Medroxyprogesterone Acetate Alters Mycobacterium
Bovis BCG-Induced Cytokine Production in Peripheral
Blood Mononuclear Cells of Contraceptive Users
Le ´anie Kleynhans1, Nelita Du Plessis1, Gillian F. Black1, Andre ´ G. Loxton1, Martin Kidd2, Paul D. van
Helden1, Gerhard Walzl1, Katharina Ronacher1*
1Division of Molecular Biology and Human Genetics, MRC Centre for Molecular and Cellular Biology, DST/NRF Centre of Excellence for Biomedical TB Research, Faculty of
Health Sciences, Stellenbosch University, Tygerberg, South Africa, 2Department of Statistics and Actuarial Sciences, Stellenbosch University, Stellenbosch, South Africa
Most individuals latently infected with Mycobacterium tuberculosis (M.tb) contain the infection by a balance of effector and
regulatory immune responses. This balance can be influenced by steroid hormones such as glucocorticoids. The widely used
contraceptive medroxyprogesterone acetate (MPA) possesses glucocorticoid activity. We investigated the effect of this
hormone on immune responses to BCG in household contacts of active TB patients. Multiplex bead array analysis revealed
that MPA demonstrated both glucocorticoid and progestogenic properties at saturating and pharmacological
concentrations in peripheral blood mononuclear cells (PBMCs) and suppressed antigen specific cytokine production.
Furthermore we showed that PBMCs from women using MPA produced significantly lower levels of IL-1a, IL-12p40, IL-10, IL-
13 and G-CSF in response to BCG which corresponded with lower numbers of circulating monocytes observed in these
women. Our research study is the first to show that MPA impacts on infections outside the genital tract due to a systemic
effect on immune function. Therefore MPA use could alter susceptibility to TB, TB disease severity as well as change the
efficacy of new BCG-based vaccines, especially prime-boost vaccine strategies which may be administered to adult or
adolescent women in the future.
Citation: Kleynhans L, Du Plessis N, Black GF, Loxton AG, Kidd M, et al. (2011) Medroxyprogesterone Acetate Alters Mycobacterium Bovis BCG-Induced Cytokine
Production in Peripheral Blood Mononuclear Cells of Contraceptive Users. PLoS ONE 6(9): e24639. doi:10.1371/journal.pone.0024639
Editor: Erika Martins Braga, Universidade Federal de Minas Gerais, Brazil
Received June 6, 2011; Accepted August 15, 2011; Published September 8, 2011
Copyright: ? 2011 Kleynhans 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: This work was supported by the National Research Foundation of South Africa [grant number FA2007022300004] to KR; The South African Tuberculosis
and AIDS Training (SATBAT) program, funded by the National Institute of Health/Fogarty International Center [grant number 1U2RTW007370/3] to KR; SATBAT
[grant number 1U2RTW007370/3] to LK and the Harry Crossley Foundation. This work was also supported by the Bill and Melinda Gates Foundation through
Grand Challenges in Global Health (GCGH), grant no. 37772. 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: firstname.lastname@example.org
Tuberculosis (TB) caused by Mycobacterium tuberculosis (M.tb) is a
major health problem and the World Health Organization has
estimated a global incidence of 9.4 million TB cases and 1.6
million TB related deaths every year . South Africa and
Swaziland are the two countries which have the highest incidence
at 960 and 1200 per 100 000 population per year, respectively.
Even though most individuals infected with M.tb contain the
infection by maintaining a balance of regulatory and effector
immune responses, this balance can be influenced by steroid
hormones such as glucocorticoids (GCs) . It has been shown
that stress-induced activation of the hypothalamus-pituitary-
adrenal axis causes reactivation of TB in mice  and that
administration of exogenous GCs induces reactivation of TB in
animal models [4,5] and increases the risk of developing TB in
humans . Furthermore endogenous concentrations of the GC
cortisol have been shown to inhibit mycobacterial antigen-induced
cell proliferation and IFNc production in peripheral blood
mononuclear cells (PBMCs) .
Medroxyprogesterone acetate (MPA) is used as a three month
injectable progestin-only contraceptive and is the most commonly
used contraceptive in South Africa and other TB endemic areas.
MPA is freely available at public health care clinics in South Africa
and is favored by women and health care workers because it is
administered only four times a year. MPA is pharmacologically
unique compared to other synthetic progestins (such as the 2
month injectable contraceptive norethisterone (NET)) as it binds
with high affinity not only to the progesterone receptor (PR), but
also to the glucocorticoid receptor (GR)  and can alter the
transcription of GR-regulated genes . Due to its GC activity it is
possible that doses of MPA used for endocrine therapy could have
significant immune modulatory effects and impact on susceptibility
as well as clinical manifestation of infectious diseases. There is
evidence that MPA increases susceptibility to vaginal simian-
human immunodeficiency virus transmission and suppresses
antiviral immune responses in Rhesus Macaques . Similarly
a study in mice found that MPA treated animals had increased
vaginal infectability with herpes simplex virus (HSV)-2 . In
women an association was found between MPA use and viral
shedding of HIV and HSV, from vaginal epithelium cells [12,13].
Another study reported a significant association between MPA use
and the acquisition of sexually transmitted bacterial infections
. Studies on whether MPA use itself increases the risk of
PLoS ONE | www.plosone.org1September 2011 | Volume 6 | Issue 9 | e24639
acquiring HIV are conflicting [15–17]. Recently, MPA use was
associated with a significantly higher risk of acquiring HIV, but
was not associated with disease progression of HIV [18,19].
The effect of hormone based contraceptives on immune
responses in the context of mycobacterial infections has never
before been investigated. This is surprising as MPA is mainly used
in low socioeconomic areas with a high TB burden. Furthermore
MPA is the recommended contraceptive for active TB patients as
anti-TB drugs like rifampicin upregulate P450 cytochromes which
rapidly metabolize estradiol containing contraceptives, rendering
them ineffective . This study is the first to show that MPA
alters the secretion of several cytokines in response to BCG in vitro
and changes the BCG-induced memory immune response in MPA
users compared to non-contraceptive users.
Materials and Methods
The Ethics Committee of the University of Stellenbosch (N05/
11/187) and the City of Cape Town City Health approved the
protocols for the study, which was conducted according to the
Helsinki Declaration and International Conference of Harmonisa-
tion guidelines. Written informed consent was obtained from all
For this cross-sectional case control study, we randomly
recruited female household contacts (HHCs) of active TB patients
between the ages of 15 and 45, who were enrolled at a TB Clinic
in the Ravensmead/Uitsig area of Cape Town during 2008.
Pulmonary TB index cases were self-reporting with a first episode
of TB and had two sputum smears positive for acid fast bacilli.
HHCs were defined as individuals living in the same house as an
adult pulmonary TB patient who was diagnosed not more than 2
months before recruitment of the HHC. HHCs were tuberculin
skin test (TST) positive with an induration of $10 mm 48–
72 hours after an intradermal injection with M.tb purified protein
derivative. Study participants were not taking any steroid
treatment other than contraceptives at the time of recruitment.
HIV positive, pregnant and sterilized women, women that used
contraceptives other than MPA and those who’s PBMCs did not
produce IFNc in response to BCG were excluded from the study.
Two hundred and seventy-four women were recruited of which
111 were using contraceptives (40.5%). The majority of contra-
ceptive users (65/111; 59.3%) used MPA; 14/111 (12.4%) used
NET; 14/111 (12.4%) used combined oral contraceptives; 14/111
(12.4%) had been sterilized and 4/111 (3.5%) used an intra-
uterine device. The mean age of MPA users and non-
contraceptive users was 2867.03 and 2668.85 years respectively
and the mean duration of MPA use 33636 months. For the
cytokine analysis, PBMCs from a subset of participants were used
as indicated in the figure legends. The mean age of MPA users in
this subset was not significantly different from controls (MPA users:
2966.4, controls: 2765.1) and the mean duration of MPA use was
39639 months. These study participants were from different
households and not related to each other with exception of two
sisters. However one of them was a MPA user and the other was
part of the control group. Therefore the cytokine expression
patterns in the two different groups cannot be attributed to
consanguinity. At time of recruitment duration of exposure of
study participants to the TB patients was recorded. The majority
of MPA users as well as the majority of controls spent more than
12 hours per day with the TB index case and there was no
significant difference in exposure score between the two groups.
Isolation and stimulation of PBMCs
PBMCs were isolated from whole blood using Histopaque
(Sigma-Aldrich, SA) density gradient centrifugation. Cells were
cultured at a density of 16106cells/well in 24-well tissue culture
plates (Greiner Bio-one, North Carolina, USA). PBMCs were
cultured as unstimulated controls and with Pasteur BCG
(56105CFU/ml, MOI 1:5) in the presence and absence of
cortisol (hydrocortisone), medroxyprogesterone 17-acetate (MPA)
and 4-pregnene-3,20-dione progesterone (Sigma-Aldrich). Prior to
use in culture the hormones were dissolved in ethanol (Merck,
New Jersey, USA) and stored at 220uC. BCG cultures were
grown in Difco Middlebrook 7H9 (BD Pharmingen, San Diago,
USA), supplemented with 0.2% glycerol (Sigma), 0.05% Tween 80
(Sigma) and 10% Middlebrook oleic acid albumin dextrose
catalase (OADC) enrichment (BD). Aliquots of BCG cultures at
logarithmic phase were frozen at 280uC in 10% glycerol (Sigma).
The number of viable bacteria was assessed by thawing a frozen
aliquot and plating serial dilutions on Middlebrook 7H11 (BD)
agar plates. Unstimulated controls and test cultures were exposed
to equal concentrations of ethanol with the final concentration not
exceeding 0.1%. Phytohaemagglutinin (PHA, Sigma-Aldrich) was
included as positive control. The plates were incubated at 37uC
and 5% CO2for six days. Supernatants were harvested on day
three and day six post stimulation and frozen at -80uC until
Quantification of cytokine levels by ELISA and Multiplex
The IFNc levels in the culture supernatants were quantified by
ELISA according to a protocol previously described , however
with different antibodies (mouse anti-human IFNc monoclonal
antibody/biotinylated detection antibody (BD) and detection
solution (Streptavidin-peroxidase together with o-phenylenedi-
amine (Sigma-Aldrich)). The standard curve ranged from 31 pg/
ml to 4000 pg/ml. Diluted (1:10) whole blood taken from a
healthy volunteer was stimulated with PHA (10 mg/ml) for four
days and the undiluted culture supernatant included on all plates
as an internal control.
Human 21-plex Luminex assays (Merck-Millipore, Missouri,
USA) were used to simultaneously quantify the levels of the
following cytokines in the day three culture supernatants:
interleukin (IL)-1a, IL-1b, IL-2, IL-4, IL-5, IL-6, IL-8 (CXCL8),
IL-10, IL-12p40, IL-13, IL-17, interleukin 1 receptor antagonist
(IL-1ra), epidermal growth factor (EGF), granulocyte colony
stimulating factor (G-CSF), granulocyte monocyte stimulating
factor (GM-CSF), interferon (IFN)c, interferon inducible protein
10 (IP-10 or CXCL10), monocyte chemotactic protein-1 (MCP-1
or CCL2), macrophage inflammatory protein (MIP)-1b (CCL4),
transforming growth factor (TGF)a and tumor necrosis factor
(TNF)a. Human 29-plex Luminex assays (Merck-Millpore) were
used to measure the levels of IL-1a, IL-1b, IL-2, IL-4, IL-5, IL-6,
IL-7, IL-8, IL-10, IL-12p40, IL-12p70, IL-13, IL-15, IL-17, IL-
1ra, soluble CD40 ligand (sCD40L), EGF, eotaxin (CCL11),
fractalkine, G-CSF, GM-CSF, IFNc, IP-10, MCP-1, MIP-1a
(CCL3), MIP-1b, TGFa, TNFa and vascular endothelial growth
factor (VEGF) in the day six culture supernatants. The assay was
done according to manufacturer’s instructions and samples were
evaluated in duplicate.
The cytokine concentrations in the unstimulated and BCG
stimulated (6cortisol, MPA and progesterone) PBMC culture
supernatants were measured on a Bio Plex platform (Bio PlexTM,
Bio Rad Laboratories). Two quality controls included in each kit
were run in duplicate on each plate. Levels of all analytes in the
quality controls were within the expected ranges. A standard curve
MPA Alters BCG-Induced Cytokine Secretion
PLoS ONE | www.plosone.org2September 2011 | Volume 6 | Issue 9 | e24639
ranging from 3.2 pg/ml to 10 000 pg/ml was used for all
cytokines. Bio-Plex Manager Software, version 4.1.1, was used to
analyze the data.
Data and Statistical analysis
For the multiplex bead array assay, cytokine levels below the
lowest and above the highest concentration of the standard curve
were extrapolated by the Bio-Plex Manager software and are
reported here. Luminex data were analyzed by a mixed model
repeated measures one-way analysis of variance (ANOVA) with a
Fisher LSD Post-Hoc test. Luminex results are shown as least-
squares (LS) means with 95% confidence intervals (CI). A p-
value#0.05 (when compared to the M.bovis BCG only responses)
was accepted as significant. Data analysis was done using Statistica
9, Statsoft (Ohio, USA). Unbiased clustering of cytokine secretion of
MPA users and non-contraceptive users and generation of a heat
map was done using Qlucore Omics Explorer (Lund, Sweden).
MPA and cortisol inhibit IFNc production in BCG-
stimulated PBMCs in vitro
Given the importance of IFNc in protective immunity to TB we
investigated whether high concentrations of MPA influence IFNc
production of BCG stimulated PBMCs in vitro. We compared
MPA to cortisol and progesterone as these hormones bind to the
GR and PR respectively. As mentioned previously MPA can bind
and activate both steroid receptors. Furthermore we used
saturating ligand concentrations (10 mM) to discount the different
binding affinities of the hormones to the respective receptors.
Supernatants of PBMCs stimulated with BCG in the presence and
absence of the three hormones were collected after three (n=29)
and six days (n=35) of culture and stored at 280uC until ELISA
analysis. Our ELISA results indicated that cortisol nearly
abolished BCG-induced IFNc response in the PBMC supernatants
after three (p,0.001; figure 1) and six days (p,0.001; data not
shown). MPA inhibited the IFNc responses to a similar extent as
cortisol (p,0.001). In contrast to MPA and cortisol, progesterone
did not down regulate the secretion of IFNc three days post
stimulation, however, at six days post stimulation, progesterone
significantly reduced IFNc production, (p,0.01), although not to
the same extent as MPA and cortisol. IFNc concentrations were
not detectable in supernatants of PBMCs cultured in the presence
of the three hormones alone; while high levels of IFNc were
detected in the supernatants of PBMCs stimulated with the
mitogen PHA at both three (2375.48 pg/ml) and six days
(2476.73 pg/ml) (data not shown).
MPA, cortisol and progesterone differentially alter BCG-
induced secretion of cytokines in vitro
To determine the effect of cortisol, MPA and progesterone on
the secretion of other cytokines and chemokines in addition to
IFNc we used a Multiplex bead array assay. Our Multiplex data
revealed that the cytokines, chemokines and growth factors could
be divided into four distinct groups; those not changed by any of
the three hormones, those influenced by MPA and cortisol, those
influenced by MPA, cortisol and progesterone and those
influenced by cortisol only (table 1).
Three days post stimulation (n=11) the secretion of IL-10, EGF
and MCP-1 was not influenced by any of the hormones (data not
shown) whereas the secretion of IL-1a, IL-17 (figure 2 A and B),
IL-6, IL-2, IFNc, IL-5, GM-CSF and MIP-1b (table 1) was
significantly down-regulated by cortisol and MPA, but not
progesterone. Cortisol and MPA were also able to up-regulate
the production of IP-10 (figure 2 F) and G-CSF (table 1).
Progesterone inhibited the day three secretion of TNFa, IL-12p40,
IL-13 (figure 2 C-E) and IL-1ra (table 1), but to a lesser extent than
cortisol and MPA. Secretion of IL-1b and IL-8 was influenced by
cortisol only with both cytokines being down-regulated (table 1).
At day six (n=10), cytokines, chemokines and growth factors
could be divided into the same groups described for day three but
the distribution of cytokines within those groups was different. The
production of EGF, MCP-1, MIP-1b and IL-8 was not influenced
by the three hormones (data not shown). Levels of IL-1a, IL-1b,
IL-1ra, IL-6, TNFa, IL-5 and IL-17 (table 1) were down-regulated
by cortisol and MPA, but not progesterone, while IP-10 was
significantly up-regulated by cortisol and MPA (table 1). The
secretion of IL-12p40, IFNc, IL-13, sCD40L, GM-CSF was
significantly down-regulated by cortisol, MPA and progesterone;
whereas the production of G-CSF was significantly up-regulated
by the three hormones (table 1). These results indicate that at high
concentrations the synthetic progestin, MPA, mimics cortisol
instead of progesterone.
MPA, cortisol and progesterone differentially affect BCG-
induced cytokine production at pharmacological doses
We found that high concentrations (10 mM) of cortisol, MPA
and progesterone alter the secretion of cytokines produced by
PBMCs stimulated with BCG. As such high concentrations do not
reflect the pharmacological doses of MPA in contraceptive users,
dose response experiments were done to determine the effect of the
hormones on cytokine production at pharmacological concentra-
tions in vitro. Serum concentrations of MPA in contraceptive users
have been shown to range from 0.1 nM to 60 nM [22–25].
PBMCs of non-contraceptive users (n=4) were stimulated with
increasing concentrations of the three hormones and the relative
potency and efficacy of each steroid to inhibit or enhance the
BCG-induced secretion of the cytokines were determined (Table 2).
Figure 1. The effect of cortisol (cort), medroxyprogesterone
acetate (MPA) and progesterone (prog) on BCG induced IFNc
responses of PBMCs of non-contraceptive users. PBMCs of
household contacts of active TB cases were cultured with AIM-V
medium only (unstimulated controls), incubated with cort, MPA and
prog alone (not shown) or M.bovis BCG (56105CFU/ml) in the presence
and absence of the three hormones (final concentration of 1025M).
IFNc concentrations were measured by ELISA in the supernatants
collected three days post stimulation (n=29). Differences among
stimulations were determined by one-way ANOVA with the Bonferroni
Post-Hoc test and the data represented as mean 6 SEM. * p,0.05,
**p,0.01, ***p,0.001 compared to BCG alone.
MPA Alters BCG-Induced Cytokine Secretion
PLoS ONE | www.plosone.org3September 2011 | Volume 6 | Issue 9 | e24639
At nanomolar concentrations MPA, cortisol and progesterone did
not alter the BCG-induced secretion of IFNc as determined by
ELISA (data not shown), however for TNFa, IL-1ra, IL-13 and
GM-CSF MPA mimicked the effect of cortisol (figure 3; pharma-
cological concentrations of MPA are indicated by the dotted lines).
MPA inhibited the production of TNFa (IC50cort=5.061028M;
IC50MPA=4.061028M), IL-1ra (IC50cort=5.961028M; IC50MPA
=761028M), IL-13 (IC50cort=6.961029M; IC50MPA=8.86
1029M) and GM-CSF<(IC50cort=2.661028M; IC50MPA=1.5
61028M) to the same extent as cortisol (Table 2). The effect of
MPA and cortisol was different to that of progesterone at similar
concentrations (p=0.0571). Compared to progesterone, MPA had
a 33-fold higher potency in inhibiting TNFa; a 6-fold higher
potency in inhibiting IL-1ra, a 53-fold higher potency in inhibiting
IL-13 and an 87-fold higher potency in inhibiting GM-CSF.
PBMCs of MPA users and non-contraceptive users
produce similar levels of IFNc in response to BCG
Saturating as well as pharmacological concentrations of MPA
altered BCG-induced cytokine production of PBMCs of non-
contraceptive users. Our data indicate that MPA mimics the effect
of cortisol rather than its intended analog, progesterone, which
could have major implications on immune function in MPA users.
Therefore we investigated whether the immune response to BCG
differs between MPA users and women not using contraceptives.
After culturing PBMCs of MPA users and controls with BCG (in
the absence of hormone) for three (controls n=29, MPA n=8)
and six (controls n=35, MPA n=15) days we found that MPA
users did not produce lower levels of IFNc at any of the two time
points (data not shown), which is consistent with our in vitro
findings where MPA only inhibited IFNc production at high but
not at concentrations equivalent to serum levels in MPA users.
PBMCs of MPA users produce lower levels of IL-1a,
IL-12p40, IL-10, IL-13 and G-CSF compared to PBMCs of
Since the BCG-induced secretion of other cytokines such as
TNFa, IL-1ra, IL-13 and GM-CSF was affected by MPA in vitro
at pharmacological concentrations; we investigated whether the
levels of these and other cytokines differ in PBMCs of MPA users.
Qlucore Omics analysis software was used to generate a heat map
of the 21-plex cytokine data from the supernatants of PBMCs of
Table 1. The effect of cortisol, MPA and progesterone on cytokine production three and six days post stimulation of PBMCs with
BCG BCG+ +Cort BCG+ +MPABCG+ +Prog
Three days post stimulation
IL-6 8117.261146.1 4454.16.6 ***5910.661160.2 *7736.161266.0
IL-2 42.2611.3 11.6612.1 *7.9611.7 **24.4614.8
778.66171.132.36188.4 **67.86179.2 **644.6649.0
IL-5 12.363.0 0.263.3 **0.363.1 **6.164.4
GM-CSF295.7667.3 78.5669.6 * 135.9668.5 *182.9677.5
5253.461359.8 3487.861380.0 **3910.161370.2 * 4748.861447.7
G-CSF597.96307.5 1285.66311.8 *** 1146.96309.7 *** 897.66326.5
231.16367.4 *** 365.1 6 441.3 *
740.86186.7 304.76191.8 ***605.66189.3596.96208.8
IL-8 7863.261121.26772.461137.0 * 7328.261129.3 8307.5 6 1190.1
Six days post stimulation
676.5699.287.36103.8 ***169.16103.8 ***432.76131.7
476.46113.598.2 6 117.3 ** 227.36117.3 *300.16138.8
255.76130.3 ***66.26130.3 *** 487.76166.5
IL-6 6744.761166.7 3550.761179.8 ***4956.3 6 1179.8 **7635.061250.7
2557.36563.3 316.86589.9 **482.66589.9 ** 1192.46752.1
IL-5 17.164.1 1.864.3 ** 1.8 6 4.3 **10.165.3
IL-1746.068.95.669.3 ***8.269.3 **31.3 6 11.8
IP-10 2472.662052.49565.762108.9 *** 8672.562108.9 **4776.762423.8
IL-12p40205.5 6 30.924.4632.2 *** 42.5632.2 ***114.3640.1 *
2890.16302.395.46316.6 ***318.96316.6 ***1805.1 6 403.8 *
IL-13253.1629.64.0 6 31.1 *** 6.9631.1 *** 88.0641.2 **
29.9 6 7.8 ***
GM-CSF455.8672.968.6675.6 ***110.7 6 75.6 ***217.3691.2 *
G-CSF799.26382.9 1547.56387.4 ***1371.86387.4 ** 1292.96411.2 *
Cytokine levels of the unstimulated supernatants were subtracted from the BCG 6 hormone stimulated supernatants as measured by multiplex bead array assay. Data
were analyzed by mixed model repeated measures ANOVA (LS means 6 SEM).
***p,0.001 compared to BCG alone.
MPA Alters BCG-Induced Cytokine Secretion
PLoS ONE | www.plosone.org4September 2011 | Volume 6 | Issue 9 | e24639
controls (n=11) and MPA (n=8) users stimulated with BCG for
three days (figure 4). This unbiased analysis approach revealed
that MPA users clearly clustered separately from control
participants (with exception of one outlier in each group)
indicating that the production of various cytokines is affected
by MPA use. Neither the outlier in the control nor the outlier in
the MPA group showed any obvious differences in demographics,
clinical aspects or medicine use compared to the other study
participants within the same group. Statistical analysis of the data
revealed that PBMCs of MPA users produce significantly lower
levels of IL-1a, IL-12p40, IL-10, IL-13 (figure 5 A-D) three days
post culture (MPA n=8 and controls n=11), as well as lower
levels of IL-1a, IL-12p40 and G-CSF six days post culture (MPA
n=10 and controls n=10; data not shown). There was no
significant difference in the age of MPA users and controls,
therefore the differences seen in immunological measurements
cannot be related to age. Due to the small sample size we were
not able to perform accurate correlation analyses of the duration
of MPA use with the immunological readouts of the luminex
analysis. Since macrophages play a prominent role in either
producing these cytokines directly or activating other cells to
secrete these cytokines, it raised the question whether the
Figure 2. The effect of cortisol, MPA and progesterone on cytokine production three days post stimulation of PBMCs with BCG.
Cortisol, MPA but not progesterone altered the expression of IL-1a (A) and IL-17 (B) three days post stimulation of PBMCs from household contacts
(non-contraceptive users, n=11). The BCG-induced expression of TNFa (C), IL-12p40 (D) and IL-13 (E) was altered by cortisol, MPA and to a lesser
extent progesterone. Cortisol and MPA also up-regulated the expression of IP-10 (F). These cytokine levels were measured by multiplex bead array
assays. Cytokine levels of the unstimulated supernatants were subtracted from the BCG 6 hormone stimulated supernatants. Data were analyzed by
mixed model repeated measures ANOVA (LS means, 95%CI). * p,0.05, **p,0.01, ***p,0.001 compared to BCG alone.
MPA Alters BCG-Induced Cytokine Secretion
PLoS ONE | www.plosone.org5 September 2011 | Volume 6 | Issue 9 | e24639
observed effect was due to lower numbers of circulating
monocytes in MPA users.
MPA users have fewer circulating monocytes, but no
change in delayed type hypersensitivity responses
Full blood counts of 228 HHCs (65 MPA users and 163
Controls) were analyzed and no differences in the white and red
cell counts between the two groups of women were observed (data
not shown). MPA users had significantly higher levels of
haemoglobin and mean corpuscular haemoglobin concentration
(data not shown) which is most likely due to amenorrhea
frequently experienced by women on MPA . No differences
in the numbers of circulating lymphocytes (figure 6 A), basophils,
neutrophils and eosinophils were observed (data not shown).
Interestingly however, MPA users had significantly lower levels of
circulating monocytes (p=0.0007; figure 6 B), which supports our
hypothesis that at least in part a reduced production of IL-1, IL-
12p40 and IL-10 can be attributed to fewer macrophages in MPA
users. No significant differences were detected in the TST
induration of these women when MPA users (19.6 6 6.4 mm)
were compared to controls (21.066.8 mm).
In our study population 60% of women using contraceptives
choose MPA, which unlike other contraceptives possesses selective
glucocorticoid activity and can alter GR regulated genes .
Therefore it is extremely important to investigate the potential
immune modulatory effects of this synthetic progestin.
This study aimed to determine whether MPA would alter the
BCG-specific memory responses of PBMCs of recently exposed
HHCs of TB patients. At high concentrations MPA generally
mimicked cortisol and unlike progesterone suppressed the
production of several cytokines including IL-1a, IL-6 and IL-17.
Even at pharmacological concentrations MPA and cortisol but not
progesterone, inhibited the production of TNFa, IL-1ra, IL-13
and GM-CSF. Our in vitro findings therefore suggest that in
general MPA mimics the effect of cortisol rather than its intended
GCs affect many aspects of immune cell function and are well
known to inhibit the production of Th1 cytokines and cause a shift
from a Th1 to a Th2 cytokine response pattern . However the
inhibitory effect of GCs goes beyond the traditional pro- and anti-
inflammatory division and it has been shown that GCs inhibit the
production of IL-2, IFNc, IL-1b, IL-8, TNFa, IL-4, IL-5, IL-6,
IL-10, IL-13, IL-17 as well as G-CSF . Furthermore GCs are
able to decrease the activity of the Th1 transcription factor T-box
expressed in T cell (T-bet)  as well as the Th2 transcription
factor GATA-binding protein 3 (GATA-3) .
As part of the classical ‘‘genomic’’ GC signaling pathway, GCs
mediate their effects by binding to cytosolic GRs which then
translocate to the nucleus and bind to the promoter regions of
genes regulated by the GR . The anti-inflammatory effects of
GCs are mainly mediated by interaction between the ligand
activated GR and transcription factors such as NF-kB and AP-1
. Similarly to GCs, MPA has been shown to inhibit the
secretion of IL-2 and IL-6 in T cells almost to the same extent as
the glucocorticoid dexamethasone, whereas progesterone only
slightly inhibited the secretion of these cytokines . During this
study we have demonstrated that as with GCs the inhibitory effect
of MPA is not restricted to Th1 cytokines alone, but MPA can also
impact on Th2 as well as Th17 cytokine responses in vitro.
Table 2. Potency and efficacy of cortisol, MPA and progesterone as determined by the dose response curves.
IC50values (potency) Efficacy (% inhibition)
CortisolMPA ProgesteroneCortisol MPAProgesterone
2.7E-08 3.5E-08NA 79.9630.4 68.1630.820.7611.6
2.5E-08 9.0E-09 3.3E-0874.8631.946.9635.4 7.762.6
IL-1ra 5.9E-08# 3.7E-08# 2.4E-0790.265.9 71.969.5 24.2617.1
IL-8 7.1E-091.2E-05 3.2E-10 2.067.5
5.0E-08# 4.0E-08# 1.3E-0681.5615.265.1617.522.2614.8
2.6E-07 2.5E-078.7E-01 92.263.346.96.522.4617.0
IL-12p404.7E-073.0E-07 4.6E-06100.060.0 50.2622.243.6620.3
IL-59.3E-09 1.0E-08 8.5E-0778.9626.6 76.0631.1 53.5613.5
IL-136.9E-09# 8.8E-09# 4.7E-07 98.461.397.461.972.7612.3
IL-17 3.9E-085.9E-08 1.9E-0688.866.268.166.1 44.569.9
GM-CSF2.6E-08# 1.5E-08# 1.3E-0689.667.966.1614.2 28.9619.7
EGF4.4E-091.8E-09 3.4E-07 29.6620.2 5.462.3
1.2E-083.0E-084.5E-08 41.4618.8 39.2612.3 39.764.5
EC50values (potency) Efficacy (% induction)
G-CSF 2.8E-08 3.5E-080 67.068.6126.96.36.1999.5
IP-10 3.4E-081.9E-08 9.6E-06* 82.3617.6 83.5615.5 87.760.3
IC50- concentration at which 50% of cytokine response is inhibited; EC50- concentration at which 50% of cytokine response is induced; # - MPA inhibited the
expression of TNFa, IL-1ra, IL-13 and GM-CSF with the same potency as cortisol, which was different to progesterone (p value=0.0571); * - Progesterone, but not MPA or
cortisol, induced the expression of IP-10 (p value=0.0571). NA - no activity for ligands was determined. Dose-response curves were generated using GraphPad Prism 5
(nonlinear regression curve; n=4). The efficacy was calculated by subtracting the percentage of cytokine expression of the highest concentration of hormone from the
BCG only stimulated (100%, no hormone present) response.
MPA Alters BCG-Induced Cytokine Secretion
PLoS ONE | www.plosone.org6September 2011 | Volume 6 | Issue 9 | e24639
Ex vivo, we saw that in response to BCG PBMCs from MPA
users produced significantly lower levels of IL1a, IL-12p40, IL-10,
IL-13 and G-CSF than PBMCs from non-contraceptive users as
determined by the Fisher LSD Post-Hoc test. IL-12 plays an
important role in the generation of protective immune responses
against TB as shown in murine models [34,35]. T cells rely on IL-
12 to differentiate into a Th1 phenotype and produce IFNc at the
site of infection . IL-12 is a heterodimeric cytokine consisting
of two subunits, p35 and p40 , and is required for optimal
differentiation and expansion of activated lymphocytes. It has been
suggested that the p40 subunit drives cellular responses  as p40
knock-out (KO) mice are more susceptible to M.tb than p35 KO
mice. IL-1 is one of the most important proinflammatory cytokines
and IL-1 type 1 receptor KO mice as well as IL-1a/b KO mice
infected with M.tb, have significantly larger granulomas with
neutrophilic infiltrates in the lungs compared to WT mice [39,40].
The IL-1 receptor KO mice were highly susceptible to M.tb
infection and had significantly more bacteria in their lungs, livers
and spleens . Susceptibility to M.tb was associated with a lack
of Th1 responses and impaired granuloma formation. IL-1 is
therefore crucial for inflammatory cell recruitment into granulo-
mas and host defense against M.tb. Lower levels of BCG-induced
IL-12p40 and IL-1a observed in MPA users in our study could
therefore suggest that they are more susceptible to M.tb and might
have a worse outcome of disease.
IL-10 regulates Th1 cell responses by inhibiting the production
of IFNc and IL-12 [41,42]. Mice deficient in IL-10 have an
enhanced ability to control M.tb infection as they have significantly
lower numbers of bacteria in the lungs and spleens . The
increase in protection observed in IL-10 deficient mice was
associated with enhanced Th1 responses at the site of disease;
however these mice succumb to the disease due to elevated IL-12
and IFNc levels which result in severe tissue pathology . The
increased mortality seen in mice during the absence of IL-10 is not
a result of the uncontrolled growth of the bacteria but rather that
of unregulated immune responses to the infection. A reduction in
IL-10 production will result in a decreased number of circulating
Th2 cells leading to less IL-13 being produced; therefore we
hypothesize that a reduction of IL-13 could be a result of an
indirect effect of the hormone. Lower levels of IL-10 and IL-13,
could result in more sever immunopathology in individuals
suffering from active TB and using MPA. Taken together it is
therefore likely that MPA use will not only have an effect on
susceptibility to TB but possibly also TB disease severity.
GC therapy not only inhibits the production of cytokines, but
has also been shown to reduce the circulating monocyte
Figure 3. Dose-response curves of cortisol, MPA and progesterone on BCG-induced cytokine expression. Culture supernatants of
PBMCs stimulated with a concentration of 56105CFU/ml BCG in the absence and presence of the three hormones were collected six days post
stimulation (n=4). TNFa, IL-1ra, IL-13 and GM-CSF (A-D) levels were measured using a Multiplex bead array assay. Hormone stimulated responses are
depicted as percentages of the BCG only (100%) responses. Dotted lines indicate the range of MPA serum concentrations measured in contraceptive
users. Dose-response curves were generated using GraphPad Prism 5 (nonlinear regression curve) and data are presented as mean 6 SEM. Data of
four study participants were analyzed to obtain IC50and EC50values.
MPA Alters BCG-Induced Cytokine Secretion
PLoS ONE | www.plosone.org7 September 2011 | Volume 6 | Issue 9 | e24639
Figure 4. Differential expression of cytokines in supernatants of BCG stimulated PBMCs of MPA users and control participants
three days post stimulation. In an unbiased analysis approach using the Qlucore Omics explorer software to generate a heat map, clustering of
the participants based on cytokine expression levels coincided with the two study groups (Controls n=11 and MPA users n=8). Over-expressed
cytokines are depicted in red and under-expressed cytokines in green.
Figure 5. Differential expression of IL-1a, IL-12p40, IL-10 and IL-13 in supernatants of BCG stimulated PBMCs of MPA users and
control participants. PBMCs of MPA users (n=8) and controls (n=11) were cultured as unstimulated controls and with M.bovis BCG. Culture
supernatants were collected after three days and cytokine responses were measured by Multiplex bead array assay. Data were analyzed by mixed
model repeated measures ANOVA with a Fisher LSD Post-Hoc test and are represented as LS means and 95%CI. The letter a, b and c indicate
statistical significance. Values with the same letters are not statistically significantly different from each other.
MPA Alters BCG-Induced Cytokine Secretion
PLoS ONE | www.plosone.org8 September 2011 | Volume 6 | Issue 9 | e24639
population in patients receiving GC therapy . Thus the
reduction in numbers of circulating monocytes in women using
MPA shown in this study could be attributed to the glucocorticoid
activity of this steroid. We could therefore speculate that at least in
part the inhibition of IL-12, IL-1a, IL-10 and IL-13 in MPA users
can be attributed to lower numbers of circulating monocytes. The
reduction of circulating monocytes together with reduced
production of IL-1 and IL-12 in response to mycobacterial
antigens could increase the susceptibility of MPA users to TB and
increase the risk of progression from latent to active disease. For
future studies with larger sample numbers it would be particularly
interesting to correlate the immunological measurements of MPA
users with duration of MPA use.
Our research study is the first to show that MPA impacts on
infections outside the genital tract due to a systemic effect on
immune function. These findings warrant further investigations
into the effect of MPA on susceptibility to TB, disease severity and
treatment outcome. Furthermore MPA use could change the
efficacy of new BCG-based vaccines and especially prime-boost
vaccine strategies which may be administered to adult or
adolescent women in the future.
The authors would like to thank Ms Kim Stanley for database support.
Conceived and designed the experiments: KR GW. Performed the
experiments: LK NDP. Analyzed the data: LK MK KR. Contributed
reagents/materials/analysis tools: GW PDvH MK KR LK. Wrote the
paper: LK KR. Patient database support: GFB AGL. Critical review of
manuscript and intellectual contributions: PDvH GFB NDP AGL GW
1. World Health Organization (2008) Global Health Observatory (GHO) TB
incidence Available: http://www.who.int/gho/mdg/diseases/tuberculosis/
situation_trends_incidence/en/index.html. Accessed 2010 Nov 19.
2. Biondi M, Zannino LG (1997) Psychological stress, neuroimmunomodulation,
and susceptibility to infectious diseases in animals and man: a review. Psychother
Psychosom 66: 3–26.
3. Howard AD, Zwilling BS (1999) Reactivation of tuberculosis is associated with a
shift from type 1 to type 2 cytokines. Clin Exp Immunol 115: 428–434.
4. Manabe YC, Kesavan AK, Lopez-Molina J, Hatem CL, Brooks M, et al. (2008)
The aerosol rabbit model of TB latency, reactivation and immune reconstitution
inflammatory syndrome. Tuberculosis(Edinb) 88: 187–196.
5. Scanga CA, Mohan VP, Joseph H, Yu K, Chan J, et al. (1999) Reactivation of
latent tuberculosis: variations on the Cornell murine model. Infect Immun 67:
6. Jick SS, Lieberman ES, Rahman MU, Choi HK (2006) Glucocorticoid use,
other associated factors, and the risk of tuberculosis. Arthritis Rheum 55: 19–26.
7. Mahuad C, Bay ML, Farroni MA, Bozza V, del Rey A, et al. (2004) Cortisol and
dehydroepiandrosterone affect the response of peripheral blood mononuclear
cells to mycobacterial antigens during tuberculosis. Scand J Immunol 60:
8. Kontula K, Paavonen T, Luukkainen T, Andersson LC (1983) Binding of
progestins to the glucocorticoid receptor. Correlation to their glucocorticoid-like
effects on in vitro functions of human mononuclear leukocytes. Biochem
Pharmacol 32: 1511–1518.
9. Koubovec D, Ronacher K, Stubsrud E, Louw A, Hapgood JP (2005) Synthetic
progestins used in HRT have different glucocorticoid agonist properties. Mol
Cell Endocrinol 242: 23–32.
10. Trunova N, Tsai L, Tung S, Schneider E, Harouse J, et al. (2006) Progestin-
based contraceptive suppresses cellular immune responses in SHIV-infected
rhesus macaques. Virology 352: 169–177.
11. Kaushic C, Ashkar AA, Reid LA, Rosenthal KL (2003) Progesterone increases
susceptibility and decreases immune responses to genital herpes infection. J Virol
12. Mostad SB, Kreiss JK, Ryncarz AJ, Mandaliya K, Chohan B, et al. (2000)
Cervical shedding of herpes simplex virus in human immunodeficiency virus-
infected women: effects of hormonal contraception, pregnancy, and vitamin A
deficiency. J Infect Dis 181: 58–63.
13. Mostad SB, Overbaugh J, DeVange DM, Welch MJ, Chohan B, et al. (1997)
Hormonal contraception, vitamin A deficiency, and other risk factors for
shedding of HIV-1 infected cells from the cervix and vagina. Lancet 350:
14. Overton ET, Shacham E, Singhatiraj E, Nurutdinova D (2008) Incidence of
sexually transmitted infections among HIV-infected women using depot
medroxyprogesterone acetate contraception. Contraception 78: 125–130.
15. Leclerc PM, Dubois-Colas N, Garenne M (2008) Hormonal contraception and
HIV prevalence in four African countries. Contraception 77: 371–376.
16. Myer L, Denny L, Wright TC, Kuhn L (2007) Prospective study of hormonal
contraception and women’s risk of HIV infection in South Africa. Int J Epidemiol
17. Pettifor A, Delany S, Kleinschmidt I, Miller WC, Atashili J, et al. (2009) Use of
injectable progestin contraception and risk of STI among South African women.
Contraception 80: 555–560.
18. Morrison CS, Chen PL, Kwok C, Richardson BA, Chipato T, et al. (2010)
Hormonal contraception and HIV acquisition: reanalysis using marginal
structural modeling. AIDS 24: 1778–1781.
19. Morrison CS, Chen PL, Nankya I, Rinaldi A, Van Der PB, et al. (2011)
Hormonal contraceptive use and HIV disease progression among women in
Uganda and Zimbabwe. J Acquir Immune Defic Syndr 57: 157–164.
20. Back DJ, Orme ML (1990) Pharmacokinetic drug interactions with oral
contraceptives. Clin Pharmacokinet 18: 472–484.
Figure 6. Differential cell counts of MPA users and control participants. Blood was drawn from each study participant (65 MPA users and
163 controls) and the counts determined by the National Health Laboratory Service. Statistical differences were determined by the non-parametric
Mann-Whitney test and represented as means and 95%CI.
MPA Alters BCG-Induced Cytokine Secretion
PLoS ONE | www.plosone.org9 September 2011 | Volume 6 | Issue 9 | e24639
21. Black GF, Fine PEM, Warndorff DK, Floyd S, Weir RE, et al. (2001) Download full-text
Relationship between IFN-gamma and skin test responsiveness to Mycobacte-
rium tuberculosis PPD in healthy, non-BCG-vaccinated young adults in
Northern Malawi. Int J Tuberc Lung Dis 5: 664–72.
22. Fang S, Sun D, Jiang H, Luo H (2004) Concentration changes if
medroxyprogesterone acetate in serum and milk in lactating women who used
depo geston. J Reprod Contraception 15: 157–162.
23. Jeppsson S, Gershagen S, Johansson ED, Rannevik G (1982) Plasma levels of
medroxyprogesterone acetate (MPA), sex-hormone binding globulin, gonadal
steroids, gonadotrophins and prolactin in women during long-term use of depo-
MPA (Depo-Provera) as a contraceptive agent. Acta Endocrinol(Copenh) 99:
24. Ortiz A, Hirol M, Stanczyk FZ, Goebelsmann U, Mishell DR (1977) Serum
medroxyprogesterone acetate (MPA) concentrations and ovarian function
following intramuscular injection of depo-MPA. J Clin Endocrinol Metab 44:
25. Shi L, Jung YJ, Tyagi S, Gennaro ML, North RJ (2003) Expression of Th1-
mediated immunity in mouse lungs induces a Mycobacterium tuberculosis
transcription pattern characteristic of nonreplicating persistence. Proc Natl Acad
Sci USA 100: 241–246.
26. Bigrigg A, Evans M, Gbolade B, Newton J, Pollard L, et al. (1999) Depo
Provera. Position paper on clinical use, effectiveness and side effects. Br J Fam
Plann 25: 69–76.
27. Elenkov IJ (2004) Glucocorticoids and the Th1/Th2 balance. Ann N Y Acad Sci
28. Liu Z, Yuan X, Luo Y, He Y, Jiang Y, et al. (2009) Evaluating the effects of
immunosuppressants on human immunity using cytokine profiles of whole
blood. Cytokine 45: 141–7.
29. Liberman AC, Refojo D, Druker J, Toscano M, Rein T, et al. (2007) The
activated glucocorticoid receptor inhibits the transcription factor T-bet by direct
protein-protein interaction. FASEB J 21: 1177–88.
30. Maneechotesuwan K, Yao X, Ito K, Jazrawi E, Usmani OS, et al. (2009)
Suppression of GATA-3 nuclear import and phosphorylation: a novel
mechanism of corticosteroid action in allergic disease. PLoSMed 6: e1000076.
31. Unterberger C, Staples KJ, Smallie T, Williams L, Foxwell B, et al. (2008) Role
of STAT3 in glucocorticoid-induced expression of the human IL-10 gene. Mol
Immunol 45: 3230–3237.
32. Smoak KA, Cidlowski JA (2004) Mechanisms of glucocorticoid receptor
signaling during inflammation. Mech Ageing Dev 125: 697–706.
33. Bamberger CM, Else T, Bamberger AM, Beil FU, Schulte HM (1999)
Dissociative glucocorticoid activity of medroxyprogesterone acetate in normal
human lymphocytes. J Clin Endocrinol Metab 84: 4055–4061.
34. Cooper AM, Roberts AD, Rhoades ER, Callahan JE, Getzy DM, et al. (1995)
The role of interleukin-12 in acquired immunity to Mycobacterium tuberculosis
infection. Immunology 84: 423–432.
35. Cooper AM, Magram J, Ferrante J, Orme IM (1997) Interleukin 12 (IL-12) is
crucial to the development of protective immunity in mice intravenously infected
with mycobacterium tuberculosis. J Exp Med 186: 39–45.
36. Manetti R, Parronchi P, Giudizi MG, Piccinni MP, Maggi E, et al. (1993)
Natural killer cell stimulatory factor (interleukin 12 [IL-12]) induces T helper
type 1 (Th1)-specific immune responses and inhibits the development of IL-4-
producing Th cells. J Exp Med 177: 1199–1204.
37. Kobayashi M, Fitz L, Ryan M, Hewick RM, Clark SC, et al. (1989)
Identification and purification of natural killer cell stimulatory factor (NKSF),
a cytokine with multiple biologic effects on human lymphocytes. J Exp Med 170:
38. Cooper AM, Kipnis A, Turner J, Magram J, Ferrante J, et al. (2002) Mice
lacking bioactive IL-12 can generate protective, antigen-specific cellular
responses to mycobacterial infection only if the IL-12 p40 subunit is present.
J Immunol 168: 1322–1327.
39. Juffermans NP, Florquin S, Camoglio L, Verbon A, Kolk AH, et al. (2000)
Interleukin-1 signaling is essential for host defense during murine pulmonary
tuberculosis. J Infect Dis 182: 902–908.
40. Sugawara I, Yamada H, Hua S, Mizuno S (2001) Role of interleukin (IL)-1 type
1 receptor in mycobacterial infection. Microbiol Immunol 45: 743–750.
41. D’Andrea A, ste-Amezaga M, Valiante NM, Ma X, Kubin M, et al. (1993)
Interleukin 10 (IL-10) inhibits human lymphocyte interferon gamma-production
by suppressing natural killer cell stimulatory factor/IL-12 synthesis in accessory
cells. J Exp Med 178: 1041–1048.
42. Powrie F, Coffman RL (1993) Cytokine regulation of T-cell function: potential
for therapeutic intervention. Immunol Today 14: 270–274.
43. Redford PS, Boonstra A, Read S, Pitt J, Graham C, et al. (2010) Enhanced
protection to Mycobacterium tuberculosis infection in IL-10-deficient mice is
accompanied by early and enhanced Th1 responses in the lung. Eur J Immunol
44. 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:
45. Heimbeck I, Hofer TP, Eder C, Wright AK, Frankenberger M, et al. (2010)
Standardized single-platform assay for human monocyte subpopulations: Lower
CD14+CD16++ monocytes in females. Cytometry A 77: 823–830.
MPA Alters BCG-Induced Cytokine Secretion
PLoS ONE | www.plosone.org10 September 2011 | Volume 6 | Issue 9 | e24639