| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Corixa Corporation, Seattle, WA 98104;
College of Medicine, King Khalid University, Abha, Saudi Arabia;
Infectious Disease Research Institute, Seattle, WA 98104;
Federal University of Bahia, Salvador, Bahia, Brazil; and
¶ Department of Biochemistry, Microbiology and Immunology, University of Ottawa, Ottawa, Ontario, Canada
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
- and 
-tubulin, histone H2b,
ribosomal protein S4, malate dehydrogenase, and elongation factor 2, as
well as two novel parasite proteins. None of these proteins have been
previously reported as T cell-stimulating Ags from
Leishmania. -tubulin-specific T cell clones generated
against Leishmania major amastigotes responded to
Leishmania-infected macrophages and dendritic cells.
IFN-
enzyme-linked immunospot analysis demonstrated the presence of
T cells specific for several of these Ags in PBMC from self-healing
cutaneous leishmaniasis patients infected with either Leishmania
tropica or L. major. The responses elicited by
Leishmania histone H2b were particularly striking in
terms of frequency of histone-specific T cells in PBMC (1 T cell
of 6000 PBMC) as well as the percentage of responding donors (86%, 6
of 7). Ags identified by T cells from immune donors might constitute
potential vaccine candidates for leishmaniasis. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
350 million people were at risk of acquiring leishmaniasis and
that 12 million were currently infected (4). In a recent
epidemic of visceral leishmaniasis
(VL)3 in southern
Sudan, Medicins San Frontièrers estimated that the excess
mortality has been about 100,000 deaths among about 300,000 people at
risk (5).
The majority of both visceral and cutaneous leishmanial
infections appear to be subclinical or self-limited. Individuals with
these relatively uncomplicated infections develop long-term immunity to
Leishmania and are protected against reinfection
(6). It is generally accepted that control of leishmanial
infection is mediated by a Th1-type immune response activation of
macrophages with IFN- to a microbicidal state (2, 7, 8). The emergence of VL as an opportunistic infection among
people with HIV infection highlights the importance of
CD4+ T cells for protective immunity against
Leishmania (1, 9). Taken together, these data
imply that development of a vaccine against leishmaniasis is feasible
but would require the induction of an appropriate T cell-mediated
immune response. Deliberate infection with in vitro-cultivated
Leishmania major promastigotes (leishmanization) has been
reportedly used in Russia, Israel, and Iran (10), and
although it is reported to be highly effective, leishmanization is not
recommended due to obvious safety concerns associated with injection of
live virulent organisms (11). Vaccination using
preparations of killed Leishmania either alone or in the
presence of bacillus Calmette-Guérin have also been attempted
(reviewed in Ref. 11). Because the protection rates
obtained in these trials varied considerably (0–82%) the efficacy of
killed promastigote preparations remains unconfirmed.
Together these findings suggest that vaccination using defined recombinant parasite Ags may be a more effective and safe alternative for eliciting protective T cell responses. Therefore, it is imperative to identify parasite proteins that are recognized by T cells during human leishmanial infections. A number of Leishmania Ags have been cloned and characterized with respect to the immune responses that they elicit during experimental murine infections (12, 13, 14, 15, 16, 17). However, there is a paucity of information regarding the Ag specificity of T cells from humans that acquired persistent immunity after leishmanial infections. Thus, we have successfully generated parasite-specific CD4+ T cell lines from the PBMC of a Montenegro (DTH) skin test-positive adult residing within an area of Northeastern Brazil that is endemic for VL. Although this individual has a positive DTH skin test he has no clinical history of disease. Therefore, it is assumed that he was previously infected by Leishmania, but developed a protective immune response and successfully controlled the infection. As mentioned above, it is estimated that a high percentage of infections with VL-causing organisms are self-resolved in this manner (18, 19, 20). To facilitate the identification of Ags that are recognized by these T cell lines we have used a direct T cell screening approach that was recently used to identify an immunodominant T cell Ag of Mycobacterium tuberculosis (21). Herein, we used Leishmania-specific CD4+ T cells lines to screen a cDNA expression library generated from L. major amastigote parasites. Using this approach we have identified, cloned, and expressed eight Leishmania proteins that are recognized by CD4+ T cells from a immune donor residing in a region endemic for VL. Furthermore, we demonstrate that PBMC and CD4+ T cell lines from donors with self-healing cutaneous leishmaniasis (CL) due to Leishmania tropica or L. major infection recognize a number of the same Ags. These results suggest that these Leishmania Ags may elicit protective immune responses in different forms of leishmaniasis caused by highly divergent Leishmania species.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Promastigotes of L. major (Friedlin strain) were
maintained at 26°C in M199 with 10% FCS, 40 mM HEPES pH 7.4, 100
µM adenine, and 0.005% hemin. Lysate of L. major
promastigotes (LPr) was prepared by sonicating promastigotes in PBS
containing protease inhibitors (one complete protease inhibitor tablet
(Boehringer Mannheim, Indianapolis, IN) per 25 ml PBS). Culture
filtrate of L. major promastigotes (CFPr) was prepared by
growing promastigotes (1 x 107/ml)
overnight in AIMV serum-free medium (Life Sciences, St. Petersburg, FL)
followed by removal of parasites by centrifugation and filtration
through a 0.2-µm sterile filter. Supernatants were concentrated
50-fold by ultrafiltration with a 3-kDa-cutoff filter unit (Amicon,
Beverly, MA). Amastigotes of L. major were recovered from
the nonulcerative lesions of SCID mice that had received an intradermal
inoculation of promastigotes at the base of the tail 6–8 wk earlier.
Culture filtrate of L. major amastigotes (CFAm) was prepared
as described above for CFPr without concentrating the supernatants.
Patients
T cell lines and clones were generated from a Montenegro skin test-positive male donor living in an endemic area for VL due to infection by Leishmania chagasi in Northeastern Brazil. The donor had no clinical history of leishmanial disease. PBMC from CL patients were obtained after spontaneous cure of CL from seven untreated patients living in Abha, Saudi Arabia. Patients were diagnosed with self-resolved CL due to L. tropica (six cases) or L. major (one case) infections. Lesions due to CL healed spontaneously in a period of 2 (five cases), 6 (one case), or 12 mo (one case), respectively.
Generation of T cell lines and clones
PBMC were obtained from the apheresis product of the Montenegro skin test-positive donors by density centrifugation over Ficoll-Hypaque (Sigma, St. Louis, MO). T cell lines were generated by stimulating PBMC (5 x 106) with either LPr or CFPr (10 µg/ml each) in TCM (RPMI 1640 containing 10% human AB serum, 25 mM L-glutamine, 25 mM HEPES, and 50 µg/ml gentamicin). After 3 days IL-2 (10 ng/ml) was added to the cultures. Cultures were split every 3–5 days and restimulated with autologous irradiated or HLA-matched PBMC and the respective Ag every 13–16 days. Before restimulation, T cell lines were cultured for at least 3 days in resting medium (TCM containing 10 ng/ml IL-7 and 0.5 ng/ml IL-2). After three rounds of Ag-specific stimulation T cells were expanded with soluble anti-CD3 Ab and irradiated feeder cells. Short-term T cell lines were generated using the same protocol as described above except that donor PBMC received only one cycle of Ag-specific stimulation. After 10 days, short-term T cell lines were cultured for 3 days in resting medium before the cellular immune responses of T cells against leishmanial Ags were determined. T cell clones were obtained from a T cell line generated against amastigote culture filtrate by limiting dilution on Terasaki microwell plates (Nunc, Naperville, IL) using a 1:10 dilution of amastigote culture filtrate, IL-2-supplemented TCM, and autologous irradiated feeder cells.
Library screening
A L. major amastigote cDNA expression library
(17, 22) prepared in the 
-ZAP vector (Stratagene, La
Jolla, CA) was chosen as the initial library to be screened. For
screening purposes, this phage library was converted to a plasmid-based
library using the mass excision protocol supplied by the manufacturer
(Stratagene). The plasmid library was transferred to 96-well plates at
a density of 40–50 clones per well, grown to mid-log phase, and frozen
at -80°C as a glycerol stock. For screening, small aliquots of the
frozen libraries were removed to fresh 96-well plates and grown to
mid-log phase. Cultures were then induced with isopropyl
-D-thiogalactoside for 3 h, after which time the
bacteria were pelleted in the 96-well plates. Bacteria were resuspended
in 200 µl of antibiotic-free RPMI 1640 containing 10% FBS, and 10
µl from each well was transferred to new well of a 96-well plate
containing 1 x 104 irradiated (3000 rad)
HLA-DR-, -DQ-matched monocyte-derived dendritic cells (MoDC)
(23). After a 90-min incubation at 37°C, the adherent
MoDC were washed twice with TCM. Parasite-specific T cells
(2.5 x 104/well) were then added and
cultured for 4 days. Proliferative response of the T cells was measured
using a standard [3H]thymidine incorporation
assay. IFN- secretion by T cells was determined by ELISA analysis of
supernatants taken after 3 days as described (24).
Recombinant proteins
All proteins were expressed in Escherichia coli as amino-terminal histidine-tagged proteins to facilitate purification by nickel-chelate affinity chromatography. Full-length cDNA clones of Ags 1G6, 4A5, and 1B11 were obtained during T cell screening, and the corresponding full-length proteins were expressed in recombinant form. A truncated cDNA clone was initially obtained for clone 8G3, and the full-length sequence was determined by 5' rapid amplification of cDNA end PCR using a spliced leader primer and a sequence-specific 3' primer. Similarly, the clone 4H6 was initially isolated as a truncated cDNA clone, and the full-length open reading frame (ORF) was obtained by PCR using primers that were based upon the previously deposited L. major sequence (accession number X93567). The full-length ORFs were amplified by PCR to facilitate the addition of N-terminal 6x histidine tags and 5' and 3' restriction sites. Sequences of oligonucleotide primers are as follows: 1G6, 5'-CAATTACATATGCATCACCATCACCATCACATGGCTTCTTCCCGCAAGGCT3' and 5'-CATGGAATTCTTAGCGGGACGCGCTCGACAC-3'; 4A5, 5'-CAATTACATATGCATCACCATCACCATCACATGAGCATTATCAAGGAGGA-3' and 5'-GTAAAACGACGGCCAGTGAAT-3'; 1B11, 5'-CAATTACATATGCATCACCATCACCATCACATGGCCAAGAAGCACCTCA-3' and 5'-GTAAAACGACGGCCAGTGAAT-3'; 8G3,5'-ATGGCTAGCCATCACCATCACCATCACATGGTGAACGTGTGCGTTGTT-3' and 5'-CATGGAATTCTGAGGTGCTACAGGTGTGAGC-3'; and 4H6, 5'-CAATTACATATGCATCACCATCACCATCACATGCGTGAGATCGTTTCCTGC-3' and 5'-CATGGAATTCCTAGTAGGCCTCCTCCTCCTCG-3'.
PCR products were digested with NdeI/EcoRI (1G6), NdeI (4A5), NdeI/XhoI (1B11), NheI/EcoRI (8G3), or NdeI/EcoRI (4H6) and were cloned into pET17b (Novagen, Madison, WI) vector digested with the same enzymes. Clones 4G2, 1E6, and 2A10 were expressed as truncated recombinant proteins. The initial 4G2 expression construct (encoding the carboxyl-terminal 595 aa) was prepared by PCR amplification using the primers 5'-CAATTACATATGCATCACCATCACCATCACAAGAACAAGAAGTGGATCAA-3' and 5'-GTAAAACGACGGCCAGTGAAT-3' followed by cloning of the PCR product into pET17b using NdeI/XhoI. The recombinant protein encoded by this construct expressed poorly in E. coli; therefore, a smaller construct called 2F11 (encoding the carboxyl-terminal 475 aa) was subsequently prepared, also by PCR amplification using the primers 5'-CAATTACATATGCATCACCATCACCATCACAAGAACAAGAAGTGGATCAA-3' and 5'-GTAAAACGACGGCCAGTGAAT-3' followed by cloning of the PCR product into pET17b using NdeI/XhoI. Clones 1E6 and 2A10 were subcloned directly from the initial pBluescript plasmids into the vectors pQE30 and pQE32, respectively, (Qiagen, Chatsworth, CA) using BamHI (5') and KpnI (3') restriction sites. Recombinant proteins containing N-terminal 6x histidine tags were purified on Ni-NTA agarose (Qiagen) according to previously described protocols (22, 25). All recombinant proteins were routinely assayed for endotoxin contamination using the Limulus amebocyte lysate assay (BioWhittaker, Walkersville, MD) and were uniformly found to contain <100 E.U./mg protein. Purified bovine tubulin was obtained from Molecular Probes (Eugene, OR) and calf thymus-derived histone was obtained from Calbiochem (La Jolla, CA).
Infection of APCs with L. major
To infect human MoDC and macrophages with metacyclic L. major promastigotes, parasites were centrifuged at 1400 x g on 1 x 104/well MoDC or macrophages in a 96-well plate for 30 min at room temperature. After centrifugation, plates were washed twice and then cultured for 7 days in 150 µl TCM before T cells were added. MoDC and macrophages were generated from adherent PBMC as described (23).
Proliferation and cytokine assays
PBMC responses to leishmanial Ags were tested by culturing PBMC
in the presence of Ag for 6 days using 3 x
105 PBMC/well in 200 µl TCM in 96-well
round-bottom plates with a [3H]thymidine pulse
(0.5 µCi/well) for the last 18 h. IL-5 and IFN- secretion was
determined in supernatants taken after 3 days by ELISA. IL-5 and
IFN- ELISA were performed as described (24). Fine
specificities of T cell lines were determined by testing their response
to leishmanial Ags using 2.5 x 104/well T
cells in the presence of either 2.5 x 104 T
cell-depleted PBMC or 1 x 104 HLA-DR-,
-DQ-matched allogenic MoDC in a total volume of 150 µl TCM. Cells
were cultured for 3 days with a [3H]thymidine
pulse (0.5 µCi/well) for the last 18 h. IL-5 and IFN-
secretion was determined in supernatants taken after 2 days. Data are
presented as the mean of triplicate cultures.
IFN- enzyme-linked immunospot (ELISPOT) assay
IFN- ELISPOT assays were performed in triplicate following
instructions provided by MABTECH (Stockholm, Sweden). Briefly, 96-well
MultiScreen-HA plates (Millipore, Bedford, MA) were coated overnight
with 10 µg/well capture Ab to human IFN- (1-1DK; MABTECH) at
4°C. After washing the plates three times with PBS, plates were
blocked with 100 µl/well TCM for 1 h at 37°C. Numbers of
IFN- spot-forming cells (SFC) were determined from wells containing
either 3 x 105 or 1.5 x
105 PBMC/well and 1.5 x
105 irradiated autologous PBMC. PBMC were
cultured in the presence of leishmanial Ag at 37°C 5%
CO2 in TCM. After 48 h cells were removed
and ELISPOT plates were washed three times with PBS containing 0.1%
Tween 20 before adding a secondary biotinylated Ab to IFN- (1
µg/ml, 7-B6-1; MABTECH). Plates were incubated for 2 h at room
temperature, then washed with PBS containing 0.1% Tween 20 before
adding 100 µl of avidin-peroxidase-complex (Vectastain). After 1
h plates were washed and developed with an AEC substrate kit
(Vectastain). A computer-assisted video image analysis system (KS
Elispot, Zeiss, Oberkochen, Germany) was used to quantify
IFN--producing cells. The cut-off value was set as the mean of
triplicates at 4.5 SFC/well, setting an arbitrary lower detection limit
of 1 in 70,000 PBMC.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To facilitate the identification of T cell-stimulating Ags from
Leishmania, T cell lines were generated from PBMC of a
Montenegro skin test-positive donor living in an area of Northeastern
Brazil that is endemic for VL due to infection by L.
chagasi. Although the donor had no clinical history of
leishmaniasis, the positive Montenegro skin test (DTH) indicates that
the donor was previously exposed to Leishmania and developed
a protective immune response against the infection. Three different T
cell lines were generated by stimulating PBMC with 10 µg/ml LPr, 10
µg/ml CFPr, or a 1:10 dilution of CFAm, respectively. T cell lines
were expanded by repeated stimulation with HLA-DR-, -DQ-matched APC and
the respective L. major Ag preparation. The established T
cell lines showed strong proliferative responses against all three Ag
preparations (Table I
), indicating that
at least some T cell-stimulating Ags were shared by all three
preparations. The established T cell lines were shown by flow cytometry
to be comprised of CD4+, TCR
+ T cells (data not shown).
|
To identify Ags that are expressed in both the promastigote and
amastigote life stages we screened an amastigote cDNA expression
library with T cell lines generated against LPr or CFPr, respectively.
The library was a L. major amastigote -ZAP phage library
(described previously; Refs. 17, 22) that was converted to
a plasmid-based library using the excision protocol supplied by the
manufacturer (Stratagene). The titer of the plasmid library was
determined by serial dilution, and pools were prepared in 96-well
format at a density of 40–50 clones per well. The library was
screened as previously described (21) except that HLA-DR-,
-DQ-matched MoDC rather than autologous MoDC were used as APC. A total
of ten 96-well plates of the cDNA library 400–48(38,400–48,000 individual
cDNA clones) were screened using both the LPr and CFPr T cell lines.
Pools that elicited positive responses from the T cell lines were
sequentially broken down until individual immunoreactive clones were
obtained. As an example, Table II shows
the identification of two positive library pools and the subsequent
breakdown of these pools to the clonal level using a subpooling
approach. The pool 1G6 induced a specific proliferative response from
the CFPr T cell line, whereas the pool 4H6 induced a specific response
from the LPr line. The estimated complexity of the initial library
pools was corroborated by the fact that only 1% of the colonies
isolated from pool 1G6 and 0.5% of the colonies from pool 4H6 were
positive during the second round of screening (data not shown).
However, 100% of the colonies identified as positive during the
second round of screening remained positive on subsequent assays. This
finding, together with a general lack of proliferation in response to
control E. coli, confirms that the library screening
protocol is both sensitive and specific for the detection of E.
coli clones that are expressing leishmanial Ags recognized by T
cells.
|
. The 39 clones comprised overlapping
derivatives of eight distinct genes, three of which were identified
with the CFPr T cell line (1G6, 1E6, and 4A5) and five with the LPr T
cell line (1B11, 2A10, 2F11, 4H6, and 8G3). Interestingly, the LPr and
CFPr T cell lines recognized distinct sets of pools with the exception
of the pool 8G3, which was strongly reactive for the LPr T cell line
and weakly reactive in screens using the CFPr T cell line. Clones 1G6,
2A10, 4H6, and 8G3 were found to comprise previously characterized
Leishmania sequences (histone H2b, -tubulin, -tubulin,
and malate dehydrogenase, respectively). The sequence of the
Leishmania histone H2b was markedly divergent from the human
homolog, particularly in the amino terminal half of the protein where
there is a large gap in the Leishmania sequence when
compared with the H2b proteins from higher eukaryotes (data not shown).
Interestingly, two distinct cDNA clones for Leishmania
histone H2b were detected by the CFPr line; these clones were divergent
between themselves in the amino terminal region. Similarly, the -
and -tubulin sequences of Leishmania contained blocks of
sequence divergence when aligned to the homologous sequences of
higher eukaryotes (data not shown). The clone originally obtained for
malate dehydrogenase (8G3) contained a severely truncated cDNA sequence
that encoded only the carboxyl-terminal 69 aa of the ORF. A peptide
comprising the carboxyl-terminal 14 aa of this protein was recognized
by the LPr T cell line as demonstrated by epitope-mapping experiments
using overlapping synthetic peptides (J. Webb, manuscript in
preparation). Interestingly, this protein contains an SHL tripeptide at
its carboxy terminus, indicating that it represents an isoform of
malate dehydrogenase that is targeted to the glycosome, an organelle
that is unique to kinetoplastid organisms (26). Two other
genes isolated herein had significant homology to known sequences
(2F11-elongation factor 2 and 1B11-ribosomal protein S4) that have not
been previously cloned from Leishmania. The two remaining
Ags exhibited either no similarity to existing databank entries (4A5)
or weak similarity to a family of cell cycle-regulated proteins from
higher eukaryotes (1E6). To our knowledge, none of the clones
identified in this study have been previously characterized as Ags
capable of eliciting responses from human
Leishmania-specific T cells.
|
All eight of the Ags identified by direct T cell screening were
expressed in E. coli as recombinant fusion proteins
containing N-terminal histidine tags and were purified to homogeneity
using nickel affinity chromatography. All eight purified recombinant
proteins elicited strong proliferative responses from the respective
LPr- or CFPr-specific T cell lines (Fig. 1) as expected from the data of the
library screening. Purified recombinant proteins and the respective
intact E. coli clones elicited a similar level of response
from the respective T cell lines (data not shown). The T cell responses
to leishmanial tubulin and histone H2b did not evoke an autoreactive
response to the mammalian homologs because both T cell lines did not
respond to bovine tubulin or calf thymus-derived histone, which are
100% homologous to the respective human proteins. The responses of
short-term T cell lines (PBMC stimulated for one round with parasite
lysate or culture filtrate) confirmed that this donor has strong
cellular immune responses against the Ags identified by CD4 T cell
expression cloning (data not shown). The immune response elicited by
these Ags may contribute to a state of protective immunity in this
donor.
|
-tubulin, and malate
dehydrogenase as well as infected macrophages and dendritic cells
Because promastigote proteins were used to generate the T cell
lines used for expression screening, we next generated a T cell line
against CFAm and evaluated the ability of this line to recognize the
eight recombinant proteins described above. As shown in Fig. 2, the CFAm T cell line recognized three
of the eight Ags identified using the LPr- and CFPr-specific T cell
lines. Recombinant Ag 1G6 (histone H2B), 4H6 (-tubulin), as well as
8G3 (malate dehydrogenase) induced a strong proliferative response in
CFAm T cells. This result indicates that these Ags are expressed in
both the promastigote and amastigote life stages at levels that are
sufficient to evoke strong cellular immune responses. Furthermore, Ags
4H6 and 8G3 were also recognized by Leishmania-specific T
cell clones that had been generated against amastigote culture filtrate
(Fig. 3). These T cell clones were
derived from a CFAm-specific T cell line by limiting dilution in the
presence of CFAm and autologous APC. In addition, MoDC and macrophages
infected with L. major were capable of evoking proliferative
responses (Fig. 3) and IFN- production (data not shown) from the
4H6-specific T cell clone CFAm15. Interestingly, no such response was
elicited from the 8G3-specific T cell clone CFAm21, suggesting that
there are Ag-dependent differences in Ag presentation during natural
infection of APC with L. major.
|
|
To further evaluate the human immune response to the eight Ags
identified by direct T cell screening, the responses of PBMC from
patients with self-resolved CL were characterized. CL patients acquire
persistent protective immune responses against Leishmania.
Therefore, Ags recognized by PBMC from these patients might be involved
in the control of a leishmanial infection. PBMC responses of six CL
patients with L. tropica infection and one CL patient with
L. major infection were characterized by proliferation and
IFN- ELISPOT. In addition to the eight Ags identified by expression
cloning, two previously described leishmanial proteins (surface
protease gp63, reviewed in Ref. 2 ; Leishmania
homolog of the intracellular receptor for activated protein kinase
(LACK); Ref. 16), leishmanial promastigote lysate, bovine
histone, bovine tubulin, and purified protein derivative (PPD) from
M. tuberculosis were tested. Data are shown in Fig. 4. All seven donors had a high frequency
of parasite-specific T cells in their PBMC with a mean of 1 in
1700 ± 430 cells producing IFN- in response to leishmanial
lysate (as determined by IFN- ELISPOT). Furthermore, six of seven CL
donors had a strong proliferative response to leishmanial lysate (mean
stimulation index (SI) of
responders4 was
29 ± 5.2 with an arbitrary cut-off SI of 4). In general, all CL
donors responded to at least one of the eight identified Ags. Most
notable were the responses to 1G6 (histone H2b) and 4H6 (-tubulin).
1G6 and 4H6 elicited responses from six of seven and seven of seven CL
patients, respectively, as measured by IFN- ELISPOT and from five of
seven and five of seven, respectively, as measured by in vitro
proliferative responses. Compared with the other recombinant Ags
tested, IFN--producing 1G6-specific T cells had the highest
frequency in PBMC from CL patients (1 in 5900 ± 2400, mean of
responders; Ref. 4) and elicited the strongest
proliferative response (SI of 21 ± 4.3, mean of responders). In
addition, 1G6 was not recognized by PBMC from uninfected North American
controls, indicating a Leishmania-specific response to 1G6
in PBMC from CL patients. None of the CL patients responded to bovine
tubulin or bovine histone, excluding the possibility of
cross-reactivity against the human homologs. Ags 1B11 (ribosomal S4
protein) and 2F11 (elongation initiation factor 2A) elicited responses
in approximately half of the patients. Interestingly, although 1E6 and
8G3 (malate dehydrogenase) induced a proliferative response in four of
seven and two of seven CL patients, respectively, IFN--secreting
cells were not detectable by ELISPOT. As shown in Fig. 4, a limited
number of normal North American control donors responded to leishmanial
lysate or to the recombinant Ags 1E6, 2F11, 4H6, or 8G3 with weak
proliferation and/or IFN- production, indicating the potential for
cross-reactivity to other microbial organisms common in North America.
However, to fully evaluate the response of normal donors, a greater
pool of donors needs be investigated.
|
). IFN- production by the short-term
T cell lines revealed a similar pattern of Ag reactivity as was
observed with the PBMC of CL patients. The short-term T cell lines did
not produce IFN- in response to malate dehydrogenase, protein 1E6,
or protein 4A5. Again, 1G6 (histone H2b) and 4H6 (-tubulin) induced
the strongest response in these short-term lines. In contrast, only one
of the short-term CL T cell lines produced significant amounts of
the Th2-associated cytokine IL-5. Interestingly, this particular line
produced both IL-5 and IFN- in the presence of either
Leishmania lysate or recombinant 1G6, indicating a Th0-like
response. However, all other T cell lines produced a Th1-like pattern
with high levels of IFN- and low to undetectable amounts IL-5 in
response to the recognized Ags.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Herein we used a recently described direct T cell-screening approach
(21) to identify Ags recognized by
Leishmania-specific human T cell lines. To identify T cell
epitopes that are shared by multiple Leishmania species, a
cDNA library of amastigote L. major (endemic in the Old
World) was screened with T cells from a donor controlling a L.
chagasi infection (a Leishmania species endemic in
Latin America). By using this approach we identified eight
leishmanial T cell-stimulating Ags. To our knowledge none of these Ags
have been previously shown to induce T cell responses in leishmaniasis
patients, although histone H2b, - and -tubulin, as well as malate
dehydrogenase have been previously characterized in terms of
biochemistry or molecular biology (27, 28, 29, 30).
Our rationale for screening an amastigote cDNA library with T cell
lines raised against promastigote lysate and culture filtrate was to
define Ags that are shared by both life stages. Indeed, at least five
of the T cell-stimulating Leishmania Ags reported herein
(- and -tubulin, histone H2b, elongation factor 2a, and ribosomal
protein S4) represent cytoplasmic or nuclear proteins that are
abundantly expressed in the promastigote as well as amastigote
life stage of the parasite (31, 32). Furthermore, - and
-tubulin are also major components of the leishmanial flagella. The
isoform of malate dehydrogenase identified herein is located within the
glycosome, a membrane-bound organelle unique to kinetoplastid
organisms.
Interestingly, the T cell lines generated against promastigote lysate and culture filtrate exhibited distinct differences in terms of the pattern of Ags recognized. Of the Ags studied herein, only malate dehydrogenase (8G3) was recognized by both long-term T cell lines. Nonetheless, both lines react strongly with preparations of total lysate or total culture filtrate. This result suggests that these lines may recognize additional Ags, which have not been cloned during this screen possibly due to either underrepresentation in the cDNA expression library or poor expression and/or processing during the screening procedure.
Interestingly, T cell reactivity to histone H2B, -tubulin, and
malate dehydrogenase was observed in T cell lines generated against
promastigote Ag preparations as well as in the T cell line generated
against amastigote culture filtrate. Although the presence of these
proteins in culture filtrate is more likely to be the result of
parasite autolysis rather than active protein secretion, a combination
of high abundance and (apparent) high solubility may favor the MHC
class II-mediated presentation of these proteins in
Leishmania-infected APC. Indeed, -tubulin-specific T cell
clones strongly recognized both L. major-infected
macrophages as well as L. major-infected dendritic cells,
indicating that -tubulin is presented by infected cells at levels
sufficient to elicit strong class II-restricted T cell responses. In
contrast, Leishmania-infected monocytes or dendritic cells
did not stimulate a malate dehydrogenase-specific T cell clone despite
the fact that these same infected cells were capable of processing and
presenting exogenous recombinant malate dehydrogenase (data not shown).
Interestingly, our data with malate dehydrogenase resemble recently
published studies showing that in the murine system
Leishmania-infected APC were not able to stimulate LACK- or
cysteine proteinase-specific T cells (33, 34).
Cellular immune responses to the eight Ags identified by CD4 expression
cloning were further evaluated with PBMC as well as short-term T cell
lines generated from patients with self-resolved CL due to L.
tropica or L. major infections. These patients control
infection and acquire persistent immunity against
Leishmania. T cells from CL donors responded to histone H2b
(1G6), -tubulin (4H6), ribosomal protein S4 (1B11), elongation
factor 2 (2F11), and -tubulin (2A10). However, most notable were the
dominant T cell responses to histone H2b and -tubulin. Histone
H2b-reactive IFN--producing T cells were present at a very high
frequency in PBMC from CL patients. In addition, histone H2b elicited
strong proliferative responses in a broad range of CL patients.
Interestingly, humoral immune responses against histone H2b were
recently reported in dogs with canine VL. Abs to histone H2b as well as
other members of the histone family were detected after the dogs
developed VL due to a Leishmania infantum infection
(35, 36). These data suggest that histone H2b is available
for immune recognition not only in situations where the immune system
has controlled the infection but also in situations where the parasite
has evaded protective immune mechanisms. Notably, neither H2b-reactive
T cells (this study) nor histone H2b-reactive dog sera recognized their
mammalian homologs, indicating that histone H2b reactivity is not
indicative of an autoreactive immune response.
Like histone H2b, -tubulin is one of the most abundant proteins in
leishmanial promastigotes and amastigotes (31). However,
unlike the histone H2b protein, which is relatively divergent from its
mammalian homolog (50% identity to human histone H2b), -tubulin
exhibits 86% identity to the human counterpart. Despite this level of
homology, the tubulin-specific T cell reactivity observed in seven of
seven tested CL patients was completely specific for parasite tubulin
as no response could be elicited by purified mammalian tubulin. This
was also indicated by preliminary epitope-mapping experiments showing
that T cell reactivity was restricted to a small block of
nonconservative amino acid substitutions (data not shown). This finding
demonstrates the exquisite sensitivity of T cells to detect even subtle
sequence differences between the proteins of an invading pathogen and
the corresponding host homologs and that recognition of such
differences may have profound consequences on the course of
infection.
Interestingly, although malate dehydrogenase (8G3) and 1E6 elicited strong T cell responses in short-term T cell lines from the Brazilian donor (data not shown), they did not evoke a specific T cell response in the CL patients. Because all these Ags were initially derived from a L. major library, the nonresponsiveness of CL donors against these three Ags may be indicative of HLA class II restriction differences. Alternatively, differences in the disease profile of these disparate types of patients may result in distinct patterns of Ag recognition.
There is currently little information available regarding the response of human T cells to LACK. In this study LACK did not induce a T cell response in PBMC from CL patients. A 24-kDa portion of LACK protected susceptible BALB/c mice when administered as a vaccine with IL-12 in the experimental L. major mouse model. Interestingly, the susceptibility of BALB/c to L. major may be mediated by an early IL-4 T cell response to LACK because mice made tolerant to LACK are ablated of early IL-4 production and exhibit a healing phenotype (reviewed in Ref. 37). However, cellular immune responses to Leishmania in humans are probably more complex and less polarized than they are in mice.
The data in this study support a recently reviewed theory about the
association of Ag presentation by Leishmania-infected
macrophages and the control of infection and its implications for the
design of subunit vaccines (38). Amastigote
Leishmania replicate in resting macrophages. Upon activation
with IFN- the macrophages kill the parasite and during the
subsequent degradation of amastigote present an extremely heterogeneous
peptide repertoire. Under these circumstances, the abundance of a
particular Ag might be an important factor in eliciting a strong immune
response. Herein, we have demonstrated that -tubulin-specific T
cells have the capacity to recognize infected macrophages and dendritic
cells. It is envisioned that, in vivo, cells such as these
-tubulin-specific T cells mediate the initial activation of infected
macrophages to kill intracellular parasites and, consequently, drive
the presentation of a more diverse repertoire of parasite Ags in the
class II pathway. The T cell response to abundant cytoplasmic proteins
like histone H2b, tubulins, and ribosomal protein S4 might then be
triggered by presentation of these killed organisms by activated APC.
In an intercellular chain reaction these T cell responses would then
activate more macrophages in the vicinity. Eventually, this combination
of parasite-specific T cell responses would be expected to reach a
point at which the parasite is forced into either a low level
subclinical latent infection or is eliminated entirely.
In summary, this report describes the identification of T
cell-stimulating Ags from Leishmania, recognized by T cells
from donors who developed a protective immune response against
leishmanial infections. Although it is unclear at this point whether T
cell responses to leishmanial histone H2b and -tubulin either alone
or in combination participate in the control of the initial infection,
the Th1-type immune responses elicited against these Ags may contribute
to a state of protective immunity in these donors. Thus it is likely
that Ags like histone H2b and -tubulin identified by a direct T
cell-screening approach might constitute potential candidates for
inclusion in a subunit vaccine against leishmaniasis.
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Peter Probst, Corixa Corporation, 1124 Columbia Street, Suite 200, Seattle, WA 98104.
3 Abbreviations used in this paper: VL, visceral leishmaniasis; CL, cutaneous leishmaniasis; MoDC, monocyte-derived dendritic cell(s); LPr, Leishmania major promastigote lysate; CFPr, L. major promastigote culture filtrate; CFAm, L. major amastigote culture filtrate; SFC, spot-forming cells; LACK, Leishmania homologue of the intracellular receptor for activated protein kinase; TCM, RPMI 1640 containing 10% human AB serum, 25 mM L-glutamine, 25 mM HEPES, and 50 µg/ml gentamicin; ORF, open reading frame; ELISPOT, enzyme-linked immunospot; PPD, purified protein derivative; SI, stimulation index.
4 Mean of responders: Mean ± SEM was calculated from responding PBMC. Responders were defined by arbitrary cutoff value of 4.5 SFC for IFN- ELISPOT assay or a SI >4 for proliferation, respectively.
Received for publication June 5, 2000. Accepted for publication October 2, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
production and expression of B7-1 in human monocyte-derived antigen-presenting cells. Eur. J. Immunol. 27:2634.[Medline]
-tubulin genes during the acquisition of promastigote infectivity. Mol. Biochem. Parasitol. 82:227.[Medline]
This article has been cited by other articles:
|
|
 |
|
  R. B. Goncalves, O. Leshem, K. Bernards, J. R. Webb, P. P. Stashenko, and A. Campos-Neto T-Cell Expression Cloning of Porphyromonas gingivalis Genes Coding for T Helper-Biased Immune Responses during Infection Infect. Immun., July 1, 2006; 74(7): 3958 - 3966. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y.-T.A. Teng Protective and Destructive Immunity in the Periodontium: Part 2--T-cell-mediated Immunity in the Periodontium. J. Dent. Res., March 1, 2006; 85(3): 209 - 219. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Kamoun-Essghaier, I. Guizani, J. M. Strub, A. Van Dorsselaer, K. Mabrouk, L. Ouelhazi, and K. Dellagi Proteomic Approach for Characterization of Immunodominant Membrane-Associated 30- to 36-Kilodalton Fraction Antigens of Leishmania infantum Promastigotes, Reacting with Sera from Mediterranean Visceral Leishmaniasis Patients Clin. Vaccine Immunol., February 1, 2005; 12(2): 310 - 320. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Campos-Neto, R. Porrozzi, K. Greeson, R. N. Coler, J. R. Webb, Y. A. W. Seiky, S. G. Reed, and G. Grimaldi Jr. Protection against Cutaneous Leishmaniasis Induced by Recombinant Antigens in Murine and Nonhuman Primate Models of the Human Disease Infect. Immun., June 1, 2001; 69(6): 4103 - 4108. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
). Proliferation was measured after 6 days with
[3H]thymidine pulse for the last 18 h. Counts of
unstimulated PBMC cultures were in the range of 200-1500 cpm. Number of
IFN-