Lepr Rev (2011) 82, 422– 431

Cytokine responses to Mycobacterium leprae

unique proteins differentiate between
Mycobacterium leprae infected and naı̈ve armadillos

MARIA PENA* , **, ANNEMIEKE GELUK***,

JOLIEN J. VAN DER PLOEG-VAN SCHIP***,
KEES L.M.C. FRANKEN***, RAHUL SHARMA* , ** &
RICHARD TRUMAN*
*Department of Health and Human Services, Health Resources and
Services Administration, Bureau of Primary Health Care, National
Hansen’s Disease Program, Baton Rouge, LA, USA, 70803
**Department of Pathobiological Sciences, School of Veterinary
Medicine, LSU, Baton Rouge, LA, USA, 70803
***Department of Infectious Diseases, Leiden University Medical
Center, Leiden, The Netherlands

Accepted for publication 15 November 2011

Summary New diagnostic tools for early detection of leprosy are necessary to help
reduce its transmission and severity. M. leprae unique proteins have been used to
assess differences in human T-cell responses in leprosy patients, household contacts
and endemic controls. In this study, we examined the response of M. leprae-infected
armadillos to a variety of M. leprae recombinant antigen candidates currently being
examined for diagnostic efficacy in humans. Among recently M. leprae infected
armadillos, IFN-g expression was enhanced after stimulation of PBMC with all
M. leprae recombinant proteins except for ML2283 (mean: 2·65 Relative Quantifi-
cation (RQ)). The group mean stimulation index for M. leprae proteins ML0009,
ML1601, ML2478 and ML2531 averaged 35·2 RQ and was significantly higher
(P , 0·05) than that measured among the non-infected, naı̈ve group (mean 6·2 RQ).
Although ML0840 tended to enhance IFN-g levels, the mean IFN-g transcript levels
of the currently experimentally inoculated group (20·1 RQ) was not significantly
different statistically (P ¼ 0·10) from the mean of the naı̈ve group (7·5 RQ). Also no
statistically significant differences were observed in IFN-g transcript levels between
the resistant and currently experimentally inoculated group (P . 0·05) or between
the resistant and the naı̈ve group (P . 0·05) after stimulation of PBMCs with all
M. leprae recombinant proteins. Only low levels of TNF-a were observed across
all groups after in vitro stimulation with all the antigens examined. These data suggest
that armadillos can be used effectively to help identify M. leprae specific proteins that

Correspondence to: Maria T Pena, NHDP Baton Rouge, Louisiana, USA (e-mail: maria65@lsu.edu)

422 0305-7518/11/064053+10 $1.00 q Lepra

 Cytokine responses by armadillos 423

may be applied for monitoring T-cell responses in M. leprae infected hosts as their
disease progresses as well as for the early diagnosis of leprosy.

Introduction

Although the prevalence of leprosy has decreased worldwide, large numbers of new cases
continue to be detected each year and the disease appears to be perpetuated through transmis-
sion from a large occult reservoir of untreated cases, subclinically infected individuals or
other unknown sources.1 Currently, no effective diagnostic tests are available that can aide
surveillance and detection of individuals who might be incubating the infection or may be
likely to progress in their disease and develop clinical leprosy. Because of the well known
spectrum of immunological responses that different individuals manifest towards
Mycobacterium leprae, development of new diagnostic tests that have good sensitivity and
specificity throughout the clinical spectrum of leprosy has been challenging. Some assays that
detect antibodies to M. leprae-specific molecules, such as phenolic glycolipid-I (PGL-I), can
identify leprosy patients with strong humoral responses to M. leprae. However, these assays
generally have low sensitivity among patients that develop mainly a cell-mediated immune
response (TT/BT) as well as patients at pre- or subclinical stages who are not exhibiting high
antibody titers.
With completion of the M. leprae genome,2 it became possible to identify large numbers
of potential diagnostic antigens and new efforts are being invested in developing molecular
tools for leprosy diagnosis. Among the most promising of these are assays based on the
selection of recombinant proteins and peptides specific for M. leprae that might be combined
in ways to allow detection of disease in its early stages and across the spectrum of leprosy.
Recently two studies simultaneously conducted in Brazil,3,15 used comparative genome
analysis to identify M. leprae-specific recombinant proteins and peptides likely to have high
T-cell recognition. In deployment, they found higher IFN-g production in PBMC assays with
samples from paucibacillary patients and household contacts (HHC) of multibacillary leprosy
patients compared to samples from tuberculosis patients and endemic controls. Similarly,
strong M. leprae-specific serological responses to recombinant proteins among paucibacillary
patients were found with little cross-reactivity among TB cases and endemic controls.5
In order to increase assay specificity Geluk et al.4 applied synthetic peptides spanning the
sequences of M. leprae-unique proteins to analyse induction of IFN-g responses in leprosy
patients, HHC, healthy controls, non-endemic TB patients, and BCG vaccinees. Using the
combined T-cell responses towards four of these peptides all (n ¼ 6) PB patients and 13/14
HHC recognised these peptides without compromising specificity.
Other than man, nine-banded armadillos (Dasypus novemcinctus) are the only natural
hosts of M. leprae. Armadillos are well developed as the hosts-of-choice for in vivo propaga-
tion of M. leprae, and they closely recapitulate many of the most significant pathological
aspects of leprosy as seen in man, such as extensive neurological involvement.6 They also
manifest granulomas to M. leprae which are indistinguishable histopathologically from those
seen in humans. Although the majority of armadillos develop a lepromatous form of leprosy,
the full spectrum of histopathological responses can be observed among armadillos and the
animals can be classified according to the Ridley-Jopling scale from lepromatous (LL) to
tuberculoid (TT) using lepromin skin testing.7 Like humans, the response of individual
armadillos is idiosyncratic, and a reliable proportion of the animals appear to be naturally
424 M. Pena et al.

resistant to experimental infection with M. leprae. Early studies showed that the PGL-I IgM
antibody response of infected armadillos generally correlated with the bacterial load within
the animal’s reticuloendothelial tissues, and could be used to monitor the progress of their
experimentally induced disease. More recently, Duthie et al.8 identified a combination of
antigens that could yield early and accurate diagnosis of experimentally infected armadillos
based on antibody responses. To better inform us about cell mediated cytokine responses
among these unique hosts, we examined the response of M. leprae-infected armadillos to a
variety of recombinant M. leprae proteins currently under evaluation for diagnosis of leprosy
in humans.

Materials and Methods

ANIMALS AND EXPERIMENTAL INOCULATION

A total of 33 armadillos caught in the wild were used in this study. The animals were
conditioned to captivity for a period of 3 to 6 months and screened for pre-existing infections
including seropositivity for PGL-I IgM antibodies. Each animal was lepromin skin tested with
heat-killed armadillo-derived M. leprae prepared as Lepromin-A using 1·6 £ 107 M. leprae in
0·1 ml of normal saline injected intradermally to the abdominal skin. After 21 days the skin test
sites were biopsied using a 4 mm biopsy punch and examined histopathologically.
The reactions were classified according to the Ridley-Jopling scale as described previously.9
Animals that are not seropositive for PGL-I IgM antibodies were considered ‘naı̈ve’.
For in vivo propagation of M. leprae, following conditioning to captivity, naı̈ve animals
(PGL-I negative) were intravenously inoculated with a suspension of 1 £ 109 viable
M. leprae bacteria derived from passage in nude mice.10 The majority of these animals
succumbed to fully disseminated infection within 18 –24 months.11 These animals were
designated ‘infected’. Animals that did not develop disseminated leprosy within 36 months
were considered ‘resistant’ and removed from propagation.9 Most armadillos will begin to
exhibit clinical signs for establishment of M. leprae infection, such as detectable PGL-I IgM
antibodies, within 6 to 12 months post infection. The majority of the animals (n ¼ 30) used in
this study were classified as LL (lepromatous) phenotype by the lepromin skin test, three were
classifed as BL/BT (Table 1).
For this study blood samples from armadillos were tested and classified into three groups:
1. naı̈ve armadillos (n ¼ 8) had been lepromin tested and classified histopathologically, but
had not been experimentally inoculated with viable M. leprae nor showed any detectable
PGL-I IgM antibodies within . 6 months of captivity. One animal was classified BL/BT.
2. Currently experimentally inoculated armadillos (n ¼ 18), that had been incubating
M. leprae infection for less than 36 months at the beginning of this study. Two of these
animals were classified as BL/BT. All of these animals exhibited detectable PGL-I IgM12
antibodies, although the titer for three of them had waned to non-detectable levels during the
course of this study. 3. Resistant armadillos (n ¼ 7) consisted of animals that had been
experimentally inoculated with M. leprae for longer than 36 months and had never shown any
detectable PGL-I IgM antibodies or other signs of systemic dissemination of M. leprae during
that period or since. All studies with animals were conducted in accordance with established
ethical guidelines of the U.S. Public Health Service for the care and use of research animals
under assurance number A3032-1.
Table 1. Comparison of Antibody and CMI responses to candidate M. leprae antigens

IFN-g (RQ) TNF-a (RQ)

ML ML ML ML ML ML ML ML ML ML ML ML
G # Lep PGL-I 0009 0840 1601 2283 2478 2531 ConA 0009 0840 1601 2283 2478 2531 ConA

I 8T14 LL þ 54·5 51·9 20 4·4 29·3 6 182 3·1 3·3 3·5 2·8 3·2 3 3·9
I 8T54 LL þ 24·6 45·1 9 1·7 9·8 5·7 189 9·8 5·6 3·1 1·9 2·6 2·7 2·9
I 7R02 LL þ 2·01 7 4·4 3·1 4·1 2·8 95·4 5·4 2 2 1·3 2·2 3·3 9·1
I 8S16 LL þþ 13·9 6·8 6·6 2·6 12·7 15·4 354 2 1·5 1·5 1·2 1·8 2 2·3
I 8T72 LL þ 4·8 2·5 3·5 2·8 4·1 2·4 28·9 3·4 1·5 1·7 0·7 2 1·9 6·3
I 9B17 LL þþ þ 59·2 26·6 33 2·7 162 140 369 5·7 4·1 3·8 2·7 41 98 6·2
I 9D91 LL þþ þ 17 10·9 20 1·8 14·8 8·6 38·9 5·5 2·8 5 1·6 4·1 3·4 7·3
I 10W21 LL þ 2·3 7·9 0·8 2·3 3·1 14·1 10·4 1·5 3·2 1·5 1·6 1·5 2·5 2·6
I 8U62 LL þ 415 50·7 255 9·2 189 5·8 305 9·4 1·8 3·7 1·4 5·2 1·8 4·3
I 8U75 LL þþ þ 10·9 3·07 10·7 0·9 21·6 6·3 96·8 3·7 1 1·7 0·8 3·1 4·1 5·3
I 8X107 LL þþ 8·2 1·3 5·2 0·9 6·6 1·9 19·9 0·7 1 1·8 1·7 2·4 1·6 4·2
I 8S05 LL þþ 7·5 9·3 15·7 0·5 10·9 2·7 54·9 2·5 1·2 2·3 0·9 1·7 1·9 3·6
I 9B58 LL þ 3·9 0·9 2·3 0·5 2·4 – 6 1·7 0·8 1·1 0·9 1·3 – 2
I 8X117 LL – 85·5 26·4 14·1 3·1 22·7 19·8 74·7 3·7 2·4 2·5 1·7 2·5 2·9 4
I 8X83 LL – 123 82·9 21·3 1·8 16·9 29·2 43·1 5·3 4·9 2·7 1·2 3·1 2·5 2·6
I 8S01 LL – 25·7 14·7 13·5 1·3 21·6 74·9 233 5 2·9 3·7 2·2 3·9 5·4 5·5
I 5R110 BL/BT þþ 7·2 2·8 5·9 1·5 8 3·2 44·1 1·5 1·2 2·1 1·7 1·7 2·3 6·9
I 9W79 BL/BT þþ þ 19·5 10·9 23·2 6·6 174 102 148 2·8 1·8 3·3 1·7 7·7 4·5 6·3
N 6-25 BL/BT – 4·4 5 6·4 2·5 8·2 3·1 65·9 1·9 1·4 1·7 1·4 1·4 2·9 74·6
N 10-32 LL – 8·9 3·3 3·2 4·6 1·9 5·3 89·8 1·1 0·8 0·9 1·3 0·7 0·4 3·5
N 10-31 LL – 2 1·5 1·8 0·6 2·9 1·5 89·8 2·8 1·1 2·2 0·9 1·3 1·3 3·5
N 8-110 LL – 0·1 0·5 0·04 0·05 0·03 0·06 42·4 2·2 1·7 1·2 1·2 1·8 1·2 2·9
N 9-78 LL – 22·6 28·9 8·5 0·7 9·8 12·1 26·1 6·7 3·1 2·1 1·3 2·6 2·7 3·2
N 9-60 LL – – – 5·7 0·2 10 3·9 17·8 – – 2·3 0·5 2·5 2·7 9·7
N 9.105 LL – 12·5 – 21·4 5·6 13·2 5·6 388 1·8 0·7 1·2 1·2 1·5 0·9 18·9
N 10-700 LL – 7·6 – 0·2 9·8 2·3 – 41·6 5·2 0·8 0·2 0·5 1·9 4·9 5·8
R 3R07 LL – 21·8 33·1 28·6 5·2 16 16 349 4·9 1·9 3·4 1·5 3·2 3·8 3·8
R 6R04 LL – 3·7 6·3 5·6 4·5 2 2·3 11·2 2·1 1·7 1·8 1·8 1·8 1·3 1·8
R 6L35 LL – 42·9 70·4 41·6 23·3 31·7 5·6 8·6 2·9 1·7 2·1 1·5 1·5 2·6 2·7
R 5L59 LL – 10·6 5 17·8 3 11·2 27·9 174 4·3 0·8 1·2 1·3 3·4 1·3 9·2
R 7R65 LL – 119 70·4 25·1 2·9 14·2 17·2 30 1·4 1·2 1·2 1·8 1·2 0·7 3·3
R 3V76 LL – 3·3 3·1 3·3 0·9 2·4 3 70·9 0·2 0·3 0·2 0·2 0·2 0·3 1·5
R 3X119 LL – 0·2 0·2 0·2 0·05 0·2 0·1 21·5 3·5 3·3 2·7 1·8 3·3 3·3 0·4
Cytokine responses by armadillos

G: group, I: Infected, N: Naive and R: Resistant. #: Animal ID, Lep: lepromin test, PGL-I: þ þþ (.0·1 OD), þ þ(0·8–0·1 OD), þ (0·7– 0·8 OD), – (undetectable). High,
mid and low positive by ELISA. Comparison of antibody and CMI responses to candidate M. leprae antigens. Results of in vitro stimulation of armadillos PBMCs with M.
425

leprae recombinant proteins are expressed as Relative Quantification (RQ) and indicates fold increase of transcript levels relative to the control stimulus (medium alone).
426 M. Pena et al.
RECOMBINANT M. LEPRAE PROTEINS

The genes for ML0009, ML0840, ML1601, ML2283, ML2478 and ML2531 were amplified
by PCR from genomic DNA of M. leprae (generously provided by the late Dr. M.J. Colston)
and cloned using the Gateway technology platform (Invitrogen, Carlsbad, CA, USA) with
pDEST17 expression vector containing an N-terminal histidine tag (Invitrogen, Carlsbad,
CA, USA).13 Sequencing was performed on selected clones to confirm identity of all cloned
DNA fragments. Recombinant proteins were overexpressed in E. coli BL21(DE3) and
purified as described to remove any traces of endotoxin. The purified protein was analysed by
12% SDS-PAGE followed by Coomassie Brilliant Blue staining and Western-blotting with
an anti-His antibody (Invitrogen, Carlsbad, CA, USA) to confirm size and purity. Endotoxin
contents were below 50 IU per mg recombinant protein as tested using a Limulus Amebocyte
Lysate (LAL) assay (Cambrex, East Rutherford, NJ, USA). To exclude protein non-specific
IFN-g release or cellular toxicity ML2531 was tested in IFN-g release assays using PBMC
or whole blood of M. leprae-unexposed, BCG-negative, Mantoux skin test negative healthy
donors. For each antigen only one batch preparation was used.

IN VITRO ASSAYS

Armadillo peripheral blood mononuclear cells (PBMCs) were purified from 8 ml peripheral
blood collected in BD Vacutainerw CPT Mononuclear Cell Preparation Tubes (BD
Biosciences, San Jose, CA) and mononuclear cells were isolated after centrifugation
(1600 £ g for 45 minutes, 25 8C). The mononuclear cell layer was removed, washed three
times with cold PBS and resuspended in culture medium (RPMI 1640 medium containing
2 mM glutamine and HEPES) supplemented with 10% fetal bovine serum (FBS). 100 ml of
the cell suspension (4 £ 106 cells/ml) was added to a 96 well round bottom tissue culture
plate containing 100 ml complete media with 2ME and the antigens (20 ml) and the plate was
incubated for 18 hours at 33 8C in 5% CO2 humidified atmosphere. Six different M. leprae
recombinant proteins (ML0009, ML0840, ML1601, ML2283, ML2478 and ML2531) at a
final concentration of 10 mg/ml were used for in vitro stimulation. Duplicate wells containing
concanavalin A (Con A 40 mg/ml) (Sigma, St. Louis, MO) or media alone were used as
positive control for cytokine production or indicator of non-stimulation, respectively.
After overnight incubation the plate was centrifuged and the supernatants removed. PBS was
added to the cell pellet followed by a second centrifugation and 350 ml of RLT buffer with
2ME (10 ml/mL) was added and then transferred to a 1·5 ml centrifuge tube which was
stored at 2 70 8C for RNA isolation. RNA was isolated using RNeasyw Plus Mini kit
(Qiagen,Valencia, CA) according to the manufacturer’s recommendations.

RT-PCR AND RELATIVE QUANTIFICATION

D. novemcinctus IFN-g cDNA sequence (GI: DQ094083) was obtained from NCBI and
primer sequences constructed as previously described.14 Bioinformatic tools were used
to identify the putative nucleotide sequence of Armadillo TNF-a. H. sapiens TNF-a
nucleotide sequence of (GI:224589818) was used to search for homologous sequences in the
D. novemcinctus whole genome sequence (WGS) trace files (6X) using BLASTn (http://
www.ncbi.nlm.nih.gov/BLAST/). Consensus sequence from three different trace files was
obtained. Armadillo specific gene specific Primers (TNF-a F: 50 - TGGTGCCTCAGCCT-
 Cytokine responses by armadillos 427
0 0 0 0
CTTCTC-3 TNF-a R: 5 - GCCGATCACCCCAAACTG-3 , IFN- gF: 5 -GAATTACAC-
GGGCTATCTCTTAGCTT-30 & IFN- g R: 50 - AAGGTCGGCCTGGCAGTAG-3) and
probes (TNF-a: CCACCGCGCTTTTCTGCCTGC and IFN-g:TCAGCTTTGCATCATTT-
TGGGTTCTTCTAGC) were designed by using Primer Expressw Software v3.0 (Applied
Biosystems, Carlsbad, CA). Briefly, cDNA was made from RNA using Advantage RT for
PCR kit (Clontech Laboratories, Mountain View, CA) according to manufacturer recom-
mendations. Real time PCR was run for 40 cycles 15 seconds at 95 8C and 1 minute at 60 8C in
ABI 7300/7500 (Applied Biosystems, Foster City, CA) Real Time PCR system. Relative
expression of these cytokines was analysed in cDNA from PBMC treated with different
M. leprae recombinant proteins and control (medium alone) using the DDCt method,13 where
GAP3DH served as a normaliser. The results are expressed as fold change increase relative
to the control stimulus (medium alone).

STATISTICAL ANALYSIS

Differences in cytokine responses after in-vitro stimulation of PBMCs from Infected,

Resistant and Naives armadillos with M. leprae recombinant proteins were analysed with
unpaired t test using Graph Pad software (version 3.0) (GraphPad Software, Inc, La Jolla, CA)
and P values , 0·05 were considered statistically significant and compared to linear regres-
sion using SigmaPlot (11·0).

Results

In this study, currently experimentally inoculated armadillos recognised five out of six
recombinant M. leprae proteins examined (ML0009, ML0840, ML1601, ML2478 and
ML2531) by producing high IFN-g responses (Figure 1).
Within this group, PBMCs from 10 armadillos showed higher IFN-g transcript levels
whereas PBMCs from eight armadillos showed lower IFN-g production after stimulation
with five out of six recombinant proteins. When we examined T-cell reactivity to M. leprae
antigens versus anti-PGL-I antibody levels, we found that three out of 10 armadillos that
showed high T-cell responses had low anti-PGL-I antibody responses. Although the trend
was not significant statistically, 8/8 armadillos that showed low IFN-g responses had high
anti-PGL-I antibody titer. Despite the variability of the response to M. leprae recombinant
proteins among the animals in the currently experimentally inoculated group, the mean IFN-g
transcript production in this group was significantly higher than that of the naı̈ve group for
ML0009 (P ¼ 0·04), ML1601 (P ¼ 0·02), ML2478 (P ¼ 0·01) and ML2531 (P ¼ 0·03). No
statistically significant differences (P ¼ 0·10) in IFN-g responses were seen for ML0840
between the currently experimentally inoculated (20·1 RQ) and the naı̈ve group (7·5 RQ).
Also no statistically significant differences were observed in IFN-g responses between the
resistant and currently experimentally inoculated group (P . 0·05) or between the resistant
and the naı̈ve group (P . 0·05) after stimulation of PBMCs with all M. leprae recombinant
proteins. For ML2283, little or no IFN-g upregulated expression was seen after stimulation of
PBMCs among all the test groups. Linear regression analysis between PGL-I as independent
variable and RQ as dependent variable resulted in no association between these variables for
all the antigens tested.
In general, M. leprae recombinant proteins did not elicit enhanced TNF-a transcript
production in PBMCs of armadillos (Figure 2).
428 M. Pena et al.

IFN-γ response to M. leprae recombinant proteins in infected-,

naïve- and resistant armadillos
P = 0·33
Fold increase relative to

P = 0·78
P = 0·04 P = 0·51
200 200 P = 0·10 P = 0·42
control (media)V

150 150

100 100

50 50

0 0

Naïve Infected Resistant Naïve Infected Resistant

ML0009 ML0840

P = 0·64 P = 0·64
P = 0·02 P = 0·22 200
200
Fold increase relative to

P = 0·26 P = 0·50
control (media)V

150 150

100 100

50 50

0 0

Naïve Infected Resistant Naïve Infected Resistant

ML1601 ML2283

P = 0·08
P = 0·23
Fold increase relative to

200 P = 0·01 P = 0·55 200

P = 0·03 P = 0·5
control (media)

150 150

100 100

50 50

0 0

Naïve Infected Resistant Naïve Infected Resistant

ML2478 ML2531
Infected armadillos with PGL-I values that waned BL/BT armadillos
Figure 1. IFN-g mRNA expression in response to M. leprae recombinant proteins in naı̈ve, infected (experimentally
M. leprae inoculated with fully disseminated leprosy) and resistant (experimentally inoculated without disease)
armadillos. IFN-g mRNA expression was analyzed after overnight in vitro stimulation of PBMCs of infected
(n ¼ 18), naı̈ve (n ¼ 8) and resistant (n ¼ 7) armadillos with M. leprae proteins. Results are expressed as Relative
Quantification (RQ) and indicates fold increase of transcript levels relative to the control stimulus (medium alone).

The currently experimentally inoculated group showed significantly higher TNF-a

expression after stimulation of PBMCs with ML1601 (P ¼ 0·007), and ML2478 (P ¼ 0·04)
when compared to the naı̈ve group. Additionally, the currently experimentally inoculated
group showed a significantly higher production of TNF-a transcript after stimulation ML1601
(P ¼ 0·04) than the resistant group. Animals in the currently experimentally inoculated group
of which PGL-I IgM antibody titers had waned showed the highest overall responses in the
 Cytokine responses by armadillos 429

TNF-α response to M. leprae recombinant proteins in infected-,

naïve- and resistant armadillos
P = 0·11
P = 0·18
14 P = 0·05 P = 0·86
Fold increase relative to

P = 0·43 P = 0·59
14
12
12
control (media)

10 10
8 8
6 6
4 4
2 2
0 0

Naïve Infected Resistant Naïve Infected Resistant

ML0009 ML0840

P = 0·4
P = 0·04
P = 0·027 P = 0·55
P = 0·007 P = 0·86 14
Fold increase relative to

14
12 12
control (media)

10 10
8 8
6 6
4 4
2 2
0 0

Naïve Infected Resistant Naïve Infected Resistant

ML1601 ML2283

P = 0·09 P = 0·05
P = 0·04 P = 0·99 >40 P = 0·08 P = 0·68
>40
Fold increase relative to

14 14
control (media)

12 12
10 10
8 8
6 6
4 4
2 2
0 0

Naïve Infected Resistant Naïve Infected Resistant

ML 2478 ML 2531
Infected armadillos with PGL-I values that waned BL/BT armadillos

Figure 2. TNF-a mRNA expression in response to M. leprae recombinant proteins in naı̈ve, infected (experimentally
M. leprae inoculated with fully disseminated leprosy) and resistant (experimentally inoculated without disease)
armadillos. TNF-a mRNA expression was analysed after overnight in vitro stimulation of PBMCs of infected
(n ¼ 18), naı̈ve (n ¼ 8) and resistant (n ¼ 7) armadillos with M. leprae proteins. Results are expressed as Relative
Quantification (RQ) and indicates fold increase of transcript levels relative to the control stimulus (medium alone).

group. The currently experimentally inoculated armadillos showed a higher IFN-g transcript
levels than naı̈ve armadillos after in vitro stimulation of PBMCs with most M. leprae
recombinant proteins tested here. Monitoring the cell mediated immune response towards
these M. leprae proteins may be an effective indicator of recent exposure to M. leprae.
430 M. Pena et al.

Discussion

These data indicate that specific M. leprae recombinant proteins can elicit strong IFN-g
responses in experimentally infected armadillos. Although these animals generally develop a
multibacillary-type of disease, T-cell based assays can be used to differentiate exposure status
to M. leprae among armadillos and may contribute to longitudinal monitoring as this
infection progresses. Armadillos are natural hosts of M. leprae which have extensive
exposure to potential cross-reactive antigens from soil-borne organisms and environmental
mycobacteria. Their response to the M. leprae -recombinant proteins tested here appeared to
be highly specific and it seems likely that armadillos can be used to help develop and evaluate
new T-cell based diagnostic assays for human leprosy.
A previous study that also used M. leprae recombinant proteins demonstrated that two
M. leprae specific antigens, ND-O-BSA and LID-1 evoked antibody responses in experi-
mentally infected armadillos. In the current study, we examined T-cell responses to
recombinant M. leprae proteins among animals currently experimentally inoculated with
M. leprae and progressing in their experimentally induced disease (infected) animals which
had been experimentally infected at least 3 years before but had resisted M. leprae (resistant)
and naı̈ve animals not yet infected with M. leprae (naı̈ve). The currently experimentally
inoculated group showed significantly higher IFN-g responses (P # 0·04) than the naı̈ve
group. When we examined T-cell reactivity to M. leprae antigens versus anti-PGL-I antibody
levels we found that all the armadillos with low IFN-g responses had high anti-PGL-I
antibody titers and three out of 10 animals with high IFN-g responses had low anti-PGL-I
antibody titers. Likewise, other studies3,4 – 16 have observed that paucibacillary leprosy
patients and household contacts who have low humoral responses tended to evoke higher
T-cell responses to M. leprae specific antigens. These findings and previous work5,15,4
suggest that M. leprae-unique antigens can be used to identify M. leprae infection exposure
in individuals in absence of detectable levels of anti-PGL-I antibodies and when used in
combination with other tests that index the humoral responses, such as PGL-I IgM antibodies,
may be an effective contribution to aid the early diagnosis of leprosy.
Armadillos showed varying responses to M. leprae -recombinant proteins and additional
longitudinal studies are needed in order to fully assess the cytokine response patterns that
may be associable with different disease states. Within the recombinant proteins we tested,
ML0840 and ML1601 are known to have homologs in M. avium,15 an environmental
mycobacterium that armadillos are commonly exposed to in nature. Although ML1601 is
present in other mycobacterial species, it was recognised here only by the currently experi-
mentally inoculated group but not by the naı̈ve group indicating that this protein is mainly
detecting an M. leprae response and that it can be considered together with ML0009, ML2478
and ML2531 for leprosy diagnosis purposes. Interestingly, ML2283, which is unique to
M. leprae, and appears to induce significant IFN-g levels in PB- and reactional leprosy
patients15,16 but not MB patients, was the only protein examined that did not induce IFN-g
transcript in armadillo PBMC cultures. In contrast to its potential for human disease, ML2283
may not be appropriate for diagnostic use in armadillos and differential recognition of
antigens can be an important limitation when using animal models. Alternatively, this could
be due to the method of analysing the IFN-g response as for humans standard ELISA were
used after PBMC stimulation for 6 days.15,3
These data show that M. leprae-unique proteins selected on the basis of their specific
recognition by T cells derived from M. leprae exposed humans also specifically produce
 Cytokine responses by armadillos 431

IFN-g transcript in M. leprae infected armadillos. Major variations in individual responses to

these antigens were observed among animals at varying stages of infection, and longitudinal
monitoring of their response over the course of experimental infection in the armadillo may
be informative for the development of improved diagnostic assays for human leprosy.

Acknowledgements

The authors thank Roena Stevenson and Vilma Marks for helping to obtain and process
the armadillo samples. This study was supported by the Netherlands Leprosy Relief
Foundation (NLR) ILEP#: 702.02.65 and ILEP#: 701.02.49, The Turing Foundation and the
Q.M. Gastmann-Wichers Foundation. LUMC and NHDP are part of the IDEAL (Initiative
for Diagnostic and Epidemiological Assays for Leprosy) Consortium. The U.S. Health
Resources and Services Administration, and the National Institutes of Allergy and Infectious
Disease IAA-2646. The authors are independent from the funders.

References
1
Truman R, Fine PE. ‘Environmental’ sources of Mycobacterium leprae: issues and evidence. Lepr Rev, 2010; 81:
89–95.
2
Cole ST, Eiglmeier K, Parkhill J et al. Massive gene decay in the leprosy bacillus. Nature, 2001; 409(6823):
1007–1011.
3
Spencer JS, Dockrell HM, Kim HJ et al. Identification of specific proteins and peptides in Mycobacterium leprae
suitable for the selective diagnosis of leprosy. J Immunol, 2005; 175: 7930–7938.
4
Geluk A, van der Ploeg J, Teles RO et al. Rational combination of peptides derived from different Mycobacterium
leprae proteins improves sensitivity for immunodiagnosis of M. leprae infection. Clin Vaccine Immunol, 2008;
15: 522–533.
5
Spencer JS, Kim HJ, Wheat WH et al. Analysis of antibody responses to Mycobacterium leprae phenolic
glycolipid I, lipoarabinomannan, and recombinant proteins to define disease subtype-specific antigenic profiles in
leprosy. Clin Vaccine Immunol, 2011; 18: 260– 267.
6
Scollard DM, Lathrop GW, Truman RW. Infection of distal peripheral nerves by M. leprae in infected armadillos;
an experimental model of nerve involvement in leprosy. Int J Lepr Other Mycobact Dis, 1996; 64: 146 –151.
7
Job CK, Kirchheimer WF, Sanchez RM. Tissue response to lepromin, an index of susceptibility of the armadillo
to M. leprae infection – a preliminary report. Int J Lepr Other Mycobact Dis, 1982; 50: 177–182.
8
Duthie MS, Truman RW, Goto W et al. Insight toward early diagnosis of leprosy through analysis of the
developing antibody responses of Mycobacterium leprae-infected armadillos. Clin Vaccine Immunol, 2011;
18: 254–259.
9
Job CK, Truman RW. Mitsuda-negative, resistant nine-banded armadillos and enhanced Mitsuda response to live
M. leprae. Int J Lepr Other Mycobact Dis, 1999; 67: 475–477.
10
Truman RW, Krahenbuhl JL. Viable M. leprae as a research reagent. Int J Lepr Other Mycobact Dis, 2001;
69: 1–12.
11
Truman RWaS RM. Armadillos: models for leprosy. Lab Animal, 1990; 22: 28– 32.
12
Truman RW, Job CK, Hastings RC. Antibodies to the phenolic glycolipid-1 antigen for epidemiologic
investigations of enzootic leprosy in armadillos (Dasypus novemcinctus). Lepr Rev, 1990; 61: 19–24.
13
Franken KLCM, Hiemstra HS, van Meijgaarden KE, Subronto Y, den Hartigh J, Ottenhoff TH, Drijfhout JW.
Purification of his-tagged proteins by immobilized chelate affinity chromatography: the benefits from the use of
organic solvent. Protein Expr. Purif., 2000; 18: 95 –99.
14
Pena MT, Adams JE, Adams LB et al. Expression and characterization of recombinant interferon gamma
(IFN-gamma) from the nine-banded armadillo (Dasypus novemcinctus) and its effect on Mycobacterium leprae-
infected macrophages. Cytokine, 2008; 43: 124 –131.
15
Geluk A, Klein MR, Franken KL et al. Postgenomic approach to identify novel Mycobacterium leprae antigens
with potential to improve immunodiagnosis of infection. Infect Immun, 2005; 73: 5636–5644.
16
Geluk A, van der Ploeg-van Schip JJ, van Meijgaarden KE et al. Enhancing sensitivity of detection of immune
responses to Mycobacterium leprae peptides in whole-blood assays. Clin Vaccine Immunol, 2010; 17: 993 –1004.