lane 3) were analyzed by Western
blotting with monoclonal antibody F29-47 (1/2,000 dilution). The
apparent molecular mass of the 19-kDa antigen expressed in E. coli
was approximately 4 kDa lower than that of the antigen in M.
smegmatis. This difference was most obvious when the two extracts
were run together in the same lane of the gel (lane 2). (B) Protein
extracts from M. tuberculosis (lane 1) and M. smegmatis transformed with vector alone (pBAK14, lane 2) or with the 19-kDa gene
(pBAK-7q, lane 3) were analyzed by SDS-PAGE and subsequent
staining with peroxidase-conjugated ConA. A ConA-positive band
was seen in the position of the 19-kDa antigen in M. tuberculosis and
in the extract from recombinant M. smegmatis/pBAK-7q.
smegmatis was estimated to be approximately ninefold
higher than that in M. tuberculosis. This increase would be
consistent with the presence of multiple copies of the plasmid-encoded gene in the recombinant strain, and it is probable that expression of the recombinant 19-kDa antigen is
regulated by recognition of its own expression signals in the
mycobacterial host. The absence of a major additional
19-kDa band in gels stained with Coomassie blue (not
shown) indicates that the recombinant product accounts for
no more than 1% of the total protein in the M. smegmatis
extracts.
Carbohydrate associated with the 19-kDa antigen. A low
level of expression of the 19-kDa antigen was detected by
immunoblot of E. coli transformed with pBAK-7q. We noted
that the antigen expressed in E. coli migrated with a different
apparent molecular weight during SDS-PAGE than that of
the native M. tuberculosis antigen and the recombinant M.
smegmatis product (Fig. 4A). This molecular mass difference, corresponding to approximately 4 kDa, was most
apparent when extracts from the two recombinant strains
were combined and run in a single lane on the gel (Fig. 4A,
lane 2). A further difference was noted when blots were
stained with peroxidase-conjugated ConA, a lectin specific
for t-D-mannose and a-D-glucose, for detection of carbohydrate residues. This procedure highlighted a number of
discrete bands in M. tuberculosis extracts, including a prominent band in the position of the 19-kDa antigen (Fig. 4B,
lane 1). Fewer ConA-positive bands were seen in extracts
from M. smegmatis, but an intense staining of the 19-kDa
protein was strikingly evident in the recombinant extract
(Fig. 4B, lane 3). ConA binding to the 19-kDa antigen in M.
tuberculosis and M. smegmatis was completely inhibited by
inclusion of a-methyl mannoside during incubation with
ConA-peroxidase (Table 1). The lower-molecular-mass 19kDa antigen expressed in E. coli/pBAK-7q was not stained
by ConA, although weak ConA binding could be detected in
blots prepared by loading high concentrations of rGST19
(data not shown). The altered electrophoretic mobility and
ConA affinity are indicative of posttranslational modification
of the mycobacterium-expressed antigen, although our results do not exclude the possibility of a very tight binding
between the 19-kDa antigen and some mycobacterium-specific carbohydrate moiety, which remains intact even during
electrophoresis in the presence of SDS. Staining of blots
with an antibody directed to mycobacterial lipoarabinomannan (LAM) (ML34 [11, 12]) demonstrated the presence of
some LAM in purified antigen preparations but did not
detect any LAM associated with the 19-kDa band on SDSPAGE (data not shown).
T-cell recognition of the recombinant 19-kDa antigen. A
murine T-cell line specific for the M. tuberculosis 19-kDa
antigen was generated to investigate the immunological
activity of the recombinant antigen expressed in M. smegmatis. The antigenic specificity of this line was confirmed by
strong in vitro proliferative responses to rGST19 and to a
peptide (p19.7, residues 61 to 80) containing the major
murine T-cell epitope in the M. tuberculosis 19-kDa antigen
(9). The T-cell line also responded vigorously to the 19-kDa
protein expressed in M. tuberculosis, whereas M. smegmatis failed to induce a significant proliferative response (Fig.
5A). The lack of response to M. smegmatis is consistent with
serological evidence indicating that M. smegmatis does not
express a protein with antigenic cross-reactivity to the M.
tuberculosis 19-kDa antigen (3, 5, 10). After transformation
with pBAK-7q, however, M. smegmatis extracts induced a
strong response and were four to five times more potent in
T-cell proliferation assays than M. tuberculosis H37Rv extracts (Fig. 5B).
The recombinant 19-kDa antigen was also efficiently presented for T-cell recognition when added to the assay in the
form of intact bacteria. Proliferative responses were induced
by both live and killed M. smegmatis/pBAK-7q, while the
control M. smegmatis/pBAK14 failed to stimulate significant
responses (Fig. 6A). Inhibitory effects resulting from the
Downloaded from http://iai.asm.org/ on December 29, 2014 by guest
kDa
264
GARBE ET AL.
0.1
00
INFECT. IMMUN.
80
40
60
30
40
20
20
10
.1-*
-
0
0.
o
0
._
E
.-S
0
0
1
00
Antigen
(gg/mi)
FIG. 5. T-cell recognition of the M. tuberculosis 19-kDa antigen. (A) Antigenic specificity of the rGST19 T-cell line. T cells (2 x 104 per
well) were cultured with irradiated spleen cells (3 x 105 per well) for 3 days in the presence of rGST19, p19.7, M. tuberculosis H37Ra, or M.
smegmatis. [3H]thymidine incorporation was determined on day 3. (B) T-cell proliferative responses to the recombinant 19-kDa antigen
expressed in M. smegmatis. The rGST19 T-cell line (2 x 104 T cells per well) was cultured as described for panel A, in the presence of various
concentrations of soluble extracts from vector-transformed M. smegmatis (pBAK14), M. smegmatis expressing the 19-kDa antigen
(pBAK-7q), and M. tuberculosis H37Rv (H37Rv). Results are expressed as mean A counts per minute ± standard deviation of triplicate
determinations. (Counts per minute without antigens were 1,245 457 for panel A and 4,229 1,285 for panel B.)
±
addition of crude extracts of M. smegmatis to T-cell proliferation assays were found to be less severe than those
observed with comparable E. coli extracts. The response of
the T-cell line to the purified 19-kDa antigen was inhibited by
50% after the addition of 1 to 2 ,ug of protein from an E. coli
sonicate, for example, while approximately 30 ,ug of protein
from an equivalent M. smegmatis extract was required to
induce the same inhibitory effect (Fig. 6B).
Purification of the recombinant 19-kDa antigen. The recombinant 19-kDa antigen was further characterized by biochemical fractionation. Like the native antigen in M. tuberculosis, the recombinant 19-kDa protein was found mainly in
the pellet fraction generated by high-speed centrifugation of
bacterial extracts (Fig. 2). For protein purification, extracts
from M. smegmatislpBAK-7q were centrifuged overnight at
48,000 x g. The resulting pellet was resuspended in urea to
remove soluble proteins, and the recombinant antigen was
recovered by further centrifugation at 230,000 x g. After
washing with Triton X-100, the partially purified protein was
dissolved in SDS and 3-mercaptoethanol and further purified
by gel filtration in the presence of SDS-13-mercaptoethanol.
During the purification procedure, the antigen was monitored by SDS-PAGE, and Fig. 7 shows analysis of the final
gel filtration fractions. Western blot analysis identified the
19-kDa antigen in fractions 27 to 32 (Fig. 7B). Similarly,
±
ConA binding and T-cell reactivity were localized in precisely the same column fractions (Fig. 7C and D). Interestingly, although gel filtration was carried out under denaturing conditions, the 19-kDa antigen did not elute along with
other similar-sized proteins from the column (Fig. 7A),
suggesting that it may retain some unusual structural features even in the presence of SDS. The purification procedure yielded approximately 1 to 2 mg of protein from 3 liters
of M. smegmatis culture.
DISCUSSION
This study demonstrates the application of recombinant
DNA expression in a rapid-growing mycobacterium for
characterization of an antigen from M. tuberculosis. Although the 19-kDa antigen of M. tuberculosis can readily be
overexpressed as a fusion protein in E. coli (9), we have
previously been unable to achieve high-level expression of
the free protein in a range of E. coli expression systems
(1Sa). It is possible that this difficulty is related to the
unusual structural features of the 19-kDa antigen. We have
previously reported evidence indicating that the 19-kDa
antigen undergoes posttranslational modification with cleavage of its signal peptide and possibly addition of fatty acid(s)
to form a lipoprotein (31). Fifis et al. (8) have demonstrated
Downloaded from http://iai.asm.org/ on December 29, 2014 by guest
:5
RECOMBINANT ANTIGENS IN M. SMEGAL4TIS
VOL. 61, 1993
77 _
265
A
66
45
50
301717r- W M 0
V--
V--
r-
*1M.
*
c
N M I.P M W r- M M 0
r--
v
0 m m m
30
45
E0.
pBAK-7q-live
0
10 10
._
!N
0
1
10
-
17
-
...,
N
-r
0
-
0
0
s.
30
4.
45
100
Organisms (106 CFU/mI)
W 0
01Y
N C)
..
41*
04wi
t4 CLO
ip
CD N C
Q
0
C<D N CO C) 0
NN
N
r-
C
-
30 17 -
L-
1N Co a) 0
_ _ _ eN
B
60 -
N
cNes c
NJ
N
e
LO
NN
CD
9
co
a
NN N
0
I
tD N co an
C O U)
C') Cl CO) CO
CN C')
X
NCOX
0
0
3C
E
QL 2C
l
40 -
ic
N
CO 0
0
N C)
R
M C
N
CO ) 0
- N CO O
O
eD
Ns
0 0o
a:
Fraction
20 -
0 TG1
# pBAK1 4
1
10
100
1000
Antigen (,g/mi)
FIG. 6. Proliferative responses of rGST19 T-cell line after stimulation with intact organisms. (A) T cells (2 x 104 per well) were
cultured and proliferation was assessed as described in the legend to
Fig. 5 in the presence of live or heat-killed M. smegmatis transformed with vector alone (pBAK14, live; pBAK14, killed) and M.
smegmatis expressing the recombinant 19-kDa antigen (pBAK-7q,
live; pBAK-7q, killed). (B) Inhibition of proliferative responses of
the rGST19 T-cell line by E. coli and M. smegmatis soluble extracts.
Serial dilutions of soluble extracts from E. coli TG1 (TG1) and M.
smegmatislpBAK14 (pBAK14) were added to a fixed concentration
(10 p,g of protein per ml) of M. smegmatis/pBAK-7q soluble extract.
0, proliferation in the presence of M. smegmatis/pBAK-7q soluble
extract without inhibitors. Results are expressed as mean counts per
minute + standard deviation of triplicate determinations. (Counts
per minute without antigen were 458 + 267 for panel A and 1,447
864 for panel B.)
carbohydrate associated with the purified 19-kDa antigen of
M. bovis, and the ConA-staining pattern seen with the
recombinant M. tuberculosis antigen in M. smegmatis
strongly supports their inference that the 19-kDa antigen
exists as a glycoprotein. In contrast to eukaryotic cells,
FIG. 7. Purification of the recombinant 19-kDa antigen. The
19-kDa antigen was isolated from M. smegmatis transformed with
pBAK-7q as described in the text and fractionated by gel filtration in
the presence of SDS. Analysis of gel filtration fractions is shown.
Positions of molecular mass markers are given in kilodaltons on the
left side. (A) Fractions were analyzed by SDS-PAGE and stained
with Coomassie brilliant blue. F, 1 ,ug of horse spleen ferritin
(Sigma) as a quantitative marker. (B) Fractions separated by SDSPAGE were transferred to nitrocellulose and stained with monoclonal antibody HYT6. (C) Fractions separated by SDS-PAGE were
transferred to nitrocellulose and stained with peroxidase-conjugated
ConA. (D) Fractions were assessed for recognition by the rGST19
T-cell line by addition at a dilution of 1:10,000 to proliferation assays
as described in the legend to Fig. 5. Results are expressed as
[3H]thymidine incorporation.
protein glycosylation is not commonly found among bacteria. Glycoproteins have clearly been demonstrated in archaebacteria, but evidence in favor of glycoproteins in
eubacteria is less definitive (16). Final proof of the glycoprotein nature of mycobacterial antigens will require chemical demonstration of a covalent carbohydrate-peptide interaction, and the availability of the defined recombinant
system described here for the 19-kDa antigen will be particularly useful in pursuing such investigations.
We do not know whether posttranslational modification of
the 19-kDa antigen affects its immunological activity. Monoclonal antibodies recognize both the M. smegmatis and E.
coli recombinant antigens and are not apparently influenced
by acylation or glycosylation. Similarly, the E. coli recom-
Downloaded from http://iai.asm.org/ on December 29, 2014 by guest
pBAK14-live
o pBAK-7q-killed
0 pBAK14-killed
B
-
266
GARBE ET AL.
Mycobacterium leprae. J. Immunol. 147:2706-2712.
10. Ishioka, G. Y., A. G. Lamont, D. Thomson, A. Bulbow, F. C. A.
Gaeta, A. Sette, and H. M. Grey. 1992. MHC interaction and T
cell recognition of carbohydrate and glycopeptides. J. Immunol.
148:2446-2451.
11. Ivanyi, J., K. Sharp, P. Jackett, and G. Bothamley. 1988.
Immunological study of defined constituents of mycobacteria.
Springer Semin. Immunopathol. 10:279-300.
12. Ivanyi, J., S. Sinha, R. Aston, D. Cussell, M. Keen, and U.
Sengupta. 1983. Definition of species specific and cross-reactive
antigenic determinants of Mycobacterium leprae using monoclonal antibodies. Clin. Exp. Immunol. 52:528-536.
13. Jackett, P. S., G. H. Bothamley, H. V. Bathra, A. Mistry, D. B.
Young, and J. Ivanyi. 1988. Specificity of antibodies to immunodominant mycobacterial antigens in pulmonary tuberculosis. J.
Clin. Microbiol. 26:2313-2318.
14. Jacobs, W. R., M. Tuckman, and B. R. Bloom. 1987. Introduction of foreign DNA into mycobacteria using a shuttle phasmid.
Nature (London) 327:532-535.
15. Laemmli, U. K. 1970. Cleavage of structural proteins during the
assembly of the head of bacteriophage T4. Nature (London)
227:680-685.
15a.Lathigra, R. B. Unpublished data.
16. Lechner, J., and F. Wieland. 1989. Structure and biosynthesis of
prokaryotic glycoproteins. Annu. Rev. Biochem. 58:173-194.
17. Matsuo, K., R. Yamaguchi, A. Yamazaki, H. Tasaka, K.
Terasaka, and T. Yamada. 1990. Cloning and expression of the
18.
ACKNOWLEDGMENTS
We are grateful to Dan Tang for technical assistance, Carlos
Moreno and Christiane Abou-Zeid for helpful discussion, and Arend
Kolk and Ase Andersen for providing monoclonal antibodies F29-47
and HYT6.
REFERENCES
1. Aldovini, A., and R. A. Young. 1991. Humoral and cell-mediated
immune responses to live recombinant BCG-HIV vaccines.
Nature (London) 351:479-482.
2. Ashbridge, K. R., R. J. Booth, J. D. Watson, and R. B. Lathigra.
1989. Nucleotide sequence of the 19 kDa antigen gene from
Mycobacterium tuberculosis. Nucleic Acids Res. 17:1249.
3. Coates, A. R. M., J. Hewitt, B. W. Allen, J. Ivanyi, and D. A.
Mitchison. 1981. Antigenic diversity of Mycobacterium tuberculosis and Mycobacterium bovis detected by means of monoclonal antibodies. Lancet ii:167-169.
4. Deres, K., H. Schild, K.-H. Wiesmuller, G. Jung, and H.-J.
Rammensee. 1989. In vivo priming of virus-specific, cytotoxic T
lymphocytes with synthetic lipopeptide vaccine. Nature (London) 342:561-564.
5. Engers, H. D., and Workshop Participants. 1986. Results of a
World Health Organization-sponsored workshop to characterize antigens recognized by mycobacterium-specific monoclonal
antibodies. Infect. Immun. 51:718-720. (Letter to the editor.)
6. Espitia, C., and R. Mancilla. 1989. Identification, isolation and
partial characterization of Mycobacterium tuberculosis glycoprotein antigens. Clin. Exp. Immunol. 77:378-383.
7. Faith, A., C. Moreno, R. Lathigra, E. Roman, M. Fernandez, S.
Brett, D. M. Mitchell, J. Ivanyi, and A. D. M. Rees. 1991.
Analysis of human T-cell epitopes in the 19,000 MW antigen of
Mycobacterium tuberculosis: influence of HLA-DR. Immunology 74:1-7.
8. Fifis, T., C. Costopoulos, A. J. Radford, A. Bacic, and P. R.
Wood. 1991. Purification and characterization of major antigens
from a Mycobacterium bovis culture filtrate. Infect. Immun.
59:800-807.
9. Harris, D. P., H. M. Vordermeier, E. Roman, R. Lathigra, S. J.
Brett, C. Moreno, and J. Ivanyi. 1991. Murine T cell-stimulatory
peptides from the 19-kDa antigen of Mycobacterium tuberculosis: epitope restricted homology with the 28kDa protein of
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
gene for the cross-reactive ot antigen of Mycobacterium kansasii. Infect. Immun. 58:550-556.
Mehlert, A., and D. B. Young. 1989. Biochemical and antigenic
characterization of the Mycobacterium tuberculosis 71 kD antigen, a member of the 70 kD heat shock protein family. Mol.
Microbiol. 3:125-130.
Moreno, C., J. Taverne, A. Mehlert, C. A. W. Bate, R. Brealey,
A. Meager, G. A. W. Rook, and J. H. L. Playfair. 1989.
Lipoarabinomannan from Mycobacterium tuberculosis induces
the production of tumour necrosis factor from human and
murine macrophages. Clin. Exp. Immunol. 76:240-245.
Nair, J., D. A. Rouse, and S. L. Morris. 1992. Nucleotide
sequence analysis and serologic characterization of the Mycobacterium intracellulare homologue of the Mycobacterium tuberculosis 19kDa antigen. Mol. Microbiol. 6:1431-1439.
Ranes, M. G., J. Rauzier, M. Lagranderie, M. Gheorghiu, and
B. Gicquel. 1990. Functional analysis of pAL5000, a plasmid
from Mycobacterium fortuitum: construction of a 'mini' Mycobacterium-Escherichia coli shuttle vector. J. Bacteriol. 172:
2793-2797.
Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular
cloning: a laboratory manual. Cold Spring Harbor Laboratory
Press, Cold Spring Harbor, N.Y.
Snapper, S. B., L. Lugosi, A. Jekkel, R. E. Melton, T. Kieser,
B. R. Bloom, and W. R. Jacobs. 1988. Lysogeny and transformation in mycobacteria: stable expression of foreign genes.
Proc. Natl. Acad. Sci. USA 85:6987-6991.
Snapper, S. B., R. E. Melton, S. Mustafa, T. Kieser, and W. R.
Jacobs, Jr. 1990. Isolation and characterization of efficient
plasmid transformation mutants of Mycobacterium smegmatis.
Mol. Microbiol. 4:1911-1919.
Stover, C. K., V. F. de la Cruz, T. R. Fuerst, J. E. Burlein, L. A.
Benson, L. T. Bennett, G. P. Bansal, J. F. Young, M. H. Lee,
G. F. Hatfull, S. B. Snapper, R. G. Barletta, W. R. Jacobs, Jr.,
and B. R. Bloom. 1991. New use of BCG for recombinant
vaccine. Nature (London) 351:456-460.
Thangaraj, H., F. I. Lamb, E. 0. Davis, P. J. Jenner, L. H.
Jeyakumar, and M. J. Colston. 1990. Identification, sequencing,
and expression of Mycobacterium leprae superoxide dismutase,
a major antigen. Infect. Immun. 58:1937-1942.
Thole, J. E. R., W. J. Keulen, A. H. J. Kolk, D. G. Groothuis,
L. G. Berwald, R. H. Thiejema, and J. D. A. van Embden. 1987.
Characterization, sequence determination, and immunogenicity
of a 64-kilodalton protein of Mycobacterium bovis BCG expressed in Escherichia coli K-12. Infect. Immun. 55:1466-1475.
Towbin, H., T. Staehelin, and J. Gordon. 1979. Electrophoretic
Downloaded from http://iai.asm.org/ on December 29, 2014 by guest
binant fusion protein, and several synthetic peptides based
on the 19-kDa sequence, have been shown to induce proliferative T-cell responses (7, 9), indicating that posttranslational modification is not essential for expression of 19-kDa
antigenicity. It remains possible, however, that acylation
and glycosylation may have more subtle effects on the
immunogenicity of the 19-kDa antigen. Immunization with
lipopeptides provides an efficient mechanism for induction of
T-cell responses restricted by class I proteins from the major
histocompatibility complex (4), for example, and glycosylation may modify the ability of peptides to bind to particular
major histocompatibility complex molecules (10). It will be
of interest, therefore, to carry out further detailed immunological comparisons of the modified and unmodified forms of
the antigen.
Expression of a recombinant M. tuberculosis gene in a
mycobacterial host has the advantage of permitting analysis
of the defined antigen presented in the context of the range of
potentially competing antigens and in the presence of additional immunomodulatory components, such as LAM (19),
which would accompany exposure to the antigen during
natural infection. Our results demonstrate the feasibility of
carrying out such experiments with M. smegmatis and
suggest that mycobacterial extracts may be less toxic in this
regard than corresponding E. coli preparations.
INFECT. IMMUN.
VOL. 61, 1993
transfer of proteins from polyacrylamide gels to nitrocellulose
sheets: procedure and some applications. Proc. Natl. Acad. Sci.
USA 76:4350-4354.
29. Vordermeier, H. M., D. P. Harris, E. Roman, R. Lathigra, C.
Moreno, and J. Ivanyi. 1991. Identification of T cell stimulatory
peptides from the 38-kDa protein of Mycobacterium tuberculosis. J. Immunol. 147:1023-1029.
30. Young, D., T. Garbe, R. Lathigra, and C. Abou-Zeid. 1990.
Protein antigens: structure, function and regulation, p. 1-35. In
J. McFadden (ed.), Molecular biology of the mycobacteria.
RECOMBINANT ANTIGENS IN M. SMEGMATIS
267
Academic Press Ltd., London.
31. Young, D. B., and T. R. Garbe. 1991. Lipoprotein antigens of
Mycobacterium tuberculosis. Res. Microbiol. 142:55-65.
32. Young, D. B., S. H. E. Kaufmann, P. W. M. Hermans, and
J. E. R. Thole. 1992. Mycobacterial protein antigens: a compilation. Mol. Microbiol. 6:133-145.
33. Zhang, Y., R. Lathigra, T. Garbe, D. Catty, and D. Young. 1991.
Genetic analysis of superoxide dismutase, the 23 kilodalton
antigen of Mycobacterium tuberculosis. Mol. Microbiol. 5:381391.
Downloaded from http://iai.asm.org/ on December 29, 2014 by guest