CD3z Expression in Pulmonary TB
• JID 2006:194 (15 November) • 1385
M A J O R A R T I C L E
Decreased Expression of CD3z and Nuclear
Transcription Factor kB in Patients with Pulmonary
Tuberculosis: Potential Mechanisms and Reversibility
Arnold H. Zea,1,2Kirk S. Culotta,1,aJuzar Ali,5Carol Mason,5Hae-Joon Park,3,aJovanny Zabaleta,3Luis F. Garcia,6
and Augusto C. Ochoa1,4
1Stanley S. Scott Cancer Center, Departments of
Louisiana State University Health Sciences Center, New Orleans;
Universitaria, Facultad de Medicina, Universidad de Antioquia, Medellı ´n, Colombia
6Grupo de Inmunologı ´a Celular e Inmunogene ´tica, Sede de Investigacion
5Section of Pulmonary/Critical Care Medicine,
presenting cells and on T lymphocytes. In patients with different forms of tuberculosis, varying degrees of T cell
function—ranging from positive delayed-type hypersensitivity, in asymptomatic infected healthy individuals, to
the absence of the response, in patients with miliary or pulmonary tuberculosis (PTB)—have been reported. The
decreased expression of CD3z reported in T cells from patients with either cancer or leprosy has provided possible
explanations for the altered immune response observed in these diseases.
The present study aimed to compare the expression of CD3z, nuclear transcription factor-kB (NF-
kB), arginase activity, and cytokine production in 20 patients with PTB, in 20 tuberculin-positive asymptomatic
subjects, and in 14 tuberculin-negative control subjects.
Compared with those in tuberculin (purified protein derivative)–negative control subjects,peripheral-
blood T lymphocytes from patients with active PTB had significantly (
absence of the p65/p50 heterodimer of NF-kB. These alterations were reversed only in patients who responded to
treatment. Also reported here for the first time is that the presence of arginase activity in peripheral-blood
mononuclear-cell lysates of patients with PTB parallels high production of interleukin-10.
The presence of arginase could, in part, explain the decreased expression of CD3z. These findings
provide a novel mechanism that may explain the T cell dysfunction observed in patients with PTB.
The protective immune response against Mycobacterium tuberculosis relies both on antigen-
) decreased expression of CD3z andP ! .001
Tuberculosis (TB) is a leading cause of death due to
infectious diseases worldwide . In some areas where
coinfection with HIV and Mycobacterium tuberculosis
is common, TB rates have doubled during recent years,
Received 2 March 2006; accepted 16 June 2006; electronically published 4
Presented in part: Experimental Biology Meeting, New Orleans, 20–24 April
2002 (poster presentation 238.12).
Financial support: National Institutes of Health (grant NCI RO1-CA82689 to
A.C.O. and grant NIH-NIAAA-11760 to C.M.); Research Institute for Children (grant
RIC-2723A to A.H.Z.).
Potential conflicts of interest: none reported.
aCurrent affiliations: University of Texas M.D. Anderson Cancer Center-PDC,
Houston (K.S.C.); MOGAM Biotechnology Institute, Yogin-City, Korea (H.-J.P.).
Reprints or correspondence: Dr. Arnold H. Zea, Louisiana State UniversityHealth
Sciences Center, Stanley S. Scott Cancer Center, 533 Bolivar St., CSRB Rm. 453,
New Orleans, LA 70112 (email@example.com).
The Journal of Infectious Diseases
? 2006 by the Infectious Diseases Society of America. All rights reserved.
despite increased availability of antimycobacterialther-
apy. Protective immunity against mycobacterial infec-
macrophages, T cells, and, possibly, natural killer cells
mediate the inhibition of mycobacteria [2–4]. Studies
of immune dysfunction in TB have reported changes
to mycobacterial antigens [7, 8], altered cytokine-
production patterns [9, 10], and altered macrophage
activation and antigen presentation [11, 12]. Further-
more, the suppression of T cell responses by myco-
bacterial infections might be mediated by the increased
production of soluble factors, including interleukin
(IL)–10 or transforming growth factor–b [13, 14]. Al-
terations in the expression of CD3z and other T cell
signal-transduction proteins have been reported in
cases of cancer [15–17], leprosy , autoimmune
1386 • JID 2006:194 (15 November) • Zea et al.
negative control subjects, PPD-positive asymptomatic subjects, and patients with
pulmonary tuberculosis (PTB).
Ages, sex, and race of tuberculin (purified protein derivative [PPD])–
Male Female White
PPD negative (14)
PPD positive (20)
aHispanic or Asian; all 11 had bacille Calmette-Gue ´rin scars (none of the white or African
Americans subjects had such scars).
diseases [19, 20], and HIV  and could partially explain the
T cell dysfunction seen in patients with these diseases. The
mechanisms leading to these alterations are poorly understood.
However, previous reports have described that, in tumors, the
amino acid l-arginine plays a central role in the regulation of
T cell function [22, 23]. Elsewhere we have reported that the
depletion of l-arginine modulates CD3z expression and T cell
function in activated human T lymphocytes . Therefore,
to determine whether M. tuberculosis infection could alter T
cell function, the present study investigated the expression of
CD3z and nuclear transcription factor kB (NF-kB) in patients
with active pulmonary TB (PTB), in tuberculin (in the form
of purified protein derivative [PPD])–positive asymptomatic
subjects, and in PPD-negative healthy control subjects. In ad-
dition, the present study also investigated in these 3 groups the
correlation between T cell dysfunction and depletion, by ar-
ginase, of l-arginine in plasma. The results of the present study
may indicate a novel mechanism by which M. tuberculosis
evades the immune response.
PATIENTS, MATERIAL, AND METHODS
Study patients and subjects.
(previously approved by the Institutional Review Board of the
Louisiana State University [LSU] Health Science Center) had
been obtained, 30 mL of peripheral blood was drawnfromeach
of (1) 20 patients with active PTB as defined by clinical symp-
toms, radiological findings consistent with TB, and bacte-
riological confirmation; (2) 20 PPD-positive asymptomatic
subjects; and (3) 14 healthy PPD-negative control subjects at-
tending the LSU-Wetmore TB clinic at the LSU HealthSciences
Center in New Orleans. Table 1 shows demographic data for
each of the 3 groups. Of the 20 PPD-positive asymptomatic
subjects, 6 (Hispanic or Asian) presented with bacilleCalmette-
Gue ´rin scars. Patients with PTB were newly diagnosed, had not
received previous anti-TB treatment, and tested negative for
HIV. Patients with hepatitis, malnutrition, immunodeficiency,
other active infectious diseases, cardiac complications, immu-
nosuppressive therapies, or cancer were excluded from the
After signed informed consent
study. Successful treatment was defined as the disappearance
of the clinical signs and symptoms of infection with M. tuber-
culosis, the resolution of the radiological changes associated
with TB, and sputum culture that was negative for PTB. Of
the 20 patients with PTB, 10 were reevaluated and tested after
4 months of treatment; an additional 5 months of therapy was
required for 3 of these 10 patients.
For the intradermal Mantoux test, 0.1 mL of 5 tuberculin
units of PPD tuberculin (Connaught) was injected intrader-
mally on the anterior part of the left forearm. The test was
read 48 h after the injection. Induration of ?5 mm was con-
sidered to be a positive reaction.
Samples of venous blood (30 mL) were
diluted 1:3 in Hanks’ balanced salt solution and were separated
over Ficoll-Paque (Pharmacia Biotech). Peripheral-bloodmono-
nuclear cells (PBMCs) enriched in T cells were obtained by use
of T cell–enrichment columns (R&D Systems). The enriched
populations were 190% CD3+T cells that were used for elec-
trophoretic mobility–shift assay (EMSA) and reverse-transcrip-
tase polymerase chain reaction (RT-PCR).
were stained, for 15 min at 4?C, with 1 mg of either anti–CD3–
fluorescein isothiocyanate (clone UCHT1) or isotype control
(Beckman-Coulter). The cells were washed and incubated, for
8 min at 4?C, in PBS containing digitonin at 500 mg/mL (Wako
BioProducts) plus 2.5 mg of anti–CD3z-phycoerythrin (clone
2H2D9) antibody (Beckman-Coulter). Then the cells were
washed, resuspended, and analyzed immediately in a Coulter-
EPICS flow cytometer (Beckman-Coulter). A similar protocol
was used for the detection of CD4z (clone RPA-T4) and CD8z
(clone HIT8a) (Beckton Dickinson).
Cytoplasmic- and nuclear-extract preparation.
mic extracts were obtained by lysing
Triton X-100 (pH 7.5) and protease inhibitors, as described
elsewhere . The lysates obtained were either analyzed im-
mediately, to test for arginase activity, or frozen at ?70?C. T
cell nuclear extracts were prepared as described elsewhere .
In brief, the nuclei were lysed in 25 mL of nuclear buffer and
PBMCs in 0.5%
CD3z Expression in Pulmonary TB • JID 2006:194 (15 November) • 1387
control subject, a PPD-positive asymptomatic subject, and a patient with pulmonary tuberculosis (PTB) (upper 3 panels) and from 14 PPD-negative
control subjects, 20 PPD-positive asymptomatic subjects, and 20 patients with PTB who were tested for expression of CD3z, CD4z and CD8z (lower
Expression of CD3z in freshly isolated peripheral-blood mononuclear cells from a tuberculin (purified protein derivative [PPD])–negative
then were kept frozen, at ?70?C, for the purpose of further
analysis of NF-kB by EMSA. Protein concentration in both
types of extract was determined by use of a BCA protein assay
kit (Pierce Chemical).
Samples (2 mg) of nuclear extracts from 6 of the
PPD-negative control subjects, 14 of the PPD-positive asymp-
tomatic subjects, and 14 of the patients with PTB were prein-
cubated, for 10 min at room temperature, in gel shift–binding
buffer and 0.5 mg/mL poly-dIdC. Then, 1 mL of
(50,000 cpm/mL/sample) (Perkin Elmer Life Sciences) NF-kB
consensus oligonucleotide (Santa Cruz Biotech) was added to
the reaction mixture, and the solution was incubated for 20
min. DNA-protein complexes were resolved by electrophoresis
on 6% polyacrylamide gels, were dried, and then were auto-
radiographed by use of BIOMAX-MR (Kodak) film at ?70?C.
Total RNA from
PPD-negative control subjects, 14 of the PPD-positive asymp-
tomatic subjects, and 14 of the patients with PTB was extracted
by use of TRIzol (Invitrogen), was treated with DNAse I (In-
vitrogen), and was reverse-transcribed by use of Superscript II
(Invitrogen). PCR amplification was performed by use of CD3z
and b-actin primers published elsewhere . Each primer
(0.25 mmol/L) and 1 U of Taq DNA polymerase in PCR buffer
containing MgCl2(2.5 mmol/L), dNTP (0.4 mmol/L), and 2
mCi of32P a-dCTP (Perkin Elmer) were added to 2-mL samples
of cDNA. For CD3z, 25 PCR cycles were run, at 96?C for 30
s, then at 55?C for 40 s, and then at 72?C for 1 min; for b-
T cells from 6 of the
actin, 27 PCR cycles were run, at 95?C for 1 min, then at 54?C
for 1.5 min, and then at 72?C for 1.5 min.
Arginase activity was determined
as described elsewhere . In brief, cell lysates were added to
25 mL of Tris-HCl (50 mmol/L; pH 7.5) containing MnCl2.This
mixture was heated at 55?C–60?C for 10 min; then 150 mL of
carbonate buffer (100 mmol/L; Sigma) and 50 mL of l-arginine
(100 mmol/L; Sigma) was added, and the solution was incu-
bated at 37?C for 20 min. The reaction from l-arginine to l-
ornithine was detected, at 515 nm, after ninhydrinsolutionhad
been added and the solution had been incubated at 95?C for
Plasma levels of l-arginine.
samples from the 14 PPD-negative control subjects and the 20
patients with PTB were determined by useof high-performance
liquid chromatography, as described elsewhere . In brief, the
plasma samples were deproteinized and derivatized, and then 50
mL of each sample was tested. A standard curve with known
concentrations of l-arginine was run for each experiment.
PBMCs from 6 of the PPD-negative
control subjects, 10 of the PPD-positive asymptomaticsubjects,
and 10 of the patients with PTB that were stimulated with PPD
(10 mg/mL) were plated at
37?C for 48 h. A 0.5-mCi portion of
Elmer) was added to each well, and the wells were incubated
for an additional 18 h, at 37?C. Each condition was run in
triplicate. Cells were lysed by freezing and thawing, were har-
Levels of l-arginine in plasma
/well and were cultured at
1388 • JID 2006:194 (15 November) • Zea et al.
berculosis who responded to treatment
Reexpression of CD3z in patients with pulmonary tu-
CD3z, mean fluorescence intensitya
PatientPretreatment 4 months 9 monthsResponseb
aThe mean intensity fluorescence for control subjects was 28.5.
bNR, nonresponder; R, responder; NR-R, patients who did not respond
(i.e., were classified as NR) during the first 4 months of treatment but who
did respond (i.e., were classified as R) after additional treatment.
vested onto a Unifilter-96 GF/B (Packard), and were counted by
use of a TOPcount Microplate Scintillation Counter (Packard).
To determine the possible role that
Th2 cytokines play in the induction of arginase, samples of
PBMCs from 6 of the PPD-negative control subjects, 14 of the
PPD-positive asymptomatic subjects, and 14 of the patients
with PTB were cultured, for 48 h, in RPMI-1640 (Cambrex)
in the presence of either 10 mg/mL PPD (Statens Seruminstitut,
Copenhagen, Denmark) or anti-CD3 (30 ng/mL) plus anti-
CD28 (100 ng/mL). The supernatants were tested for human
IL-2, interferon (IFN)–g, IL-4, and IL-10, by use of ELISA
(Biosource). Samples of unstimulated PBMCs were used as
Intergroup comparisons were per-
formed by use of either Student’s t test or 2-way analysis-of-
variance using the Graph-Pad statistical program (Prism 3.0;
Graph-Pad). The association between CD3z and arginase was
estimated by use of a regression model and Pearson’s corre-
lation coefficient. was considered to be significant.P ? .05
Diminished expression of CD3z-chain in T cells from patients
Expression of CD3z was tested in samples of
PBMCs from 14 PPD-negative control subjects, 20 PPD-pos-
itive asymptomatic subjects, and 20 patients with PTB. The
upper panel of figure 1 shows representative scatter plots com-
paring the expression of CD3z and CD3e in samples of PBMCs
from a PPD-negative control subject, a PPD-positive asymp-
tomatic subject, and a patient with PTB; expression of CD3z
in the patient with PTB was lower than that in either the PPD-
negative control subject orthePPD-positiveasymptomaticsub-
ject—a decrease that was not due to a decrease in the number
of T cells, because the expression of CD3e was similar (range,
76%–85%) in all 3 of the groups. The lower panel of figure 1
shows that the mean fluorescence intensity (MFI) of CD3z was
significantly ( ) lower in the patients with PTB (meanP ! .001
? SD, 14.35 ? 2.5) than in the PPD-negative control subjects
(mean ? SD, 28.55 ? 1.2), and similar and significant de-
creased expression of z-chain was also observed in CD4 and
CD8 T cell subsets. Of the 20 PPD-positive asymptomatic sub-
jects, 4 (20%) had markedly decreased expression of CD3z,
although none of these 4 had signs or symptoms of TB.
Reexpression of CD3z to normal levelsaftersuccessfultreat-
ment of patients with PTB.
Of the 20 patients with PTB who
were receiving anti-TB therapy, 10 were retested for expression
of CD3z. As summarized in table 2, 7 of these 10 patients
responded to treatment after 4 months, with expression of
CD3z (MFI, 28.5) having recovered to levels similar to those
in the PPD-negative control subjects; in contrast, the other 3
of these 10 patients did not respond to treatment and had
decreased expression of CD3z, whichrecoveredtonormallevels
after additional treatment.
Diminished expression of CD3z in patients with PTB—not
due to decreased expression of CD3z mRNA.
used to investigate whether the decreased expression of CD3z
in patients with PTB was the result of a decrease in expression
of CD3z mRNA. The levels of CD3z mRNA were normalized
by use of b-actin as internal control. Representative results of
the integrated densitometry values of the ratio between ex-
pression of CD3z and expression of b-actin, in PPD-negative
control subjects (3/6), PPD-positive asymptomatic subjects (4/
14), and patients with PTB (5/14), are shown in the bar graph
in figure 2, wherethisratioisexpressedasapercentage.Analysis
comparing the expression of CD3z mRNA and of b-actin
mRNA failed to show significant intergroup differences in the
expression of CD3z mRNA (P p .81
Decrease in nuclear expression of p65 of NF-kB in T cells
from patients with PTB.
NF-kB plays an important role in
T cell activation after antigenic stimulation. After freshly iso-
lated T cells had been activated with anti-CD3 plus anti-CD28
antibodies, nuclear extracts were obtained and were tested, by
EMSA, for NF-kB. Figure 3 shows representative EMSA data
for 5 PPD-negative control subjects, 5 PPD-positive asymp-
tomatic subjects, and 5 patients with PTB. The expression of
both the p65/p50 heterodimer and the p50/p50 homodimer of
NF-kB was observed in T cells from the 5 PPD-negativecontrol
subjects (lanes 1–5) and the 5 PPD-positive asymptomaticsub-
jects (lanes 6–10); in contrast, T cells from 4 of the 5 patients
with PTB (lanes 11, 12, 14, and 15) lacked heterodimer p65/
p50, expressing only homodimer p50/p50. Data from all pa-
tients and controls are summarized, in tabular format, under
CD3z Expression in Pulmonary TB • JID 2006:194 (15 November) • 1389
protein derivative [PPD])–negative subjects, 4 PPD-positive asymptomatic subjects, and 5 patients with pulmonary tuberculosis (PTB) are shown. In all
cases, b-actin was used as an internal control. Densitometry analysis (by analysis of variance), using the ratio between expression of CD3z and
expression of b-actin, did not show statistically significant intergroup differences in expression of CD3z mRNA.
Detection of CD3z mRNA in T cells by reverse-transcriptase polymerase chain reaction. Representative samples from 3 tuberculin (purfied
the gels in figure 3. The T cells of 5 (36%) of the 14 patients
with PTB who were tested and of 13 (93%) of the 14 PPD-
positive asymptomatic subjects who were tested expressed the
p65/p50 heterodimer of NF-kB. Interestingly, all of the patients
with PTB who lacked NF-kB p65 (pretreatment) also had de-
creased expression of CD3z—and heterodimer p65/p50 was
restored after successful treatment (data not shown).
Correlation between elevated levels of arginase and low
expression of CD3z in PBMCs from patients with PTB.
vivo, levels of l-arginine are tightly regulated by arginase.
Therefore, we tested the levels of arginase inPBMClysatesfrom
patients with PTB. As shown in figure 4A, arginase activity was
significantly elevated in patients with PTB (mean, 127.8 ? 88.1
nmol [ ]) and in PPD-positive asymptomatic subjectsP p .002
(mean, 22.67 ? 28.4 nmol [P p .024
in PPD-negative control subjects (mean, 3.03 ? 3.3 nmol). As
shown in figure 4B, statistical analysis using Pearson’s corre-
lation coefficient revealed that, in the 20 patients with PTB,
there was a highly significant (
( ) between decreased expressionof CD3zandhighr p ?0.8256
arginase activity as measured by the conversion of l-arginine
to l-ornithine. As shown in figure 4C, the levels of l-arginine
in the plasma of patients with PTB (mean, 88 ? 7 mmol/L)
]), compared with that
) inverse associationP ! .001
were significantly less (
(mean, 153 ? 16 mmol/L).
Alteration of cell proliferation and cytokine production in
patients with PTB.
In response to stimulation with PPD,
PBMCs from patients with PTB had a significantly lower pro-
liferation rate than did those from PPD-negative control sub-
jects ( ) and PPD-positive asymptomatic subjects (PP p .002
p.007), as shown in figure 5C; the rate was not different
between the PPD-negative control subjects and the PPD-pos-
itive asymptomatic subjects. As shown in figures 5A and 5B,
in the presence of PPD the PBMCs did not produce IFN-g and
shifted to a predominantly down-regulatory IL-10 response,
especially in the PPD-positive control subjects and the patients
with PTB. In addition, these cells did not produce either IL-2
or IL-4 (data not shown). In light of these findings, a test was
performed to determine whether the inability, by PBMCs from
PPD-positive control subjects and from patients with PTB, to
produce Th1 cytokines was specific to stimulation with my-
cobacterial antigens; stimulation of PBMCs by anti-CD3 plus
anti-CD28 abolished the effect of mycobacterial antigens, and,
as seen in figure 5A, the cells were able respond normally,
indicating that T cell unresponsiveness in patients with TB is
antigen specific and related to mycobacterial antigens.
) than those in control subjectsP p .004
1390 • JID 2006:194 (15 November) • Zea et al.
from 6 tuberculin (in the form of purified protein derivative [PPD])–negative control subjects, 14 PPD-positive asymptomatic subjects, and 14 patients
with pulmonary tuberculosis (PTB). To determine translocation of NF-kB to the nucleus, T cells were stimulated for 1 h in the presence of 30 ng/mL
anti-CD3 plus 100 ng/mL anti-CD28. Representative data for 5 PPD-negative control subjects, 5 PPD-positive asymptomatic subjects, and 5 patients
with PTB show that heterodimer p65/p50 was absent from T cells from most of the patients with PTB but was present in most of the PPD-negative
control subjects and PPD-positive asymptomatic subjects. Homodimer p50/p50 was expressed consistently in all groups. NF-kB dimers are indicated
by arrows. The tabular data below the gels summarize the results for the 3 groups.
Electrophoretic mobility-shift assay using nuclear transcription factor (NF)–kB consensus oligonucleotides and nuclear extracts of T cells
A large body of clinical and experimental evidence has shown
that the level of immune competence varies among patients
with mycobacterial diseases. The different clinicalpresentations
of TB appear to be associated with the degree of alteration of
the immune response against the mycobacteria. The present
study has shown that patients with active PTB (1) have a sig-
nificant decrease in the expression of CD3z, which is an im-
portant signal-transduction protein in T cell activation, and(2)
lack the p65/p50 NF-kB heterodimer, which is important in
the activation of certain T cell genes. The data suggest that
these alterations could in part explain the loss of a protective
immune response against M. tuberculosis, and they may help
us to understand the mechanisms that leadtoTcelldysfunction
The results of the present study show that, in a high per-
centage of patients with PTB, the expression of CD3z is sig-
nificantly lower than that in PPD-negative control subjects.
Interestingly, although the PPD-positive asymptomaticsubjects
studied did not show any signs of detectable active disease, a
small percentage (20% [4/20]) also showed decreased expres-
sion of CD3z, suggesting that some individuals might develop
alterations in the expression of signal-transduction proteins
during latent infection due to M. tuberculosis. Therefore, it will
be important to continue to monitor these PPD-positiveasymp-
tomatic subjects, to determine whether these signal-transduc-
tion changes precede the development of the active disease; in
addition, similar decreased expression of CD3z was observed
of CD3z has been reported in infectious diseases suchasleprosy
and AIDS and at the site of mycobacterial infection in TB [18,
21, 28]. As has clearly been seen in patients with leprosy ,
changes in the expression of CD3z and p56lckand the absence
of NF-kB p65 in the nucleus are more predominant in patients
with lepromatous leprosy than in either patients with tuber-
culoid leprosy or healthy control subjects, and therefore the
alterations seen in those patients might be a reflection of im-
munological incompetence. Likewise, the signal-transduction
alterations seen in patients with PTB may reflect both the dif-
ferent stages of immune competence and T cell–function im-
pairments that may result in the induction of anergy , as
is commonly seen in miliary TB.
Nuclear transcription factors such as NF-kB appear to play
a central role in both T cell activation and the regulation of
certain cytokine genes . The present study found that the
64% of patients with PTB who lacked the p65/p50 heterodimer
CD3z Expression in Pulmonary TB • JID 2006:194 (15 November) • 1391
cells (PBMCs) from 14 tuberculin (in the form of purified proteinderivative
[PPD])–negative control subjects, 20 PPD-positive asymptomatic subjects,
and 20 patients with pulmonary tuberculosis (PTB), which were tested
for arginase activity measured in terms of conversion of L-arginine to L-
ornithine. PBMCs from patients with PTB produced more arginase
( ) than did those from healthy control subjects. B, Analysis byP p .02
Pearson’s correlation coefficient, showing negative correlation (rp
?0.8256) between arginase activity and expression of CD3z in PBMCs
from patients with PTB. C, Levels of L-arginine, which were significantly
higher ( ) in plasma from PPD-negative control subjects than inP p .004
that from patients with PTB.
A, Cytoplasmic extracts of peripheral-blood mononuclear
clear cells (PBMCs) from tuberculin (in the form of purified protein de-
rivative [PPD])–negative control subjects, PPD-positive asymptomaticsub-
jects, and patients with pulmonary tuberculosis (PTB), stimulated with
either PPD (10 mg/mL) or anti-CD3 plus anti-CD28 and then measured by
ELISA of supernatants after 48 h in culture. C, T cell proliferation, which,
after stimulation with PPD for 48, was significantly decreased (P p
) in patients with PTB. Unstimulated (NS) PBMCs from all 3 groups.002
were used as controls in the experiments.
A and B, Cytokine production by peripheral-blood mononu-
also had decreased expression of CD3z, suggesting that it is
possible, although not yetproven,thattheTcell–receptorsignal
via CD3z could be a determining factor in the release of NF-
kB p65 from its inhibitor, IkB, and in its subsequent translo-
cation to the nucleus to initiate cell activation and cytokine-
gene regulation. Therefore, a more definitive understanding of
these interactions in TB will require both an extensive analysis
of more patients and testing for other transcription factors also
known to play important roles in cytokine-gene regulation.
in disease are poorly understood. In vivo, levels of l-arginine
are regulated, in great part, by a balance between arginase and
inducible nitric oxide (NO) synthase, 2 enzymes using l-ar-
ginine for the production of l-ornithine and NO, respectively
[22, 30]. NO is an important cytotoxic mechanism in mac-
1392 • JID 2006:194 (15 November) • Zea et al.
rophages, whereas l-ornithine is a precursor of polyamines
essential for cell proliferation [31, 32]. Arginase can be induced
in macrophages by a variety of cytokines, including IL-4, IL-
13, and IL-10  and transforming growth factor–b , as
well as by lipopolysaccharides . In murine peritoneal mac-
rophages, stimulation with IL-4 plus IL-13 induces arginase
production that rapidly reduces levels of l-arginine, resulting
in the induction of T cell dysfunction . We recently have
reported that, in the absence of l-arginine, stimulated T cells
have both decreased CD3z expression and low proliferation,
observations similar to those is seen in patients with cancer
and other infectious diseases .
The high levels of arginase seen in the PBMCs of patients
with PTB correlates with decreased expression of CD3z and
contrasts with the low arginase levels and normal expression
of CD3z that were seen in the PPD-negative control subjects;
therefore, arginase may play a role in the metabolic regulation
of plasma levels of l-arginineininfectedindividuals.Theresults
of the present study also show that levels of l-arginine in pa-
tients with PTB were significantly lower than those seen in the
PPD-negative control subjects—allowing one to speculate that
these low levels may contribute to decreased expression of
CD3z, as has been seen in T cells cultured in the absence of
l-arginine . Although the number of samples tested in the
fore cannot be drawn, the overall results suggest that patients
with PTB produce high levels of arginase that deplete levels of
l-arginine, which, in turn, induces T cell dysfunction charac-
terized by decreased cell proliferation, increased production of
IL-10, and decreased production of both IL-2 (data not shown)
and IFN-g. It is unclear which cells are producing arginase—
or which signals trigger the increase in arginase production—
in patients with PTB; however, it is possible that mycobacterial
infection of macrophages and/or increased IL-10 production
may be responsible for this event.
Preliminary data show that patients responding to anti-TB
therapy reexpress all of the signal-transduction proteins, includ-
ing CD3z and NF-kB p65. The recovery of these proteins could
be dueto different factors, suchasdecreaseineithertheantigenic
burden or the inflammatory response (i.e., IL-10), increase in
Th1-type cytokines, alternative macrophage activation, or de-
crease in the levels of arginase. The findings of the present study
also suggest that the reexpression of signal-transduction mole-
cules may benefit the clinical outcome in patients, because those
patients who did not respond to treatment maintained dimin-
ished expression of CD3z and NF-kB and showed no signs of
an improved immune response. In addition, preliminary obser-
vations show that these patients still have high levels of arginase
activity and high production of IL-10. However, further testing
of a larger number of patients, along with a more complete
follow-up, will be necessary to confirm these findings.
In summary, the present study found that a significant num-
ber of the patients with PTB who were studied had alterations
in the expression of several T cell signal-transduction proteins,
alterations that were similar to those observed in patients with
cancer or other diseases. The similarity of T cell signal-trans-
duction alterations in diseases withdifferentpathophysiological
characteristics suggests the possibility that a common mecha-
nism causes such changes. The presence of increased levels of
arginase in macrophages of patients with PTB suggests that this
enzyme plays a role in regulation of the immune response, in
light of the fact that high levels of arginase play an important
role in both the down-regulation of CD3z and the induction
of T cell dysfunction. These observations of molecular and
functional characteristics in TB may provide new tools to study
and monitor patients, to determine how these characteristics
effect the development of immune dysfunction, and to study
new pathways to block suppressor mechanisms. This endeavor
would reestablish that the function of the immunesystemcom-
bined with anti-TB therapy will benefit the clinical outcome in
patients with TB.
The authors thank the patients who participated in the study; the nurses
and staff of the LSU-Wetmore TB clinic in New Orleans; Dr. Seth Pincus
(Research Institute for Children), for his critical review of the manuscript;
and Dr. Cruz Velasco (LSU Health Science Center School of PublicHealth),
for his contribution to the statistical analysis.
1. Bloom BR , Small PM. The evolving relation between humans and
Mycobacterium tuberculosis. N Engl J Med 1998;338:677–8.
2. Tsukaguchi K, Balaji KN, Boom WH. CD4+ab T cell and gd T cell
responses to Mycobacterium tuberculosis. Similarities and differencesin
Ag recognition, cytotoxic effector function, and cytokine production.
J Immunol 1995;154:1786–96.
3. Flynn JL, Chan J. Immunology of tuberculosis. Annu Rev Immunol
4. Cooper AM, Flynn JL. The protective immune response to Mycobac-
terium tuberculosis. Curr Opin Immunol 1995;7:512–6.
5. Tsuyuguchi I, Shiratsuchi H, Fujiwara H, Teraoka O. Nonspecific re-
cruitment of lymphocytes in purified protein derivative-induced lym-
phocyte proliferative response of patients with tuberculosis. Infect Im-
6. Beck JS, Potts RC, Kardjito T, Grange JM. T4 lymphopenia in patients
with active pulmonary tuberculosis. ClinExpImmunol1985;60:49–54.
7. Toossi Z, Kleinhenz ME, Ellner JJ. Defective interleukin 2 production
and responsivenessin humanpulmonarytuberculosis.JExpMed1986;
8. Kleinhenz ME, Ellner JJ. Antigen responsiveness during tuberculosis:
regulatory interactions of T cell subpopulations and adherent cells. J
Lab Clin Med 1987;110:31–40.
9. Lowrie DB. Is macrophage death on the field of battle essential to
victory, or a tactical weakness in immunity against tuberculosis? Clin
Exp Immunol 1990;80:301–3.
10. Johnson BJ, McMurray DN. Cytokine gene expression by cultures of
human lymphocytes with autologous Mycobacterium tuberculosis-in-
fected monocytes. Infect Immun 1994;62:1444–50.
CD3z Expression in Pulmonary TB • JID 2006:194 (15 November) • 1393
11. Sibley LD, Hunter SW, Brennan PJ, Krahenbuhl JL. Mycobacterial
lipoarabinomannan inhibits gamma interferon-mediated activation of
macrophages. Infect Immun 1988;56:1232–6.
12. Hmama Z, Gabathuler R, Jefferies WA, de Jong G, Reiner NE. Atten-
uation of HLA-DR expression by mononuclear phagocytes infected
with Mycobacterium tuberculosis is related to intracellularsequestration
of immature class II heterodimers. J Immunol 1998;161:4882–93.
13. Boussiotis VA, Tsai EY, Yunis EJ, et al. IL-10-producing T cellssuppress
immune responses in anergic tuberculosis patients. J Clin Invest 2000;
14. Toossi Z, Gogate P, Shiratsuchi H, Young T, Ellner JJ. Enhanced pro-
duction of TGF-b by blood monocytes from patients with active tu-
berculosis and presence of TGF-b in tuberculous granulomatous lung
lesions. J Immunol 1995;154:465–73.
15. Mizoguchi H, O’Shea JJ, Longo DL, Loeffler CM, McVicar DW, Ochoa
tumor-bearing mice. Science 1992;258:1795–8.
16. Finke JH, Zea AH, Stanley J, et al. Loss of T-cell receptor zeta chain
and p56lck in T-cells infiltrating human renal cell carcinoma. Cancer
17. Zea AH, Curti BD, Longo DL, et al. Alterations in T cell receptor and
signal transduction molecules in melanoma patients. Clin Cancer Res
18. Zea AH, Ochoa MT, Ghosh P, et al. Changes in expression of signal
transduction proteins in T lymphocytes of patients with leprosy. Infect
19. Liossis SN, Ding XZ, Dennis GJ, Tsokos GC. Altered pattern of TCR/
CD3-mediated protein-tyrosyl phosphorylation in T cellsfrompatients
with systemic lupus erythematosus. Deficient expression of the T cell
receptor zeta chain. J Clin Invest 1998;101:1448–57.
20. MatsudaM, UlfgrenAK, Lenkei R, etal.Decreasedexpressionofsignal-
transducing CD3 z chains in T cells from the joints and peripheralblood
of rheumatoid arthritis patients. Scand J Immunol 1998;47:254–62.
21. Stefanova I, Saville MW, Peters C, et al. HIV infection–induced post-
translational modification of T cell signaling molecules associatedwith
disease progression. J Clin Invest 1996;98:1290–7.
22. Albina JE, Caldwell MD, Henry WL Jr, Mills CD. Regulation of mac-
rophage functions by L-arginine. J Exp Med 1989;169:1021–9.
23. Brittenden J, Park KG, Heys SD, et al. L-arginine stimulates host de-
fenses in patients with breast cancer. Surgery 1994;115:205–12.
24. Zea AH, Rodriguez PC, Culotta KS, et al. l-Arginine modulates CD3z
expression and T cell function in activated human T lymphocytes.Cell
25. Ghosh P, Sica A, Young HA, et al. Alterations in NF kappa B/Rel family
proteins in splenic T-cells from tumor-bearing mice and reversal fol-
lowing therapy. Cancer Res 1994;54:2969–72.
26. Rodriguez PC, Quiceno DG, Zabaleta J, et al. Arginase I production
in the tumor microenvironment by mature myeloid cells inhibits T-
cell receptor expression and antigen-specific T-cell responses. Cancer
27. Zea AH, Rodriguez PC, Atkins MB, et al. Arginase-producingmyeloid
suppressor cells in renal cell carcinoma patients: amechanismoftumor
evasion. Cancer Res 2005;65:3044–8.
28. Seitzer U, Kayser K, Hohn H, et al. Reduced T-cell receptor CD3z-
chain protein and sustained CD3e expression at the site of mycobac-
terial infection. Immunology 2001;104:269–77.
29. Grilli M, Chiu JJ, Lenardo MJ. NF-kappa B and Rel: participants in a
multiform transcriptional regulatory system. Int Rev Cytol 1993;143:
30. Wu G, Morris SM Jr. Arginine metabolism: nitric oxide and beyond.
Biochem J 1998;336:1–17.
31. Hibbs JB Jr, Taintor RR, Vavrin Z. Macrophage cytotoxicity: role for
L-arginine deiminase and imino nitrogen oxidation to nitrite. Science
32. Tabor CW, Tabor H. Polyamines. Annu Rev Biochem 1984;53:749–90.
33. Munder M, Eichmann K, Moran JM, Centeno F, Soler G, Modolell
M. Th1/Th2-regulated expression of arginase isoforms in murine mac-
rophages and dendritic cells. J Immunol 1999;163:3771–7.
34. Boutard V, Havouis R, Fouqueray B, Philippe C, Moulinoux JP, Baud
L. Transforming growth factor-beta stimulates arginase activityinmac-
rophages. Implications for the regulation of macrophagecytotoxicity.
J Immunol 1995;155:2077–84.
35. Ryan JL, Yohe WB, Morrison DC. Stimulation of peritoneal cell ar-
ginase by bacterial lipopolysaccharides. Am J Pathol 1980;99:451–61.
36. Rodriguez PC, Zea AH, DeSalvo J, et al. l-Arginine consumption by
macrophages modulates the expression of CD3 z chain in T lympho-
cytes. J Immunol 2003;171:1232–9.