Biomarker Changes Associated with Tuberculin Skin Test
(TST) Conversion: A Two-Year Longitudinal Follow-Up
Study in Exposed Household Contacts
Rabia Hussain1*, Najeeha Talat1, Firdaus Shahid1, Ghaffar Dawood2
1 Department of Pathology and Microbiology, The Aga Khan University, Karachi, Pakistan, 2 Masoomeen General Hospital, Kharadar, Karachi, Pakistan
Abstract
Background: A high prevalence (50–80%) of Tuberculin Skin Test Positivity (TST+ $10 mm indurations) has been reported
in TB endemic countries. This pool forms a huge reservoir for new incident TB cases. However, immune biomarkers
associated with TST conversion are largely unknown. The objective of this study was to identify immune biomarkers
associated with TST conversion after acute Mycobacterium tuberculosis (MTB) exposure.
Methodology/Principal Findings: A 24 month longitudinal study was carried out in a recently MTB exposed cohort of
household contacts (HC = 93; 75% TST+). Control group consisted of unexposed community controls (EC = 59; 46%TST+).
Cytokine secretion was assessed in whole blood cultures in response to either mycobacterial culture filtrate (CF) antigens or
mitogens (PHA or LPS) using Elisa methodology. Compared to the EC group, the HC group at recruitment (Kruskal-Wallis
Test) showed significantly suppressed IFN c (p = 0.0001), raised IL-10 (p = 0.0005) and raised TNF a (p = 0.001) in response to
CF irrespective of their TST status. Seventeen TST-HC, showed TST conversion when retested at 6 months. Post TST
conversion (paired t tests) significant increases were observed for CF induced IFN c (p = 0.038), IL-10 (p = 0.001) and IL-6
(p = 0.006). Cytokine responses were also compared in the exposed HC group with either recent infection [(TST converters
(N = 17)] or previous infection [TST+ HC (N = 54)] at 0, 6, 12 and 24 months using ANOVA on repeated measures. Significant
differences between the exposed HC groups were noted only at 6 months. CF induced IFN c was higher in previously
infected HC group (p = 0.038) while IL-10 was higher in recently infected HC group (p = 0.041). Mitogen induced cytokine
secretion showed similar differences for different group.
Conclusions/Significance: Our results suggest that TST conversion is associated with early increases in IFN c and IL-10
responses and precedes latency by several months post exposure.
Citation: Hussain R, Talat N, Shahid F, Dawood G (2009) Biomarker Changes Associated with Tuberculin Skin Test (TST) Conversion: A Two-Year Longitudinal
Follow-Up Study in Exposed Household Contacts. PLoS ONE 4(10): e7444. doi:10.1371/journal.pone.0007444
Editor: Philip Campbell Hill, MRC Laboratories, Gambia
Received June 10, 2009; Accepted September 16, 2009; Published October 14, 2009
Copyright: ß 2009 Hussain et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: Funding was provided by National Commission on Biotechnology (PCST/NCB-AC3/2003) & Higher Education Commission (HEC# 20-796/R&D/07)
Government of Pakistan. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
* E-mail: rabia.hussain@aku.edu
in the M. tuberculosis complex [10]. Commercial IFNc release
assays (IGRA) using ESAT-6 stimulated whole blood cultures are
increasingly available for detection of recent infection [11] or
disease [12]. However, longitudinal studies suggest IGRA test
cannot replace TST in high TB incidence areas for detection of
latent infection [13]. The high incidence of HIV (,11%) in 8.8
million new TB cases [14] further confounds the immune
correlates associated with latency as immune suppression results
in lower TST diameters in HIV+ individuals with latent TB
infection. There is an urgent need to carry out such studies in TB
endemic areas with a low HIV prevalence to understand
unmitigated immune correlates associated with TST conversion.
Immune activation after a recent exposure may be associated with
activation and modulation of several immune markers and the
magnitude and timing of pro- and down-regulatory cytokines may
be very important for TST conversion during a natural exposure.
IFN c is a known correlate of delayed type hypersensitivity as well
as TST skin test diameter [15]. Early activation of IFN c is
therefore linked to TST conversion and development of effective
Introduction
Pakistan ranks 8th in terms of global tuberculosis (TB) disease
burden [1]. One third of the world population is considered to be
latently infected as assessed by Tuberculin Skin Test positivity
(TST+ $10 mm of indurations) [2] resulting in a huge reservoir
for new incident cases. In a high transmission setting such as
Pakistan the rate of TST positivity can be as high as 50% in the
community[3] and household contacts [4] increasing up to 80% in
recently exposed household contacts [5,6]. PPD administered in
TST is a mixture of proteins prepared from M. tuberculosis (MTB)
culture filtrate and is highly cross reactive with other mycobacterial antigens including Mycobacterium bovis. Pakistan has wide
BCG vaccination coverage at birth (.70%) through the expanded
program for immunization (EPI) [7] and the high prevalence of
TST+ in the community may be due to either previous BCG
vaccination [8] or booster effects with environmental mycobacteria [9]. ESAT-6, induced interferon (IFN c) is considered to be a
MTB specific marker for infection and disease as it is present only
PLoS ONE | www.plosone.org
1
October 2009 | Volume 4 | Issue 10 | e7444
Biomarkers Associated with TST
immunity. TNFa in conjunction with IFN c has been shown to
play a role in granuloma formation [16] and in maintenance of
granuloma [17]. IL-10 is a down-regulatory cytokine and plays a
key role in limiting pathology by inhibiting overproduction of proinflammatory cytokines [18,19] particularly in the chronic phase
of infection [20,21]. IL-6 is a pro-inflammatory cytokine secreted
by both macrophages and T cells and plays a critical role in
driving the differentiation of B cells [22]. In the mouse gene
knockout model IL-6 has also been shown to participate in the
early induction of IFN c production [23]. We have therefore opted
to analyze these biomarkers post acute exposure due to their
identified roles in control of mycobacterial infections. Due to slow
replication of mycobacteria, immune responses are usually slow to
evolve requiring serial testing over several months post infection.
Longitudinal studies for immune correlates of tuberculosis
infection and disease have been carried out in African countries
with relatively high HIV prevalence [24–27]. Pakistan still has
a relatively low HIV (0.8/100000) prevalence [1] compared to
1–3% in eastern Africa and 10.8% in South Africa [28] and
therefore provides a useful setting for such studies. The objective of
the current study was to evaluate multiple interrelated biomarkers
serially in a MTB exposed cohort to identify biomarkers associated
with TST conversion and to analyze differences in the evolution of
biomarkers in household contacts with either recent or past
infection post acute exposure.
previously [31]. Briefly MTB strain H37 Rv was grown in
Prouskeur Beck medium. After 8–10 weeks of culture, bacilli were
removed by sedimentation and filtration. The preparation
contained minimal amounts of endotoxin as determined by
inhibition with polymyxin B [32]. Pairs of mouse monoclonal
antibodies specific for each cytokine were purchased from
Pharmingen (San Diego, CA) for the assessment of cytokines. All
reagents were used according to manufacturer’s instructions.
Whole blood stimulation assay
Whole blood stimulation assay has been described in detail
previously [33]. Briefly, 5 ml of blood collected by venipuncture from
each donor was mixed with sodium heparin (20 U/ml; Leo
pharmaceutical, Ballerup, Denmark) in 15 ml plastic centrifuge
tubes (BD Falcon). Blood was further diluted 1:10 with RPMI 1640
tissue culture medium containing 100 U/ml of penicillin+100 mg/ml
of streptomycin (Sigma Chemicals St. Louis, MO. USA) and 2 mM
of L-glutamine(Sigma Chemicals). Diluted blood (900 ml/well) was
dispensed in 24-well tissue culture plates (Flow laboratories, Irvine,
CA) within two hours of collection. The blood was subsequently
stimulated with 100 ml of PHA [5.0 mg/ml] or LPS [1.0 mg/ml] or
CF [5.0 mg/ml] in tissue culture plates and further incubated at 37uC
in 5% CO2. Blood cultures were stimulated in separate wells for
either 2 days (TNF a, IL-10 and IL-6) or 5 days (IFN c). Supernatants
of whole blood cultures were collected and stored as 46200 ml of
aliquots at 235uC until use. Selection of mitogen/antigen concentrations and optimal days for collection of supernatants for various
cytokines has been reported previously [33].
Materials and Methods
The study cohort comprised of healthy household contacts
(HC = 93) from 20 families. Each family had at least one recently
diagnosed, untreated pulmonary tuberculosis patient. The intensity and duration of exposure as assessed by history of symptoms
and the extent of disease in the index case was similar in all
families (Table S1). Inclusion criteria: household contacts who had
no previous history of anti-tuberculous treatment (ATT) and who
remained disease-free during a 24 month follow up period [5,29].
Exclusion criteria: Children under the age of five or subjects with
underlying diseases which could result in immune suppression or
individuals on steroids. Endemic controls (EC, N = 59) were
employees working in various capacities at The Aga Khan
University with no history of recent exposure to tuberculosis.
Lady Health Visitors (LHV) made home visits, obtained informed
written consent from all adult participants or guardians in case of
children ,18 years. There is no routine testing for HIV in
Pakistan because of the low incidence [1].
Cytokine assessment
Tuberculin Skin Tests (TST) were carried out by injecting 5
tuberculin units intra-cutaneously in the volar surface of the arm.
A single tester administered and read the indurations after
48 hours using a caliper. A diameter$10 mm was considered
positive [30]. At intake TST was carried out on all HC (N = 93)
and EC (N = 59). Repeat testing was carried out in TST-HC
(N = 23). No further TST testing was done on the TST+HC group
and no prophylactic treatment was provided for latent tuberculosis
infection.
The Elisa assay was optimized in-house and details of the assay
have been reported previously [33]. Briefly Immulon 4 plates were
coated with capture antibodies and subsequently incubated with
100 ml of un-stimulated or stimulated culture supernatants overnight
at 4uC. Subsequently the probing and revealing antibodies were
added (100 ml) stepwise for appropriate incubations. The plates were
washed between incubations. All probing antibodies were biotin
labeled and the revealing antibodies were labeled with avidin bound
to horse- radish peroxidase (HRP) (Sigma St Louis MO). HRP
substrate was added and color development was stopped with 1N
NaOH after 30 minutes incubation. The optical densities were read
using an Elisa reader (Biorad microplate reader 680). Mitogen
stimulated cytokines were assessed at a dilution of 1:10 and 1:100 and
antigen stimulated cytokines were assessed at neat and a dilution of
1:10 of the supernatants. Optical density values falling in the
midrange of the standard curve were used to calculate the
concentration of cytokine in the supernatants and final values were
obtained after multiplying by the dilution at which the cytokines were
assessed. In case of PHA stimulated cultures occasionally (IFN c in 3
donors) a dilution of 1:1000 was needed to calculate the final value.
Results were expressed as d pg/ml, after deducting secretion in the
absence of stimulus to normalize the expression of mitogen/antigen
specific secretion. The sensitivity and range of cytokine detection was
as follows: TNF a (7.8–1000 pg/ml); IL-6 (7.8–1000 pg/ml); IL-10
(15–250 pg/ml); Interferon c (50–2000 pg/ml).
Reagents
Statistical analysis
Purified protein derivative from Mycobacterium tuberculosis (MTB)
was obtained from Staten Serum Institute (batch RT47;
Copenhagen, Denmark). LPS and PHA were purchased from
Sigma (St. Louis MO, USA). MTB Culture Filtrate (CF) proteins
were prepared in-house at Case Western Reserve University,
Cleveland, Ohio and provided by Dr Robert Wallis as described
SPSS soft ware (version 16.0) was used for the statistical
analyses. Kruskal-Wallis analysis was carried to compare overlapping test groups with control group. Paired student t tests were
applied to compare two related samples. Analysis of variance
(ANOVA) on repeated measures was carried out using General
Linear Model (GLM) for comparison of group cytokine profiles.
Tuberculin Skin Tests
PLoS ONE | www.plosone.org
2
October 2009 | Volume 4 | Issue 10 | e7444
Biomarkers Associated with TST
observed in the HC group compared to the EC groups (Table 2).
No difference in CF induced IL-6 secretion was observed in the two
groups. Thus the hallmarks of recent infection may be suppressed T
cells and unregulated macrophage response.
Ethics statement
The study protocol received approval of the Ethical Review
Committee of The Aga Khan University and Hospital.
Results
Comparison of mycobacterial antigen (CF) and mitogen
(PHA or LPS) stimulated cytokine levels pre- and post TST
conversion in exposed disease free household contacts
(HC)
Study subjects
Table 1 shows the characteristics of the study groups at
recruitment (0 months). The EC group was healthy and had no comorbid conditions that could compromise the immune response.
The intensity and duration of Mycobacterium tuberculosis (MTB)
exposure was similar in the household contacts (HC) (Table S1).
Although it is difficult to exclude TB exposure in high TB burden
countries, none of the community controls (EC) gave a history of
recent exposure to a tuberculosis patient and were considered
unexposed. TST diameter in the TST+HC and the TST+EC
groups was comparable (Table 1). The TST positivity in the EC
group in the absence of MTB exposure is therefore most likely due
to either previous BCG vaccination or exposure to environmental
mycobacteria. The TST positivity in the HC group at recruitment
is more likely due to previous infection as these individuals were
from a high transmission pocket [6]. We have therefore,
considered TST+HC group as having latent infection at the time
of recruitment for the current study. The TST-HC group showed
slightly higher TST diameter (mean indurations 3.563.0)
compared to the TST-EC group (mean indurations 0.862.3).
The absence of TST positivity in the HC group (23/93) was not
related to BCG vaccination as there was no difference in the
presence of BCG scar in TST-HC (40%) compared to TST+HC
(38.5%) (Table S1). The slightly higher TST diameter in the
TST-HC is probably due recent exposure. The TST- HC group
was therefore considered a recently infected group.
Repeat skin testing was carried out in 17/23 TST-HC donors at
6 months. TST conversion was noted in all donors (N = 17; TST
diameters pre-conversion, mean = 3.1863.55 SD; post conversion,
mean = 1160.6 SD mm). The remaining donors (6/23) could only
be retested at 24 months and were excluded from these analyses as
the time of TST conversion could not be ascertained. To assess
biomarker changes associated with TST conversion, we analyzed
mycobacterial antigens or mitogen induced cytokine secretion preand post TST conversion.
a.
b.
Cytokine secretion in recently exposed HC and non
exposed EC in mycobacterial antigen (CF) stimulated
whole blood cultures
To assess early changes in biomarkers associated with recent
exposure to MTB, we first compared CF induced cytokine
responses in the exposed HC group (TST-HC = 23; TST+HC = 70)
with the unexposed EC (TST-EC = 32; TST+EC = 27) group
(Figure 1 and Table 2). Kruskal-Wallis ANOVA tests were applied
to determine the significance of differences in cytokines in the
exposed HC groups compared to the non exposed EC groups. The
distribution of CF induced cytokines is shown as a box plot (Figure 1)
with the median and inter-quartile ranges (IQR) shown in Table 2.
CF induced IFN c was significantly suppressed (p = 0.0001) in the
HC group irrespective of the TST status. Increased CF induced
TNF a (p = 0.001) and CF induced IL-10 (p = 0.0005) were
Mycobacterial antigen induced cytokine secretion in
HC pre- and post TST conversion. Figure 2 (Panel A)
shows a comparison of CF induced cytokine responses (log n) in
TST-HC group pre- and post TST conversion (N = 17).
Significant increases (paired t tests) in median values were
observed for CF induced IFN c (pre- 4.3060.51; post
5.1960.55; p = 0.038); CF induced IL-10 (pre- 5.2560.20;
post 6.0760.14; p = 0.001) and CF induced IL-6 (pre7.4060.17; post 7.7760.15; p = 0.006). The only cytokine that
was not significantly different post TST conversion was CF
induced TNF a (pre- 6.1160.19; post 5.9560.14; p = 0.43).
Mitogen induced cytokine secretion in HC pre- and
post TST conversion. We also analyzed changes in cytokine
secretion in response to mitogenic stimuli pre- and post TST
conversion which would reflect non-specific cellular activation.
While it is difficult to define the cellular source of cytokine in
whole blood cultures we have used PHA as a pan T cell activator
and LPS which is potent activator of monocytes/macrophages.
Mitogen stimulated cytokine pre- and post TST conversion
(Figure 2, panel B) also showed significant differences with the
same three cytokine [PHA induced IFN c (pre- 9.0960.18; post
9.6360.28; p = 0.028); LPS induced IL-10 (pre- 5.6360.20; post
6.7060.13; p = 0.0004) and LPS induced IL-6 (pre- 7.5060.17;
post 8.046.11; p = 0.002)]. LPS induced TNF a again showed
comparable responses for post TST conversion (pre- 6.5060.25;
post 6.5160.15; p = 0.96). As expected the overall levels of PHA
induced IFN c secretion was 15–18 fold higher compared to
mycobacterial antigen induced IFN c. On the other hand, LPS
induced TNFa (a prototypic marker of macrophage activation)
Table 1. Characteristics of the study group at intake.
Group ID*
n
Age (X61SD)
Male/Female
Diameter of indurations (X61SD)
TST+HC
70
25.8615.9
34/36
17.666.6
TST-HC
23
20.0610.7
11/12
3.563.2
TST+EC
27
31.369.4**
10/17
16.664.1
TST-EC
32
24.065.0
18/14
0.862.3
*HC = Healthy household contacts. EC = Community controls.
TST+ = Positive Tuberculin Skin Test diameter $10 mm.
TST- = Negative Tuberculin Skin Test diameter ,10 mm.
**One way ANOVA, p = .008.
doi:10.1371/journal.pone.0007444.t001
PLoS ONE | www.plosone.org
3
October 2009 | Volume 4 | Issue 10 | e7444
Biomarkers Associated with TST
Figure 1. Comparison of CF induced cytokine responses in MTB exposed HC and unexposed EC. Cytokine secretion was assessed in
supernatants of stimulated whole blood cultures. CF [5 mg/ml] induced IFNc secretion was assessed at day 5 and CF induced TNF a, IL-10 and IL-6
secretion was assessed at day 2 post- stimulation. Results are expressed after deducting background secretion in the absence of stimulation. Results
are given as box plots representing 25th, 50th and 75th percentiles (logn pg/ml). The number of donors in each group is given in brackets. Significant
differences between the HC and EC groups for different cytokines (ANOVA; Kruskal-Wallis test) are given as p values in Figure 1.
doi:10.1371/journal.pone.0007444.g001
was only 2–3 fold higher compared to CF induced TNFa. These
results are in line with previously published reports that
mycobacterial antigens are potent stimulators of macrophages
[31,34]. The overall cellular activation was parallel to mycobacterial antigen stimulated cytokine secretion.
group with previous infection (TST+HC = 54) analysis of variance
(ANOVA) was carried on repeated measures over a two year
period. The reference group (TST+HC) had similar MTB
exposure and demographics at recruitment including the presence
of BCG scar (Table S1). Only donors sampled at all time points
(54/70) were included in the analysis. Figure 3 (IFN c and IL-10)
and Figure 4 (TNF a and IL-6) show comparative profiles for
mycobacterial antigen (CF) or mitogen (PHA or LPS) induced
cytokine secretion.
Both groups showed dynamic changes in cytokines after acute
exposure. Not surprisingly, CF induced IFN c (Figure 3, panel A)
showed significantly lower responses in recently infected compared
Longitudinal assessment of CF and mitogen induced
cytokine secretion post exposure in recently infected and
previously infected HC groups over a 2-year period
To identify the cytokine dynamics post exposure in the HC
group with recent infection (TST-HC = 17) compared to the HC
Table 2. Cytokine secretion in recently exposed HC and non exposed EC in mycobacterial antigen induced whole blood cultures.
Groups
TST- HC
TST-EC
TST+HC
TST+EC
p=
Biomarker (pg/ml)
median (IQR)
median (IQR)
median (IQR)
median (IQR)
IFNc
397(1009)
283 (758)
873(2724)
1720(2619)
0.0001
TNFa
544(452)
541 (574)
216(420)
441(750)
0.001
IL-10
212(351)
263(472)
141(212)
195(258)
0.0005
IL-6
2078(2913)
2232(3932)
2918(3697)
3429(3351)
.0.1
Note: Values in bold signify significant up regulation and values underlined indicate significant suppression Kruskal Wallis analysis were applied for statistical
significance.
doi:10.1371/journal.pone.0007444.t002
PLoS ONE | www.plosone.org
4
October 2009 | Volume 4 | Issue 10 | e7444
Biomarkers Associated with TST
Figure 2. Comparison of CF and mitogens induced cytokine secretion pre- and post TST conversion. Cytokine secretion was assessed in
supernatants of stimulated whole blood cultures. IFNc secretion was assessed at day 5 and TNF a, IL-10 and IL-6 secretion at day 2 post stimulation in
whole blood culture supernatants. Panel A. CF [5 mg/ml] stimulated cytokine secretion. Panel B. PHA [5 mg/ml] stimulated IFNc and LPS [1 mg/ml]
stimulated TNF a, IL-10 and IL-6 secretion. Results are given after deducting background secretion in the absence of stimulation and are depicted as
box plots representing 25th, 50th and 75th percentiles (log n pg/ml) on the Y axis. Paired t test were applied to determine the significance of
differences between pre- and post TST conversion (N = 17) and p values for each cytokine is given.
doi:10.1371/journal.pone.0007444.g002
to previously infected HC group at 6 months [TST-HC; lower
IFN c (p = 0.038)]. In response to PHA (Figure 3, panel C), the
magnitude of IFN c response did not differ between the two HC
groups, but the timing of peak responses were different. The
previously infected HC group showed peak IFN c response at 6
months, while the recently infected HC group peaked at 12
months (Figure 3, panel C) suggesting that TST conversion may
be related to the strength of mycobacterial specific T cell response
rather than the overall strength of T cell activation. The trend with
CF induced IL-10 (Figure 3, panel B) in the two groups was
opposite to that observed with CF induced IFN c with IL-10
responses being significantly higher in recently infected HC at 6
months (p = 0.041). Parallel trends were observed with LPS
PLoS ONE | www.plosone.org
induced IL-10 (Figure 3, panel D). Both LPS and CF induced
IL-10 showed a much more dynamic profile with greater
variability in recently infected HC group (Table S2) compared
to a relatively flat response over time in the previously infected HC
group indicating that immune responses in newly infected HC take
at least a year to stabilize.
No significant differences were noted with TNFa (Figure 4,
panel A and C) or IL-6 (Figure 4, panel B and D) in response to
CF or mitogens between recently infected compared to previously
infected HC (Figure 4) The only exception was LPS induced IL-6
(Figure 4D) which was significantly higher (p = 0.039) in the
previously infected HC group at 0 months compared to the HC
group with recent infection at recruitment.
5
October 2009 | Volume 4 | Issue 10 | e7444
Biomarkers Associated with TST
Figure 3. CF and mitogen induced IFN c and IL-10 in TST+HC and TST-HC groups. Cytokine secretion was assessed in supernatants of
stimulated whole blood cultures for all donors at 0, 6, 12 and 24 months. Previously infected (TST+HC = 54) HC are depicted as closed circles and solid
lines; recently infected (TST-HC = 17) as open circles and broken lines. Concentrations of CF and mitogens, and days of stimulation are same as in
Figure 2. Panel A and B show CF induced and panel C and D show mitogen induced cytokines. Months at which cytokines were assessed post
exposure are shown on the X axis. The cytokine values are log n transformed and means are shown on the Y axis. ANOVA test for repeated measures
using GLM was carried to assess the differences in the two groups. Vertical line indicates the error bars (95% confidence intervals) at each time point.
P values are shown for each time point. A p value,0.05 was considered significant.
doi:10.1371/journal.pone.0007444.g003
stimulated whole blood have been recommended for early
diagnosis of latent infection especially in a resource strapped
countries [25]. When cytokines responses were compared in the
exposed and the non exposed groups, elevated responses were
observed with mycobacterial antigen induced TNFa and IL-10
which were most likely derived from innate cells. This was not
surprising as mycobacterial proteins are known to be potent
stimulators of macrophages [24, 33, 35, and 36]. However,
suppressed mycobacterial antigen induced IFN c responses in the
exposed HC group compared to the EC group was surprising.
Pakistan has wide BCG vaccination coverage at birth (.70%)
through the expanded program for immunization (EPI) [7] and a
high prevalence of TST+ in the non exposed community control
group may be due to either previous BCG vaccination [8] or
booster effects with environmental mycobacteria [9]. As to why we
did not observe such a booster effect in the recently exposed cohort
may lie with CF induced IL-10 which was selectively raised in the
exposed HC groups but not in the non exposed EC group. IL-10 is
a down-regulatory cytokine and inhibits pro-inflammatory cytokines such as IFN c [18,19] and may be the reason for the
observed lower CF induced IFN c responses in the exposed HC
group. The HC groups had been exposed to MTB for at least 3
Discussion
This is the first report on serial assessment of multiple
biomarkers to identify changes associated with TST conversion
in a recently MTB exposed cohort in a low HIV setting. Natural
evolution of biomarkers post exposure was possible in those with
recent and past infection as no prophylaxis was given for latent
tuberculosis. The cohort remained disease free over a 4 year
period suggesting that these biomarker profiles may represent
protective profiles in the HC group post exposure. The strength of
the study was therefore, careful selection of a cohort with no
history of TB or of anti-tuberculous treatment in the past.
Variability in the duration and intensity of exposure can change
the dynamics of immune parameters particularly with respect to
innate responses that are activated early during infection. Both
these parameters are very difficult to control in a high transmission
setting. We have tried to limit this variability by including families
where the symptoms in the index cases initiated no more than 3
months and the pulmonary disease was restricted to moderate lung
disease. We have used MTB culture filtrate antigens which are
comparable in activity to PPD antigens used in TST. In household
contacts IFN c secretion in response to culture filtrate antigens in
PLoS ONE | www.plosone.org
6
October 2009 | Volume 4 | Issue 10 | e7444
Biomarkers Associated with TST
Figure 4. CF and mitogen induced TNF a and IL-6 in TST+HC and TST-HC groups. Previously infected (TST+HC = 54) HC are shown as closed
circles and solid lines and recently infected (TST-HC = 17) by open circles and broken lines. Concentrations of CF and mitogens, and days of
stimulation are same as in Figure 2. All other parameters are same as in Figure 3.
doi:10.1371/journal.pone.0007444.g004
marker of classically activated macrophages (CAM Q) while
alternatively activated macrophages (AAM Q) secrete high levels of
IL-10. [39]. Mycobacterial antigens can signal macrophages via
the Toll like receptors [40] and recently TLR9 has been
implicated in the activation of alternatively activated macrophages
(AAM Q) [41]. It is therefore tempting to speculate that post TST
conversion there is concurrent elevation of Th1 and AAM Q which
down regulate CAM Q. Increase in IL-6 post TST conversion was
also intriguing. Although the role of IL-6 in human tuberculosis is
not clear, IL-6 has been shown to be elevated in established disease
[33]. In the experimental IL-6 gene knockout mouse, a delay in
early induction of IFN c [23] as well a delay in the initial
differentiation of Th1 cells, but not in expansion, has been
reported [42].
To further understand the evolution of cytokines post MTB
exposure, we compared the HC group with recent infection (TST
converters) or past infection (TST+HC at recruitment). Again
significant differences post exposure in the two groups, were noted
only at 6 months. These results suggest that TST conversion may
precede latency and stable granuloma formation as cytokine
responses in the two groups (recent infection Vs previous infection)
become comparable beyond 12 months. The cytokines likely to
play a role in TST conversion such as IFN c show a plateau
beyond 12 months while IL-10 continues to increase in both
groups beyond 12 months and may play a greater role in
establishing dormancy and latency. These findings therefore
months, which may be sufficient to activate the innate arm of the
immune responses. CF induced IL-6 was the only cytokine which
did not show differences between the HC and the EC groups. The
HC groups stratified on the basis of TST status did not show
significant differences (Mann Whitney U tests) in CF induced
cytokine responses (Figure 3 &4). The differences between the non
exposed EC and the exposed HC groups are therefore most likely
due to recent MTB exposure.
In relation to TST conversion, the HC group did show an
increase in CF induced IFN c secretion post TST conversion.
Similar increases in IFN c to mycobacterial antigens have been
reported post exposure [37] and post TST conversion to ESAT-6
which is a MTB specific antigen [27]. Antigen induced IFN c has
been reported to correlate with TST diameter [38], but the
coordinate expression of CF induced IL-10 was surprising. A
similar LPS induced IL-10 secretion was noted, suggesting an
overall activation of macrophages post TST conversion. IL-10
may be elevated to reduce collateral tissue damage [19] in
response to a heightened proinflammatory response during the
early phase of infection. The time frame of TST conversion (6
months) and establishment of stable cytokine responses is also the
time frame where the highest reported incidence of disease post
MTB exposure was observed in this cohort [5]. Disease
progression in this cohort (N = 7) was associated with depressed
IFN c/IL-10 ratio [5]. Surprisingly there was no concurrent
elevation of TNF a post TST conversion. TNF a is prototypic
PLoS ONE | www.plosone.org
7
October 2009 | Volume 4 | Issue 10 | e7444
Biomarkers Associated with TST
supports the notion that latent tuberculosis actually represents a
subclinical spectrum which can only be defined by the host
immune response [43] and is supported by the variability of
responses in recently infected groups while previously infected
groups represent a more stable host response.
Our studies therefore suggest that while IFN c may play an
important role in resolution of infection and TST conversion,
IL-10 may be critical in dampening the pro inflammatory arm of
the immune response as well as in initiation of dormancy during
the later phase of infection. The current study therefore, raises
important questions with regards to the source of various
biomarkers and their role in establishing TST conversion,
dormancy and latency.
scar positive HC, 71% were TST positive, and in BCG scar negative
HC 78.5% were TST positive. Reference: Crofton J (1990) Clinical
features of tuberculosis. In: Seton D, Gordon A, editors. Crofton
and Douglas Respiratory Diseases. London: Blackwell Scientific. pp.
395–421.
Found at: doi:10.1371/journal.pone.0007444.s001 (0.05 MB
DOC)
Table S2 Repeated measure on TST+ HC and TST- HC.
Intergroup comparison of cytokine profiles in the HC group with
previous infection (TST+HC at intake) with recent infection (TSTHC at intake) in response to mycobacterial antigens (CF) and
mitogens (PHA or LPS). This is a companion table for Figures 3
and 4. ANOVA test for repeated measures using GLM was carried
to assess the differences in the two groups. The cytokine values are
log n transformed and the estimated marginal means, SEM, b
estimates, p value and 95% CI for each of the stimulant and
cytokines shown in Figure 3 and Figure 4. A p value,0.05 was
considered significant.
Found at: doi:10.1371/journal.pone.0007444.s002 (0.12 MB
DOC)
Supporting Information
Table S1 Intensity and duration of exposure to M.tuberculosis in
Household Contacts at Recruitment. a Cough and or low grade fever
and or weight loss. b Intensity of Acid Fast Bacilli in sputum smear. c
Determined by radiology (According to Crofton et al 1990[30] d
ATT (days) started prior recruitment. TST+HC were not given
prophylactic ATT. e # HC in the family. # TST-HC in the HC
within each family is shown in brackets. * repeat TST available at 24
months only. Cytokine levels were available at all time points (0, 6, 12
and 24 months) on 77 HC (TST+ = 54; TST- = 23). Secondary cases
(N = 8) diagnosed over 4 years follow up, and contacts previously
treated (N = 5) were excluded. 1 One co-prevalent case (on ATT) in
the family BCG scar was present in 40% (38/94) HC. Among BCG
Author Contributions
Conceived and designed the experiments: RH. Performed the experiments:
NT FS. Analyzed the data: RH NT. Contributed reagents/materials/
analysis tools: RH. Wrote the paper: RH. Statistical analysis and data
presentation: NT. Data entry and logistical support: FS. Patient
recruitment, diagnosis and clinical follow-up: GD.
References
1. World Health Organization (2008) WHO Report 2008 - Country Profile
Pakistan. Global Tuberculosis Control. pp 133–136.
2. Dye C, Scheele S, Dolin P, Nathanial V, Raviglione MC (1999) Global Burden
of Tuberculosis: Estimated incidence, Prevalence, and Mortality by Country.
JAMA 282: 677–686.
3. Hussain H, Akhtar S, Nanan D (2003) Prevalence of and risk factors associated
with Mycobacterium tuberculosis infection in prisoners, North West Frontier
Province, Pakistan. Int J Epidemiol 32: 794–799.
4. Rathi SK, Akhtar S, Rahbar MH, Azam SI (2002) Prevalence and risk factors
associated with tuberculin skin test positivity among household contacts of
smear-positive pulmonary tuberculosis cases in Umerkot, Pakistan. Int J Tuberc
Lung Dis 6: 851–857.
5. Hussain R, Talat N, Shahid F, Dawood G (2007) Longitudinal tracking of
cytokines after acute exposure to Tuberculosis: Association of Distinct Cytokine
patterns with protection and disease development. Clin Vaccine Immunol 14:
1578–1586.
6. Akhtar S, White F, Hasan R, Rozi S, Younus M, et al. (2007) Hyperendemic
pulmonary tuberculosis in peri-urban areas of Karachi, Pakistan. BMC Public
Health 7: 70.
7. World Health Organization (2009) Review of National Immunization coverage
1980-2008 (Pakistan). .
8. Diel R, Nienhaus A, Lange C, Meywald-Walter K, Forssbohm M, et al. (2006)
Tuberculosis contact investigation with a new, specific blood test in a lowincidence population containing a high proportion of BCG-vaccinated persons.
Respir Res 7: 77.
9. Menzies D (1999) Interpretation of Repeated Tuberculin Tests. Am J Respir
Crit Care Med 159: 15–21.
10. Sorensen AL, Nagai S, Houen G, Andersen P, Andersen AB (1995) Purification
and characterization of a low-molecular mass ‘T’ cell antigen secreted by
Mycobacterium tuberculosis. Infect Immun 63: 1710–1717.
11. Streeton JA, Desem N, Jones SL (1998) Sensitivity and specificity of a gamma
interferon blood test for tuberculosis infection. Int J Tuberc Lung Dis 2:
443–450.
12. Britton W, Gilbert GL, Wheatley J, Leslie D, Rothel JS, et al. (2005) Sensitivity
of human gamma interferon assay and tuberculin skin testing for detecting
infection with Mycobacterium tuberculosis in patients with culture positive
tuberculosis. Tuberculosis 85: 137–145.
13. Hill PC, Jackson-Sillah DJ, Fox a, Brookes RH, de Jong BC, et al. (2008)
Incidence of tuberculosis and the predictive value of ELISPOT and Monteux
tests in Gambian case contacts. PLoS ONE 3: e1379.
14. World Health Organization (2007) WHO: Global Tuberculosis Control:
Surveillance, Planning, Financing.
15. Black GF, Fine PEM, Warndorff DK, Floyd S, Weir RE, et al. (2001)
Relationship between IFN gamma and skin test responsiveness to Mycobacte-
PLoS ONE | www.plosone.org
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
8
rium tuberculosis PPD in healthy, non BCG vaccinated young adults in
Northern Malawi. Int J Tuberc Lung Dis 5: 1–9.
Kindler V, Sappino AP, Grau GE, Piguet PF, Via LE, et al. (1989) The inducing
role of tumor necrosis factor in the development of bactericidal granulomas
during BCG infection. Cell 56: 731–740.
Keane J (2005) TNF-blocking agent and tuberculosis: new drugs illuminate an
old topic. Rheumatology 44: 714–720.
Fiorentino DF, Zlotnik A, Mosmann TR, Howard M, O’Garra A (1991) IL-10
inhibits cytokine production by activated macrophages. J Immunol 147:
3815–3822.
O’Garra A, Vieira PL, Vieira P, Goldfeld AE (2004) IL-10 producing and
naturally occurring CD4+ Tregs: limiting collateral damage. J Clin Invest 114:
1372–1378.
Abebe F, Mustafa T, Nerland AH, Bjune GA (2005) Cytokine profile during
latent and slowly progressive primary tuberculosis: a possible role for interleukin15 in mediating clinical disease. Clin Exp Immunol 143: 180–192.
Higgins DM, Sanchez-Campillo J, Rosas-Taraco AG, Lee EJ, Orme IM, et al.
(2009) Lack of IL-10 alters inflammatory and immune responses during
pulmonary Mycobacterium tuberculosis infection. Tuberculosis (Edinb) 89:
149–157.
Lue C, Kiyono H, McGhee JR, Fujihashi K, Kishimoti T, et al. (1991)
Recombinant human interleukin 6 (rhIL-6) promotes the terminal differentiation
of in vivo activated human B cells into antibody secreting cells. Cellular
Immunology 132: 423–432.
Saunders BM, Frank AA, Orme IM, Cooper AM (2000) Interleukin-6 induces
early gamma interferon production in the infected lung but is not required for
generation of specific immunity to Mycobacterium tuberculosis infection. Infect
Immun 68: 3322–3326.
Elliot AM, Hodsdon WS, Kyosiimire J, Quigley MA, Nakiyingi JS, et al. (2004)
Cytokine responses and progression to active tuberculosis in HIV-1-infected
Ugandans: a prospective study. Trans Roy Soc Trop Med & Hyg 98: 660–670.
Whalen CC, Chunda A, Zalwango S, Nshuti L, Jones-Lopez E, et al. (2006)
Immune correlates of acute Mycobacterium tuberculosis infection in household
contacts in Kampala Uganda. Am J Trop Med Hyg 75: 55–61.
Demissie A, Wassie L, Abebe M, Aseffa A, Rook G, et al. (2006) The 6
Kilodalton early secreted antigenic target-responsive, asymptomatic contacts of
tuberculosis patients express elevated levels of interleukin-4 and reduced levels of
gamma interferon. Infect Immun 74: 2817–2822.
Hill PC, Brookes RH, Fox A, Jackson-Sillah D, Jeffries DJ, et al. (2007)
Longitudinal assessment of an ELISPOT test for Mycobacterium tuberculosis
infection. PLoS Med 4: e192.
[Anonymous] (2009) World Health Organization statistics 2009.
Talat N, Shahid F, Dawood G, Hussain R (2009) Dynamic changes in
Biomarker Profiles Associated with Clinical and Subclinical Tuberculosis in a
October 2009 | Volume 4 | Issue 10 | e7444
Biomarkers Associated with TST
30.
31.
32.
33.
34.
35.
36. Gaikwad AN, Sinha S (2008) Determinants of natural immunity against
tuberculosis in an endemic setting: factors operating at the level of macrophageMycobacterium tuberculosis interaction. Clin Exp Immunol 151: 414–422.
37. Ewer K, Millington KA, Deeks JJ, Alvarez L, Bryant G, et al. (2006) Dynamic
antigen-specific T-cell responses after point-source exposure to Mycobacterium
tuberculosis. Am J Respir Crit Care Med 174: 831–839.
38. Black GF, Weir RE, Chaguluka SD, Warndorff D, Crampin AC, et al. (2003)
Gamma interferon responses induced by a panel of recombinant and purified
mycobacterial antigens in healthy, non-Mycobacterium bovis BCG-vaccinated
malawian young adults. Clin Diag Lab Immunol 10: 602–611.
39. Katakura T, Miyazaki M, Kobayashi M, Herndon DN, Suzuki F (2004) CCL17
and IL-10 as effectors that enable alternatively activated macrophages to inhibit
the generation of classically activated macrophages. J Immunol 172: 1407–1413.
40. Stenger S, Modlin RL (2002) Control of Mycobacterium tuberculosis through
mammalian Toll-like receptors. Curr Opin Immunol 14: 452–457.
41. Ito T, Schaller M, Hogaboam CM, Standiford TJ, Chensue SW, et al. (2007)
TLR9 activation is a key event for the maintenance of a mycobacterial antigenspecific pulmonary granulomatous response. Eur J Immunol 37: 1–9.
42. Leal IS, Smedegard B, Andersen P, Appelberg R (1999) Interleukin-6 and
interleukin-12 participate in induction of a type 1 protective T-cell response
during vaccination with a tuberculosis subunit vaccine. Infect Immun 67:
5747–5754.
43. Young DB, Gideon HP, Wilkinson RJ (2009) Eliminating latent tuberculosis.
Trends Microbiol 17: 183–188.
High Transmission Setting: A Four-Year Follow-Study. Scand J Immunol 69:
537–546.
Crofton J (1990) Clinical features of tuberculosis. In: Seton D, Gordon A, eds.
Crofton and Douglas Respiratory Diseases. London: Blackwell Scientific. pp
395–421.
Hussain R, Shiratsuchi H, Philips M, Ellner J, Wallis RS (2001) Opsonizing
antibodies (IgG1) up-regulate monocyte proinflammatory cytokines tumor
necrosis factor-alpha (TNF-a) and IL-6 but not anti-inflammatory cytokine
IL-10 in mycobacterial antigen-stimulated monocytes-implications for pathogenesis. Clin Exp Immunol 123: 210–218.
Thakurdas SM, Hasan Z, Hussain R (2004) IgG1 antimycobacterial antibodies
can reverse the inhibitory effect of pentoxifylline on tumour necrosis factor alpha
(TNF-a) secreted by mycobacterial antigen stimulated adherent cells. Clin Exp
Immunol 136: 320–327.
Hussain R, Kaleem A, Shahid F, Dojki M, Jamil B, et al. (2002) Cytokine
profiles using whole blood assays can discriminate between tuberculosis patients
and healthy endemic controls in BCG-vaccinated population. J Immunol Meth
264: 95–108.
Wallis RS, Amir-Tahmasseb M, Ellner JJ (1990) Induction of interleukin-1 and
tumor necrosis factor by mycobacterial proteins: The monocyte western blot.
Proc Natl Acad Sci USA 87: 3348–3352.
Wallis RS, Paranjape R, Philips M (1993) Identification of two dimensional Gel
Electrophoresis of a 58-kilodalton tumor necrosis factor introducing protein of
mycobacterium tuberculosis. Infect Immun 61: 627–632.
PLoS ONE | www.plosone.org
View publication stats
9
October 2009 | Volume 4 | Issue 10 | e7444