Cystatin cures visceral leishmaniasis by NF-κB-mediated proinflammatory response through co-ordination of TLR/MyD88 signaling with p105-Tpl2-ERK pathway.

Abstract

Cystatin could completely cure experimental visceral leishmaniasis by switching the differentiation of Th2 cells to Th1 type, as well as upregulating NO, and activation of NF-κB played a major role in these processes. Analysis of upstream signaling events revealed that TLR 2/4-mediated MyD88-dependent participation of IL-1R-activated kinase (IRAK)1, TNF receptor-associated factor (TRAF)6 and TGFβ-activated kinase (TAK)1 is essential to induce cystatin-mediated IκB kinase (IKK)/NF-κB activation in macrophages. Cystatin plus IFN-γ activated the IKK complex to induce phosphorylation-mediated degradation of p105, the physiological partner and inhibitor of the MEK kinase, tumor progression locus 2 (Tpl-2). Consequently, Tpl-2 was liberated from p105, thereby stimulating activation of the MEK/ERK MAPK cascade. Cystatin plus IFN-γ-induced IKK-β post-transcriptionally modified p65/RelA subunit of NF-κB by dual phosphorylation in infected phagocytic cells. IKK induced the phosphorylation of p65 directly on Ser-536 residue whereas phosphorylation on Ser 276 residue was by sequential activation of Tpl-2/MEK/ERK/MSK1. Collectively, the present study indicates that cystatin plus IFN-γ-induced MyD88 signaling may bifurcate at the level of IKK, leading to a divergent pathway regulating NF-κB activation by IκBα phosphorylation and by p65 transactivation through Tpl-2/MEK/ERK/MSK1.

Full text

Page 1

Cystatin cures visceral leishmaniasis by NF-jB-
mediated proinflammatory response through
co-ordination of TLR/MyD88 signaling
with p105-Tpl2-ERK pathway
Susanta Kar1, Anindita Ukil2and Pijush K. Das1
1Molecular Cell Biology Laboratory, Infectious Diseases and Immunology Division, Indian
Institute of Chemical Biology, Kolkata, India
2Department of Biochemistry, Calcutta University, Kolkata, India
Cystatin could completely cure experimental visceral leishmaniasis by switching the
differentiation of Th2 cells to Th1 type, as well as upregulating NO, and activation of NF-jB
played a major role in these processes. Analysis of upstream signaling events revealed
that TLR 2/4-mediated MyD88-dependent participation of IL-1R-activated kinase (IRAK)1,
TNF receptor-associated factor (TRAF)6 and TGFb-activated kinase (TAK)1 is essential to
induce cystatin-mediated IjB kinase (IKK)/NF-jB activation in macrophages. Cystatin plus
IFN-c activated the IKK complex to induce phosphorylation-mediated degradation of p105,
the physiological partner and inhibitor of the MEK kinase, tumor progression locus 2 (Tpl-
2). Consequently, Tpl-2 was liberated from p105, thereby stimulating activation of the
MEK/ERK MAPK cascade. Cystatin plus IFN-c-induced IKK-b post-transcriptionally modi-
fied p65/RelA subunit of NF-jB by dual phosphorylation in infected phagocytic cells. IKK
induced the phosphorylation of p65 directly on Ser-536 residue whereas phosphorylation
on Ser 276 residue was by sequential activation of Tpl-2/MEK/ERK/MSK1. Collectively, the
present study indicates that cystatin plus IFN-c-induced MyD88 signaling may bifurcate at
the level of IKK, leading to a divergent pathway regulating NF-jB activation by IjBa
phosphorylation and by p65 transactivation through Tpl-2/MEK/ERK/MSK1.
Key words: Cystatin.Leishmaniasis.Macrophage.NF-kB.Tumor progression locus 2
Supporting Information available online
Introduction
Innate immunity coordinates the inflammatory response to
pathogens and the essential role of TLR in this is widely
recognized. Following infection, components of exogenous
pathogens bind to TLR on host cell, triggering the activation of
signaling pathways that stimulate the production of inflammatory
mediators such as pro-inflammatory cytokines, IFN and chemo-
kines. All these mediators are released into the circulation to elicit
effective immune response against invading microorganisms. All
TLR activate a common signaling pathway that culminates
in the activation of transcription factor NF-kB as well as MAPK
(ERK1/2, p38 and JNK) [1]. Following the recognition of
pathogen-associated molecular pattern, all of the TLR except
TLR3 recruit a signaling complex to the receptors that includes
an adaptor protein, MyD88 and two protein kinases, termed
Correspondence: Dr. Pijush K. Das
e-mail: pijushdas@iicb.res.in
& 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eji-journal.eu
DOI 10.1002/eji.201040533Eur. J. Immunol. 2011. 41: 116–127Susanta Kar et al.
116

Page 2

IL-1R-activated kinase (IRAK)1 and IRAK4 [2]. IRAK1 is then
phosphorylated by IRAK4 as well as by autophosphorylation
[3, 4], resulting in recruitment of TNF receptor-associated factor
(TRAF)6 [5]. Phosphorylated IRAK1 and TRAF6 are thought to
dissociate from the receptors [6], which is followed by TRAF6
autoubiquitination with lys-63-linked polyubiquitin chains [7].
TAB2 and TAB3, the regulatory subunits of the TGFb-activated
kinase (TAK) 1 complex, specifically interact with ubiquitinated
TRAF6, which activates TAK1 [8]. Activated TAK1 is then thought
to phosphorylate and activate downstream IkB kinase (IKK)
complex as well as MAPK kinase 6 to activate NF-kB and MAPK
signaling pathway. Activation of MAPK in macrophages requires
another serine/threonine kinase, tumor progression locus 2 (Tpl-
2) [9], also known as Cot/MAP3K8. Tpl-2 is critical for host
defense and appropriate cytokine response against pathogens like
Toxoplasma gondii and Listeria monocytogenes [10, 11]. It
functions as a MAPK kinase kinase, which phosphorylates and
activates MEK, the upstream activator of ERK1/2 kinase [12].
Tpl-2 is critically involved in linking TLR to TNF-a production
through its activation of ERK1/2 and also stimulates MEK
activation in presence of TNF-a and CD40 ligand, suggesting its
important role in both innate and adaptive immune response
[13]. In unstimulated macrophages, Tpl-2 is complexed with NF-
kB1 (p105), an IkB family member, which blocks its MEK kinase
activity [14, 15]. Stimulation of macrophages with LPS facilitates
the release of Tpl-2 from p105 and thereby activates its kinase
activity, which has been shown to occur as a consequence of p105
phosphorylation and proteolysis by the proteasome [14]. p105
phosphorylation is triggered by IKK complex, which is also critical
for regulating the activation of NF-kB transcription factors
through phosphorylation of IkB and p65/RelA subunit [15, 16].
Leishmaniases are spectrum of vector-borne parasitic diseases
that are a major international public health problem, affecting
the lives of millions of people worldwide with significant
morbidity and mortality [17]. Gene knock-out studies in mice
revealed an essential role of TLR signaling in the immune
responses against Leishmania parasites [18]. For instance,
MyD88-dependent pathway is required for the secretion of IL-1a
by mouse peritoneal macrophages following infection with
L. major [19] and MyD88-deficient mice developed progressive
lesion with dominant Th2 response [20]. In vivo studies in mice
revealed an important role for TLR4 in the control of L. major
infection, possibly through the early induction of iNOS [21].
TLR9 is involved in NK-cell activation and pro-inflammatory
cytokine synthesis in animal models of visceral (L. infantum) and
cutaneous (L. major and L. braziliensis) leishmaniasis [22, 23],
while Leishmania lypophosphoglycan, an abundant molecule on
the surface of promastigotes, signals via TLR 2 [24]. Moreover,
silencing of IRAK1, MyD88, TLR2 or TLR3 significantly atte-
nuated the secretion of nitric oxide and TNF-a following infection
of IFN-g-primed macrophages by L. donovani promastigotes [25].
In addition, a recent study suggests that TLR2 and MyD88
appeared to have a critical role in both recognition as well as
modulation of immune response against Leishmania parasites
[26]. Our earlier work showed that cystatin, a natural cysteine
protease inhibitor, in synergy with IFN-g, could completely cure
visceral leishmaniasis in BALB/c mice by inducing proin-
flammatorycytokinesynthesis
Furthermore, this induction was found to be mediated by ERK 1/
2/mitogen- and stress-activated protein kinase 1 (MSK1) leading
to the activation of NF-kB-dependent gene expression [28]. In the
present study, we sought to determine the detailed upstream
signaling events involved in cystatin-induced activation of NF-kB,
which is essential for its anti-leishmanial effector response.
and NOgeneration [27].
Results
Activation of NF-jB is essential for cystatin plus IFN-c-
mediated anti-leishmanial response
Cystatin could completely cure experimental visceral leishmania-
sis and this was shown to be associated with strong upregulation
of Th1 cytokine and iNOS [27]. In the in vitro situation, a sub-
optimal dose of IFN-g was required for cystatin-induced proin-
flammatory cytokine response and iNOS expression whereas IFN-g
was not a prerequisite in vivo. Because iNOS and many
proinflammatory gene expressions are regulated by NF-kB, the
effect of cystatin plus IFN-g on NF-kB activation was assessed in
infected macrophages using EMSA. DNA binding activity of NF-kB
was found to be markedly enhanced in infected BM-derived
macrophages (BMM) following cystatin plus IFN-g treatment,
being maximum at 4h (3.8-fold) and still significant after 6h (1.9-
fold) (Fig. 1A). For NF-kB nuclear translocation, cytoplasmic IkBa
must be phosphorylated, ubiquitinated and degraded. Because
IkBa is phosphorylated by the IKK multiprotein complex, the effect
of cystatin on the status of intrinsic cellular IKK activation was
assessed. Accordingly, L. donovani-infected macrophages were
treated with cystatin plus IFN-g; IKKb was immunoprecipitated
and tested for the ability to phosphorylate GST-IkBa substrate. As
shown in Fig. 1B, cystatin plus IFN-g strongly induced IKKb
activity in infected BMM. In order to further ascertain the role of
NF-kB in the modulation of disease progression, anti-leishmanial
activity of cystatin plus IFN-g was evaluated in the presence of
BAY 11-7082 or BAY 11-7085, chemical compounds that block
NF-kB expression by inhibiting IkB phosphorylation. In the in vitro
situation of amastigote multiplication within macrophages, BAY-
11-7082 (5mM) could substantially reverse the parasite killing
(78% reduction) by cystatin (Fig. 1C). In the in vivo situation of
L. donovani-infected BALB/c mice, administration of cystatin
(5mg/kg/day) plus IFN-g (5?105
consecutive days after 15 days of infection completely suppressed
the parasite burden in the spleen cells of mice (Fig. 1D). However,
a dose of 5mg/kg/day of BAY-11-7085 given three times weekly
for 4wk starting at 15 days after infection greatly reduced
cystatin-mediated protection (74% reduction in parasite killing).
In vivo administration of BAY-11-7085 could also result in marked
reduction of cystatin-mediated increase in TNF-a synthesis (77%
reduction) (Fig. 1E) as well as iNOS expression (4.3-fold
reduction) in infected mice (Fig. 1F). Taken together, these
U/kg/day) given for 4
Eur. J. Immunol. 2011. 41: 116–127
Immunity to infection
117
& 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eji-journal.eu

Page 3

results strongly suggest that the therapeutic effect of cystatin may
be attributed to the activation of IKK-NF-kB pathway.
Cystatin plus IFN-c-induced Tpl-2/MEK/ERK pathway
Apart from activation of NF-kB, IKK is also known to induce
activation of Tpl-2, a serine-threonine kinase, which plays an
important role in both innate and adaptive immune response
[13, 14]. Since cystatin induced activation of IKK, we, therefore,
examined whether cystatin plus IFN-g could activate Tpl-2/MEK/
ERK pathway. Treatment of infected BMM with cystatin plus IFN-g
(0?60min) caused a marked increase in Tpl-2 kinase activity
reaching a maximum at 60min (Fig. 2A). The densitometric
analysis in the right-hand panel of immunoprecipitated Tpl-2
(consisting of p58 and p52 isoforms) demonstrated the disappear-
ance of p58 isoform at later time points (60 and 90min) (Fig. 2A)
suggesting its phosphorylation-mediated proteasomal degradation.
Figure 1. Involvement of NF-kB in cystatin plus IFN-g-mediated anti-leishmanial response. (A) BMM were infected with L. donovani promastigotes
(L. d) (cell:parasite ratio, 1:10) for 4h. Noningested promastigotes were removed, and macrophages were cultured for another 20h. L. donovani-
infected (24h) BMM were further treated with cystatin (0.5mM) plus IFN-g (100U/mL) for 0–6h and EMSA of NF-kB was performed with nuclear
extract. Extracts were also run with cold competitor oligonucleotides. Bands were analyzed densitometrically, and fold changes are indicated at
the bottom. (B) Whole-cell extracts from L. donovani-infected (24h) and cystatin plus IFN-g-treated (1h) BMM were immunoprecipitated with Ab
against IKKb. IKKb was assayed using GST-IkBa as substrate, and GST-phosphorylated IkBa was visualized by autoradiography. Relative amount of
IKKb in the whole-cell extracts was determined by Western blots. (C) Infected (24h) and cystatin plus IFN-g-treated (24h) BMM were pretreated
with BAY11-7085 (1 and 5mM) for 1h. Intracellular parasite number was determined by Giemsa staining. (D) Mice were challenged with 107
promastigotes, and after 15 days of infection, they were treated with cystatin plus IFN-g daily for 4 consecutive days. Spleen parasite burdens were
determined weekly after infection and were expressed as the mean of log10LDU. SD for six animals per group. BAY11-7085 (5mg/kg/day) was used
along with cystatin plus IFN-g in a separate set of experiments. (E and F) Infected mice were treated either with cystatin plus IFN-g or with cystatin
plus IFN-g plus BAY11-7085 as indicated in (D). ELISA was performed for the level of TNF-a (E) and real-time PCR analysis was performed for
the expression of iNOS (F) in the splenocytes. The cultures were set in triplicate and the experiments were done a minimum of three times.
Animal experiments were done with six animals per group and the results are representatives of four independent experiments. Data represent
the mean1SD.??p o 0.01,???p o 0.001; Student’s t-test.
Eur. J. Immunol. 2011. 41: 116–127Susanta Kar et al.
118
& 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eji-journal.eu

Page 4

Proteasomal degradation of p58 isoform was further confirmed by
the use of MG 132, a proteasomal inhibitor, which almost
completely abrogated degradation of p58 as well as activation of
Tpl-2 in cystatin plus IFN-g-stimulated BMM (data not shown). This
was further reflected in the activation of downstream kinases, MEK
and ERK1/2. A marked phosphorylation of MEK and ERK1/2 were
found in infected cells following cystatin plus IFN-g treatment,
which was significantly abrogated (2.3- and 2.2-fold for MEK and
ERK1/2, respectively) in cells treated with Tpl-2-specific siRNA
(Fig. 2B). To further determine whether cystatin plus IFN-g induced
the dissociation of Tpl-2 from p105, cell lysates were pre-cleared of
p105 by immunodepletion and then subjected to Western blotting
with anti-Tpl-2 antibody. As depicted in Fig. 2C, very little p105-free
Tpl-2 was detected in infected cells. However, cystatin plus IFN-g
treatment induced the appearance of Tpl-2 in the p105-depleted
lysate of infected BMM (Fig. 2C) suggesting a release of Tpl-2 from
its inhibitor NF-kB p105. Moreover, the appearance of Tpl-2 in the
p105-depleted lysate peaked at 60min, the same time point at
which the Tpl-2 MEK kinase activity was maximum. Tpl-2 activation
is associated with p105 phosphorylation followed by its proteolysis.
We found that treatment of infected BMM with cystatin plus IFN-g
induced phosphorylation of p105 on serine 927 followed by its rapid
disappearance (Fig. 2D). The kinetics of p105 phosphorylation and
degradation was correlated with Tpl-2 release from p105. To further
address whether IKK is required for p105 phosphorylation as well as
activation of Tpl-2, BMM were treated with the IKK inhibitor BAY-
11-7082 prior to infection. Western blotting of cell lysates revealed
that this inhibitor blocked cystatin plus IFN-g-induced p105
phosphorylation and degradation (Fig. 2D). BAY-11-7082 also
blocked cystatin plus IFN-g-induced release of Tpl-2 from p105
(Fig. 2C) and Tpl-2 MEK kinase activity (Fig. 2F) in infected cells.
Moreover, cystatin plus IFN-g-mediated induction of MEK and ERK
phosphorylation were also significantly reduced by BAY-11-7082
treatment (Fig. 2E). Collectively, these data suggest that IKK activity
is required for cystatin plus IFN-g-mediated activation of Tpl-2/
MEK/ERK signaling pathway.
Cystatin plus IFN-c induced dual phosphorylation of
p65 by IKK
Phosphorylation of p65 at multiple residues has been implicated in
p65 nuclear translocation and transcriptional activation of NF-kB
[15, 16]. To investigate the possibility that IKK regulates NF-kB
by a mechanism involving p65 phosphorylation, infected BMM
were treated with cystatin plus IFN-g for different time periods
(0?3h). Immunoblot analysis showed a strong induction of p65
phosphorylation at both Ser-536 and Ser-276 residues (Fig. 3A),
Figure 2. Effect of cystatin plus IFN-g on Tpl-2/MEK/ERK pathway. (A) Infected BMM were treated with cystatin plus IFN-g for the indicated times.
Tpl-2 was immunoprecipitated and its MEK kinase activity was determined by coupled MEK/ERK kinase assay. Phosphorylated MBP substrate was
visualized by autoradiography and the levels of immunoprecipitated Tpl-2 were determined by Western blotting to normalize the kinase assay.
(B) RAW 264.7 cells were transfected (24h) with either control or Tpl-2 siRNA, infected with L. donovani promastigotes (24h) and further stimulated
with cystatin plus IFN-g for 4h. The levels of p-MEK and p-ERK1/2 were measured by Western blotting. (C) L. donovani-infected BMM were treated with
cystatin plus IFN-g either alone or in combination with BAY11-7082 (5mM) (pretreated for 1h) for the indicated times. p105-cleared cell lysates were
subjected to Western blotting for Tpl-2 and p105. (D) Infected BMM (24h) were treated with cystatin plus IFN-g and BAY11-7082 as indicated in
(C). Levels of total and phosphorylated p105 were measured by immunobloting. (E) Infected BMM (24h) were treated with cystatin plus IFN-g either
alone or in combination with BAY11-7082 (1 and 5mM) (pretreated for 1h) for 4h. Levels of p-MEK and p-ERK1/2 were measured by Western blotting.
(F) Infected BMM (24h) were treated with cystatin plus IFN-g either alone or in combination with BAY11-7082 (1 and 5mM) (pretreated for 1h) for 1h.
Tpl-2 was immunoprecipitated from total-cell lysates and then assayed for MEK kinase activity as indicated in (A). Densitometries are shown as bar
graphs on the right-hand side of each panel. Results are expressed as mean1SD, n53. The results are representative of three separate experiments.
Eur. J. Immunol. 2011. 41: 116–127
Immunity to infection
119
& 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eji-journal.eu

Page 5

which was maximum at 2h and almost completely abrogated in the
presence of IKK inhibitor BAY-11-7082, suggesting the involvement
of IKK in cystatin plus IFN-g-induced dual phosphorylation of p65
(Fig. 3B). The decrease in p65 phosphorylation was not due to p65
degradation because the level of p65 remained constant (Fig. 3B).
To further ascertain whether dual p65 phosphorylation was carried
out by IKK directly or by some intermediate kinase, BMM were
pretreated with apigenin, an inhibitor of ERK1/2 prior to cystatin
plus IFN-g stimulation. Interestingly, cell exposure to apigenin
markedly reduced the phosphorylation of p65 (3.6-fold) at Ser-276
residue whereas the level of Ser-536 phosphorylation was unaltered
(Fig. 3B). MSK1, a downstream target of ERK, phosphorylates
multiple substrates including Ser-276 of p65. As depicted in Fig. 3C,
phosphorylation of p65 at Ser-276 residue by cystatin was
significantly abrogated (3.2-fold) in infected cells transfected with
dominant-negative (dn) constructs of MSK1 with no alteration on
p65 Ser-536 phosphorylation. Moreover, MSK1 activity could be
significantly inhibited by apigenin (86%) and BAY11-7082 (91%)
(Fig. 3D) suggesting IKK-ERK-mediated MSK1 activation, which in
turn regulates Ser-276 phosphorylation of p65. Collectively, these
results suggest that IKK induced the phosphorylation of p65 directly
on Ser-536 residue and indirectly on Ser 276 residue by activation of
ERK/MSK1 in cystatin plus IFN-g-treated cells.
TAK1 is involved in cystatin plus IFN-c-mediated
NF-jB activation
The MAP3 kinase kinase kinase, TAK1 is known to mediate the
pathway leading to IKKa/b phosphorylation resulting in classic
NF-kB activation through IkB phosphorylation and degradation
[8]. We, therefore, examined the possibility of TAK1 involvement
in intermediate signaling events of cystatin plus IFN-g-induced
NF-kB activation. Kinetic analysis (0–60min) following stimula-
tion with cystatin plus IFN-g revealed time-dependent phosphory-
lation of TAK1 in both normal and infected BMM with lesser
induction and slower kinetics in infected cells (Fig. 4A). We
further analyzed the role of cystatin-mediated TAK1 induction on
both IKK activation and NF-kB-driven luciferase expression.
Transient transfection of RAW 264.7 cells with dn constructs of
TAK1 significantly reduced the amount of cystatin-induced
phospho-IkBa and prevented IkBa degradation (Fig. 4B) suggest-
ing a role of TAK1 in cystatin-mediated IKK activation. In
addition, overexpression of dn-TAK1 markedly inhibited cystatin
plus IFN-g-induced IKKb activity (Fig. 4C) as well as NF-kB-
driven luciferase expression (2.5-fold reduction) (Fig. 4D).
Cystatin-mediated anti-leishmanial activity was also significantly
decreased in infected cells (68% reduction in parasite killing)
transfected with dn TAK1 (Fig. 4E) implying the importance of
TAK1 in cystatin-mediated activation of NF-kB pathway.
TLR/MyD88-dependent activation of NF-jB is essential
for cystatin-mediated anti-leishmanial response
In the TLR/MyD88-signaling pathway, activated IRAK1 interacts
with TRAF6, which in turn leads to the activation of TAK1-
mediated NF-kB and MAPK [5, 6, 8] We, therefore, checked
whether MyD88-dependent signaling is necessary for cystatin-
induced activation of TAK1. To address this, RAW 264.7 cells
Figure 3. Role of cystatin plus IFN-g on p65 phosphorylation. (A) L. donovani infected (24h) BMM were treated with cystatin plus IFN-g for the
indicated times. Cells were lysed and levels of p-ser536p65, p-ser276p65 and p65 were determined by Western blotting. (B) Infected cells (24h) were
left untreated or pretreated with apigenin (40mM) or BAY11-7082 (5mM) for 1h, followed by incubation with cystatin plus IFN-g for 2h. Levels of
p-ser536p65, p-ser276p65 and p65 were determined by Western blotting. (C) Infected BMM (24h) were transiently transfected with WTor dn-MSK1
expression plasmid. After 24h of transfection, cells were stimulated with cystatin plus IFN-g for 2h. Levels of p-ser536p65, p-ser276p65 and p65
were determined by Western blotting. (D) Infected cells were incubated with cystatin plus IFN-g for various time periods (0–120min). In a separate
set of experiment, infected cells were treated with apigenin (40mM) or BAY11-7082 (5mM) for 1h before treatment with cystatin plus IFN-g for
90min. Cell lysates were then immunoprecipitated with anti-MSK1 Ab, and the immunoprecipitate was used to measure the phosphorylation of
CREBTIDE as described in Materials and methods section (n54). Bands were analyzed densitometrically. Error bars represent mean1SD, n53. The
data shown are representative of three independent experiments.???p o 0.001; Student’s t-test.
Eur. J. Immunol. 2011. 41: 116–127Susanta Kar et al.
120
& 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eji-journal.eu