Journal of Neuropathology & Experimental Neurology Advance Access published February 17, 2016

J Neuropathol Exp Neurol

Vol. 0, No. 0, pp. 1–12
doi: 10.1093/jnen/nlv027

ORIGINAL ARTICLE

Inflammatory Cytokines Are Involved in Focal Demyelination

in Leprosy Neuritis
Priscila Ribeiro Andrade, PhD, Marcia Rodrigues Jardim, MD, PhD,
Ana Caroline Costa da Silva, PhD, Paula Saraiva Manhaes, MD,
Sérgio Luiz Gomes Antunes, MD, PhD, Robson Vital, MD, PhD, Rhana Berto da Silva Prata, BSc,
Rafael Braga Petito, PhD, Roberta Olmo Pinheiro, PhD, and Euzenir Nunes Sarno, MD, PhD

Abstract serious sensory and motor dysfunctions that lead to the devel-

Downloaded from http://jnen.oxfordjournals.org/ by guest on February 21, 2016

Mycobacterium leprae (ML) infection causes nerve damage that of- opment of permanent deformities and/or disabilities (1). The
ten leads to permanent loss of cutaneous sensitivity and limb deformi- etiological agent of leprosy, Mycobacterium leprae (ML), is
ties, but understanding of the pathogenesis of leprous neuropathy that an obligatory intracellular pathogen that preferentially infects
would lead to more effective treatments is incomplete. We studied macrophages and Schwann cells (SCs) (2, 3).
reactional leprosy patients with (n ¼ 9) and without (n ¼ 8) acute neuri- Leprosy remains a major health challenge due to the pe-
tis. Nerve conduction studies over the course of the reactional epi- ripheral nerve damage and irreversible physical deformities it
sode showed the findings of demyelination in all patients with neuritis. causes (4). Despite advances in knowledge regarding the path-
Evaluation of patient sera revealed no correlation of the presence of ogenesis of the leprosy spectrum, understanding of the mecha-
antibodies against gangliosides and the clinical demyelination. In nerve nisms involved in nerve damage and regeneration in leprosy-
biopsies of 3 patients with neuritis, we identified tumor necrosis factor associated neuropathy remains incomplete. The natural affinity
(TNF), TNF receptors, and TNF-converting enzyme in Schwann cells of ML for peripheral nerves, particularly for SCs, makes it
(SCs) using immunofluorescence. To elucidate immunopathogenetic likely that leprosy nerve damage starts at a very early stage of
mechanisms, we performed experiments using a human SC line. ML infection but the mechanisms underlying nerve damage in
induced transmembrane TNF and TNF receptor 1 expression in the early disease have not been elucidated (5). Nerve conduction
SCs; TNF also induced interleukin (IL)-6 and IL-8 production by the studies (NCS) suggest the existence of a defined disease pro-
SCs; and ML induced IL-23 secretion, indicating involvement of this gression pattern despite clinically undetected alterations (6).
previously unrecognized factor in leprosy nerve damage. These data There are, however, a number of controversies in the literature
suggest that ML may contribute to TNF-mediated inflammation and regarding the most highly affected parameters and the preva-
focal demyelination by rendering SCs more sensitive to TNF within lence and evolution of different types of nerve lesions (7).
the nerves of patients with leprous neuropathy. Studies in experimental animal models and with neuron
Key Words: Demyelination, IL-23, Leprosy, Mycobacterium lep-
and SC cocultures indicate that nonmyelinated axonal units
rae, Neuropathy, Schwann cell, Tumor necrosis factor.
are more susceptible to ML infection than myelinated ones,
rendering the former the main target of this mycobacterium
in the PNS (8). It has recently been reported that SCs in mice
undergo genetic reprogramming upon long-term ML infec-
INTRODUCTION tion and that they assume a state of progenitor/stem-like cells
Leprosy is an infectious disease characterized by skin uncommitted to the SC lineage (9). Moreover, ML can induce
lesions and peripheral nerve damage, which often results in a large number of immune-related genes during the early
stages of infection, rendering SC capable of eliciting an in-
flammatory response in the tissue and thus contributing to
From the Leprosy Laboratory, Oswaldo Cruz Institute, FIOCRUZ, Rio de Ja- nerve damage (10, 11).
neiro, Brazil (PRA, MRJ, ACCdS, SGA, RV, RBdSP, RBP, ROP, ENS); Loss of the myelin sheath is among the most damaging
and Pedro Ernesto University Hospital, Rio De Janeiro State University, effects of nerve injury in many human diseases, and progres-
Rio de Janeiro, Brazil (MRJ, PSM). sive demyelination can lead to permanent neurological dys-
Send correspondence to: Euzenir Nunes Sarno, MD, PhD, Leprosy Laboratory–
Oswaldo Cruz Foundation, Av. Brasil 4365-Manguinhos, Rio de Janeiro function (12). The identification of antiganglioside antibodies
21045-900, Brazil; E-mail: [email protected] that target neural molecules, including myelin itself, in pa-
This work was supported by FAPERJ (Fundação de Amparo à Pesquisa do tients with peripheral neuropathies such as Guillain-Barré
Estado do Rio de Janeiro) and CNPq (Conselho Nacional de Desenvolvi- syndrome have been described (13). However, there are no
mento Cientı́fico e Tecnologico).
The authors declare no financial or commercial conflicts of interest.
data linking this mechanism to the nerve injury observed in
Supplementary Data can be found at http://www.jnen.oxfordjournals.org. leprosy patients.

C 2016 American Association of Neuropathologists, Inc. All rights reserved

V 1
Andrade et al J Neuropathol Exp Neurol  Volume 0, Number 0, Month 2016

On the other hand, experimental evidence from both in included in the study (Table 1). All of the patients had neuro-
vivo and in vitro studies point to the involvement of immune logical and neurophysiological examinations to diagnose lep-
mediators such as tumor necrosis factor (TNF) in the immuno- rous neuropathy. They were then sorted into 2 groups: pa-
pathogenesis of several inflammatory demyelinating disorders tients with neuritis and those without neuritis (Table 1).
in the PNS and CNS (14). This cytokine is one of the first medi-
ators to appear subsequent to experimental peripheral nerve
damage (15, 16). High levels of TNF in neurological disorders
have been shown to promote demyelination, axonal degenera- Clinical Evaluation
tion, increased nerve blood barrier permeability, and immune A detailed neurological examination was performed to
cell recruitment to the injury site (17). Moreover, TNF has been record the number and distribution of affected nerves. The
detected in the dermis and epidermis of leprosy reactional skin neurological examination evaluated the following: the motor
lesions and in the sera of reactional patients (18–21). In patients strength and tactile sensation of large myelinated nerve fi-
with reverse reaction and the pure neuritic form of the disease, bers; thermal and pain sensation; the presence of erythrocya-
an increase in the mRNA of this cytokine was observed in the nosis on the palms and/or soles; paresthesia; and nerve pain.
peripheral nerves, suggesting that TNF plays an important role Sensory impairment, motor deficit, and disability/deformity
in the pathogenesis of leprosy nerve injury (22). status were assessed using standard methods. In brief, the tac-
The lack of knowledge concerning the molecular and tile threshold was tested via Semmes-Weinstein monofila-
immunological mechanisms of ML-induced nerve damage in ments (25, 26). Thermal sensation was determined by the use
of cold metal (15 C) objects, and a safety pin was utilized to

Downloaded from http://jnen.oxfordjournals.org/ by guest on February 21, 2016

leprosy precludes the development of effective therapies for
leprosy neuropathy. Although immunosuppression does not ascertain pain perception. Disability was recorded in accor-
prevent nerve damage, corticosteroids are able to curb in- dance with the standard World Health Organization grading
flammation-associated symptoms such as pain. criteria (27).
The existence of a National Reference Center for Lep-
rosy Neuropathy allowed us to select a group of reactional
patients with and without acute neuritis in an attempt to clarify Neurophysiological Evaluation
the potential mechanisms involved in the immunopathogenesis The NCS was performed using the Nihon-Kohden ap-
of nerve lesions in leprosy patients. We also evaluated the roles paratus using standard procedures (28). Amplitude, velocity,
of serum antiganglioside antibodies and of TNF, including and latency were recorded for the median, radial, ulnar, and
their possible applications as predictive markers. sural sensory nerves in addition to the median, ulnar, and pe-
roneal motor nerves. The results of the conduction studies
MATERIALS AND METHODS were used to identify: (1) normal nerves, (2) axonal lesions,
which were defined by a reduction in compound muscle ac-
Patients tion potentials (CMAP), and/or sensory nerve action poten-
The study was conducted at the Leprosy Laboratory and tials (SNAP), with the amplitude being less than 30% of the
Leprosy Out-Patient Unit of the Oswaldo Cruz Foundation in reference value and/or the maximum conduction velocity
Rio de Janeiro, RJ, Brazil. Reactional leprosy patients were (MCV) being above 70% of the reference value (7), (3) de-
selected and classified according to the Ridley-Jopling scale myelination lesions, which were defined in cases when the
(23). All patients provided their informed written consent, CMAP and/or SNAP latency were prolonged in comparison
and the acquisition of all specimens was approved by the to the reference together with a reduction in sensory conduc-
Human Ethics Committee of the Oswaldo Cruz Foundation. tion velocity and/or an MCV below 85% of the reference
Over a 1-year period, patients diagnosed with any type value, (4) mixed lesions, which were defined when both axo-
of leprosy reaction were evaluated for neurological complica- nal and demyelinating lesions were detected in the same
tions. Only those who presented leprosy-related neuropathy nerve, and (5) no conduction, defined as an increase in the du-
and no contraindications for steroid treatment were included ration of the proximal distal CMAP of over 30% (29).
in the study.
For this work, reactions were defined as: (1) reverse re-
action (ie, increased inflammation of existing lesions with or
without nontender new lesions and/or acroedema that could Histopathological Analysis
be associated or not with neuritis (24)), (2) erythema nodo- The sural nerves of 3 patients were biopsied for histo-
sum leprosum (ie, the sudden appearance of inflamed pap- pathological diagnosis. These biopsies were performed be-
ules, nodules, and plaques that are tender upon palpation). In cause there was a suspicion of pure neural leprosy or another
this case, patients may be ill with inflammatory systemic superimposed neuropathy in addition to leprosy. These patients
symptoms associated or not with neuritis, and (3) neuritis (ie, had a persistent and prolonged course of reactional episodes,
one or more nerves may be enlarged, painful, or show loss of refractoriness to antireactional treatment and leprosy relapse.
function). Neuritis may or might not be accompanied by skin The samples were submitted for routine histopathological pro-
lesions, in which case it is referred to as “isolated neuritis” cessing followed by staining with hematoxylin and eosin and
(7). Wade stain for detection of acid-fast bacilli; 500-lm-thick sec-
Out of the 35 leprosy patients diagnosed as having a tions were then stained with toluidine blue and examined in de-
reaction within the year, 17 were found eligible and were tail for detection of fine nerve fiber alterations.

2
J Neuropathol Exp Neurol  Volume 0, Number 0, Month 2016 Cytokines and Demyelination in Leprous Neuritis

TABLE 1. Clinical Data

Patient No. Clinical Form Reaction Multidrug therapy Age Gender Degree Mean Bacilloscopy
(Years) of Disability Index
Patients without neuritis
1 LL ENL During 24 M 0 5.0
2 BL RR During 39 F 0 3.2
3 BT RR After 67 F 1 0
4 LL RR During 30 M 1 5.0
5 BT RR After 51 F 0 0
6 BB RR During 56 F 1 0.5
7 BT RR After 57 F 0 0
8 BL RR During 66 M 1 0
Patients with neuritis
9 BT Neuritis After 33 M 0 0
10 BL RR þ Neuritis During 66 M 1 1.5
11 BL Neuritis During 57 M 1 2.5
12 BL Neuritis After 19 M 0 4.25

Downloaded from http://jnen.oxfordjournals.org/ by guest on February 21, 2016

13 LL ENL þ Neuritis After 34 M 2 5.0
14 BT RR þ Neuritis During 58 M 2 0
15 BB RR þ Neuritis During 55 F 1 0.75
16 LL ENLþ Neuritis After 75 M 1 2.75
17 BT Neuritis During 54 M 1 0
ENL, erythema nodosum leprosum reaction; F, female; M, male; RR, reverse reaction.

Schwann Cell Line and Culture In Vitro Assay Conditions

The human SC line ST8814 isolated from a malignant For TNF protein expression evaluation, ST8814 SCs
schwannoma of a patient with neurofibromatosis 1 (30) was were cultured in 12-well plates at 37 C/5% CO2 and stimulated
generously donated by Dr J.A Fletcher (Dana Farber Cancer with irradiated ML for 3, 5, and 7 hours. Whole-cell extracts
Institute, Boston, MA). The cells were cultured in RPMI media were obtained for Western blot and enzyme-linked immuno-
(Gibco BRL, Grand Island, NY) supplemented with 100 U/mL sorbent assay (ELISA) analyses. TNF surface levels were in-
of penicillin, 100 mg/mL of streptomycin, 2 mM of L-glutamine, vestigated by immunofluorescence after the cells were cultured
and 15% fetal bovine serum (Hyclone Laboratories, South Lo- on glass coverslips in 24-well plates and stimulated with irradi-
gan, UT) in a humidified CO2 incubator at 378C. For experi- ated ML for 3 and 7 hours. Phenotypic characterization of the
mental assays, the cells were detached from the culture bottles ST8814 line was performed in the same way in the absence of
via trypsin/EDTA (0.25%, 1 mmol/L) (Gibco BRL), washed, stimuli. TNF secretion was evaluated through analysis of the
suspended in complete medium, and cultured in plates as needed culture supernatants after stimulation of ST8814 SCs seeded in
for in vitro assays. 12-well plates with irradiated ML for 3, 6, and 24 hours by
Western blot and ELISA.
TNFR1 and TNFR2 gene expressions were measured
Reagents and Stimuli in ST8814 SCs cultured in 6-well plates and stimulated with
Irradiated armadillo-derived ML were provided by Dr RhTNF and irradiated ML associated or not for 24 hours. The
P. Brennan (Department of Microbiology, Colorado State RNA of these cells was obtained for semiquantitative reverse
University, Fort Collins, CO) and used at a dosage (lg/mL) transcriptase-polymerase chain reaction (RT-PCR) analysis,
corresponding to a multiplicity of infection of 50 bacteria:cell and the culture supernatants were evaluated by ELISA to as-
in each experiment. The endotoxin level, measured using a lim- sess inflammatory cytokine secretion.
ulus amoebocyte lysate assay, was found to be below 50 IU/mg
(Cambrex, East Rutherford, NJ) . Recombinant human TNF
(RhTNF) (Calbiochem, Merck-Millipore, Darmstadt, Germany) Western Blot
was used in a 25 ng/mL concentration. Soluble and transmembrane TNF expression was as-
For protein expression analyses, the following mono- sessed by Western blot analysis. Whole-cell extracts were ob-
clonal antibodies (mAbs) were purchased: anti-TNF (mAb tained after detachment of ST8814 SCs from the culture
210, R&D Systems, Minneapolis, MN); anti-TNF receptor 1 plates. The cells were washed twice with ice-cold phosphate-
(TNFR1) (mAb 225, R&D Systems); anti-TNF receptors 2 buffered saline (PBS) and incubated for 30 minutes with lysis
(TNFR2) (mAb 226, R&D Systems); anti-TNF converting buffer and protease inhibitor cocktail II (Calbiochem), as de-
enzyme (TACE, mAb 9301, R&D Systems); and anti-CD44 scribed elsewhere (31). Proteins (30 lg) from extracts and
(M7082, Dako, Glostrup, Denmark). Polyclonal anti-S100 culture supernatants were resolved in 12% sodium dodecyl
antibody (Z0311, Dako) was also obtained. sulfate polyacrylamide gels and blotted onto nitrocellulose
3
Andrade et al J Neuropathol Exp Neurol  Volume 0, Number 0, Month 2016

membranes (Bio-Rad Laboratories, Hercules, CA). Membranes GTCGTGTTGGAGAACG-30 (Applied Biosystems, Foster
containing supernatant samples were stained with Ponceau S for City, CA). The primers were designed to avoid genomic DNA
protein-loading evaluation. After blocking with a solution of 5% amplification. PCR was performed, as previously described
bovine serum albumin (BSA) in tris-buffered saline (TBS) (33), and the samples were amplified in a DNA thermocycler
0.15% Tween (TBS-T), blots were incubated for 1 hour at room 2400 (Perkin Elmer, Waltham, MA). PCR products were sub-
temperature with anti-TNF (0.3 lg/mL). Blots with the extracts jected to electrophoresis in 1.7% agarose gels at which time the
were also probed for a-tubulin (T6074, Sigma-Aldrich, St. specificity of the amplified bands was validated by their pre-
Louis, MO) (0.4 lg/mL), as controls of the protein load. Anti- dicted size (a-actin, 661 bp; tnfrsf1a, 586 bp; and tnfrsf1b,
mouse horseradish peroxidase-conjugated immunoglobulin G 402 bp). Densitometry analysis was performed by scanning the
(IgG) (Dako) was used as a secondary antibody in a 1:2000 dilu- gel images (Video Documenting System, Amersham-Pharma-
tion. An enhanced chemiluminescence detection system (Amer- cia Biotech inc., Piscataway, NJ), and values were obtained via
sham Biosciences, Piscataway, NJ) was used. The densities of Image Master VDS Software (GE Healthcare Life Sciences,
the bands were determined using Adobe Photoshop CS5 (Adobe UK).
Systems, San Jose, CA).
ELISA
Immunofluorescence IgG and IgM antibodies to the gangliosides asialo-GM1
Immunofluorescence assays were performed in SC cul- (GA1), GM1, GM2, GD1a, GD1b, GQ1b were determined by
Ganglio Combi ELISA (Bühlmann Laboratories, Schönen-

Downloaded from http://jnen.oxfordjournals.org/ by guest on February 21, 2016

tures and frozen tissue sections to evaluate surface-level TNF
and TNF-related molecules. ST8814 SCs were cultured onto 24- buch, Switzerland) in serum samples from 13 leprosy pa-
well plates containing glass coverslips covered with 4% silane tients. Results were expressed as a percent ratio of a highly
(Sigma-Aldrich). The samples were washed with PBS and fixed positive control and categorized as either negative (<30%), bor-
with 4% paraformaldehyde. To evaluate surface-level proteins, derline (30%–50%), or moderately (50%–100%) or strongly
samples were blocked with 10% goat serum and 10% BSA in positive (>100%). The values obtained from 3 healthy blood
PBS for 1 hour at room temperature and were then incubated donors served as a reference group.
with anti-TNF (1:50), anti-CD44 (1:100), and anti-S100 (1:200) TNF, IL-23, IL-8, IL-6, and IL1-b concentrations in
in 1% BSA/PBS overnight at 48C. cell-free ST8814 culture supernatants (30 mg of total protein)
Frozen sections (5 lm thick) of leprosy nerve tissues and patient sera were determined using commercial ELISA
(n ¼ 3) were also obtained and fixed with cold acetone. For kits and processed according to the manufacturer’s specifica-
protein surface-level evaluation the sections were blocked tions. Specific kits for IL-23, IL-8, IL-6, and IL1-b were pur-
with 5% goat serum and 10% BSA in PBS for 1 hour at room chased from R&D Systems. Evaluations of TNF expression
temperature. The samples were then incubated with anti-TNF and secretion in whole-cell extracts and supernatants were per-
(1:50), anti-TNFR1 (1:50), anti-TNFR2 (1:100), and anti- formed using a high sensitivity kit (lowest detection limit of
TACE (1:10) associated or not with anti-S100 (1:200) in 1% 4 pg/mL) from Ebioscience (San Diego, CA).
BSA/PBS overnight at 48C.
After rinsing with PBS, either Alexa 532 goat anti-
mouse or Alexa 633 goat anti-rabbit IgG secondary antibody Statistical Analysis
was added to both SC cultures and frozen sections for 1 hour at Results were expressed as mean 6 SE (standard error).
room temperature. Nuclei were stained with 4’,6-diamidino-2- The Kolmogorov-Smirnov test was performed to evaluate if
phenylindole (DAPI) (D9542, Sigma-Aldrich) and coverslips samples came from a Gaussian distribution as well as to de-
were mounted with Permafluor (Thermo Scientific, Waltham, termine the appropriate statistical test. Significant differences
MA). Samples were analyzed by an Axio Observer Z1 Colibri between 2 groups were determined by the Wilcoxon nonpara-
microscope (Zeiss, Oberkochen, Germany) and images were metric test or the paired t-test in the experimental data. The
processed via AxioVision software (Zeiss). Kruskal–Wallis test and a post-hoc test were used to compare
more than 2 groups. Pearson chi-squared test corrected by
Fisher exact test was employed for clinical data evaluation.
RNA Isolation and Semiquantitative RT-PCR The adopted statistical significance level was p  0.05. Anal-
ST8814 SC culturesV plated in 6-well plates were sus- yses were performed using Windows GraphPad Prism version
R
pended in 1 mL of TRIzol reagent (Life Technologies, Carls- 5.0 (GraphPad Software, San Diego, CA).
bad, CA). Total RNA was extracted according to the manufac-
turer’s instructions. cDNA was synthesized via Superscript III RESULTS
first-strand RT-PCR Kit (Invitrogen, Carlsbad, CA). Semiquan-
titative analysis of TNFR1 (tnfrsf1a) and TNFR2 (tnfrsf1b) Leprosy Patients With Neuritis Exhibit Evidence
mRNA was performed in ST8814 cultures as previously de- of Nerve Demyelination
scribed (32). Sequences for tnfrsf1a (gene ID: 7132) were: 50 - The 17 patients (64.7% male, mean age 49.4 6 1.6
ATTTGCTGTACCAAGTGCCACAAAGGAACC-30 and 50 -T years) evaluated were sorted into 2 groups. The group without
CGATTTCCCACAAACAATGGAGTAGAGC-30 . Sequences neuritis consisted of 8 patients (47.1%) and the group with neu-
for tnfrsf1b (Gene ID: 7133) were: 50 -GAATACTATGACCA- ritis included 9 patients (52.9%). The demographic findings of
GACAGCTCAGATGTGC-30 and 50 -TATCCGTGGATGAA the patients are presented in Table 1. Fifty-nine percent (n ¼ 10)

4
J Neuropathol Exp Neurol  Volume 0, Number 0, Month 2016 Cytokines and Demyelination in Leprous Neuritis

were receiving multidrug therapy. Most (64.7%, n ¼ 11) were were found at the same frequency (37.5%) in the neuritis
treated with the multibacillary scheme because they had the group; 25% of patients without neuritis were normal. Among
POLAR lepromatous (LL), borderline lepromatous (BL), or the patients with neuritis, 22.2% displayed polyneuropathy and
mid-borderline (BB) forms. Only 2 patients (11.7%) had perma- 66.6% had mononeuropathy multiplex. Importantly, the NCS
nent disabilities and/or acral ulcers at the time of diagnosis (ie, a results of all the individuals in the study were altered.
disability grade 2, according to the World Health Organization In the sensory NCS, in the group without neuritis, the
grading system (34)). number of nerves with no conduction disturbances (normal
The histopathological evaluation of the nerve biopsy nerves) was significantly higher than in the neuritis group
specimens in the neuritis group revealed a significant reduc- (p < 0.000482). However, in this group, the number of nerves
tion in the quantity of myelinated nerve fibers. Moreover, with no conduction was significantly more frequent (p < 0.00
there was endoneurial fibrosis along with an inflammatory in- 0001). The motor NCS showed a higher demyelination rate in
filtrate that consisted of lymphocytes and macrophages dis- the neuritis group (15 motor nerves [27.8%] vs 5 [10.4%] in the
tributed across the 3 nerve compartments (Supplementary group without neuritis, p ¼ 0.027502). When the total number of
Data 1). Foamy macrophages forming perivascular clusters nerves presenting any type of alteration was considered in each
were present in 1 section while an excessive extracellular ma- group (25% in the group without neuritis vs 61.2% in the neuri-
trix with a fibrotic appearance occupied the whole endoneu- tis group), the frequency was significantly higher in the neuritis
rium of the fascicles, leaving some residual inflammatory group (p < 0.0005; X2 ¼ 24.99) (Table 2).
cells and microvessels behind. Wade staining revealed acid- Demyelination was detected in 3 patients (5 motor nerves)

Downloaded from http://jnen.oxfordjournals.org/ by guest on February 21, 2016

fast bacilli in all of the samples (Supplementary Data 1). without neuritis in the motor NCS. Demyelination in 2 of these
Clinical examination showed that the neuritis group ex- patients could not be attributed to leprosy reaction because both
hibited a greater frequency of nerve enlargement (66.6% vs presented with carpal tunnel syndrome in the median nerve;
the group without neuritis [0%], p < 0.004), nerve pain (77.8% however, the possibility that the third patient was afflicted with
vs 12.5%, p < 0.007), sensory impairment (88.9% vs 50%, silent neuritis cannot be ruled out (Table 2). As in the sensory
p < 0.07), and motor impairment (66.7% vs 0%, p < 0.004). NCS, the number of nerves without any detectable injury (nor-
Paresthesia frequency was not significantly different between mal nerves) was significantly greater in the group without neuri-
the groups (66.7% in the neuritis group vs 37.5% in the non- tis (p ¼ 0.000246).
neuritis group, p ¼ 0.22). Seven patients in the neuritis group
felt pain during nerve palpation; 1 patient who reported nerve
pain had characteristics of neuropathic pain. Antiganglioside Antibody Screening Is Not a
The NCS studies indicated that the most frequent pattern Reliable Indicator of Peripheral Nerve
in the sensory examinations in the group without neuritis was Demyelination in Leprosy Patients
mononeuropathy multiplex (62.5%). Patients with neuritis Serum samples from 13 patients (8 with and 5 without
showed a greater percentage of polyneuropathy (55.5%). In the neuritis) were screened for the presence of antibodies against
motor NCS, mononeuropathy and mononeuropathy multiplex GA1, GM1, GM2, GD1a, GD1b, and GQ1b (Table 3). All of

TABLE 2. Neurophysiological Findings in Patients With and Without Neuritis

Without Neuritis Neuritis Total p value
(n ¼ 8) (n ¼ 9) (n ¼ 17)
Sensory nerves
Number of nerves evaluated n ¼ 64 n ¼ 72 n ¼ 136
Normal 27 (42.2%) 11 (15.2%) 38 (28%) <0.000482
Axonal 13 (20.3%) 19 (26.4%) 32 (23.5%) 0.404
Demyelinating 0 0 0 -
Mixed 0 0 0 -
No conduction 6 (9.4%) 30 (41.7%) 36 (26.5%) < 0.000001
No classification 18 (28.1%) 12 (16.6%) 30 (22%) < 0.107
Motor nerves
Number of nerves evaluated n ¼ 48 n ¼ 54 n ¼ 102
Normal 36 (75%) 21 (38.9%) 57 (56%) 0.000246
Axonal 1 (2.1%) 3 (5.55%) 4 (3.9%) 0.367194
Demyelinating 5 (10.4%) 15 (27.8%) 20 (19.6%) 0.027502
Mixed 0 0 0 -
No conduction 0 3 (5.55%) 3 (2.9%) 0.097408
No classification 6 (12.5%) 12 (22.2%) 18 (17.6%) 0.198581
Temporal dispersion 0 3
Compound muscle action potentials (CMAPs) and sensory nerve action potentials (SNAPs) were defined by combining the nerve conduction study parameters as defined in
“Materials and Methods.” The definition of partial conduction block (CB) depended on a 50% or higher reduction in the proximal compared to the distal amplitudes (29).

5
Andrade et al J Neuropathol Exp Neurol  Volume 0, Number 0, Month 2016

TABLE 3. Antiganglioside Antibodies in Leprosy Patient and Control Sera

Patient No. Reaction Nerve IgG IgM
Demyelination
GA1 GM1 GM2 GD1a CD1b GQ1b GA1 GM1 GM2 GD1a CD1b GQ1b
Patients without
neuritis
1 ENL No 20.7 16.4 18.2 22 23.1 16 59.4* 76.8* 15.1 15.1 31.1# 11
3 RR No 10.5 9.3 11.6 23.8 22 15.8 4.3 4.8 5.5 4.8 4.5 3.6
4 RR Yes 21.5 23.5 22.1 24 25.2 19.3 67.2* 47.6# 28.4 11.3 45.1# 16.1
5 RR Yes 10.2 9.6 10.4 25.3 12.9 11.5 8.7 12.4 4.1 8.2 4.6 4.1
7 RR No 7.9 7.2 8.3 8.1 8.5 9.6 23.3 12.1 11.5 6.7 8.9 8.1
Patients with
neuritis
10 RR þ Neuritis Yes 23 16.6 23.1 23.1 31.2# 19.7 64.9* 17.5 23.9 17.6 17.3 19.3
11 Neuritis Yes 14 12.7 14.9 31.6# 26.9 18 14.1 9.7 5.8 9.7 9.5 6.7
12 Neuritis Yes 21.2 19.8 19.8 19.8 17.1 17.3 46.4# 74.6* 18.8 15.2 20.5 14.9
13 ENLþ Neuritis Yes 11.2 8.4 11.6 18.7 8.5 9.7 15.2 18.1 10.8 4.7 5.6 5.5

Downloaded from http://jnen.oxfordjournals.org/ by guest on February 21, 2016

14 RR þ Neuritis Yes 12 10.2 11.1 15.6 15.5 12.5 9.6 7.2 4 8.8 6 4.2
15 RR þ Neuritis Yes 11.3 10.9 10.9 18.7 13.1 10.5 15 7.1 4.2 8.5 5.8 4.6
16 ENL þ Neuritis Yes 16.9 11.5 14.4 18.2 20.2 31.6# 13 6.1 6.3 6.3 4.7 4
17 Neuritis Yes 35# 26.5 40.5# 37# 20.7 57* 23.8 19.5 10.4 13.9 10.6 9.3
Healthy controls
18 Healthy control NE 32.1# 23.2 29.2 20.8 19.9 21.1 31.7# 18.8 18.9 18.4 45.9# 35.5#
19 Healthy control NE 156.9** 18.7 20.0 25.8 22.5 19.9 55.4* 13.4 17.1 15.4 14.1 90.2*
20 Healthy control NE 31.13# 21.2 52.1 35.9# 21.6 23.5 71.0* 21.3 30.7# 21.7 24.6 39.8#
ENL, erythema nodosum leprosum; NE, not evaluated; RR, reverse reaction.
Levels of antiganglioside antibodies (immunoglobulin G [IgG] and M [IgM]) were assessed in the sera of 13 leprosy patients with or without neuritis by enzyme-linked immuno-
sorbent assay (ELISA). Results of the IgG and IgM groups are expressed as % ratio of a highly positive control and categorized as negative (<30%), #borderline (30%–50%), or
*moderately (50%–100%) or **strongly positive (>100%). Values obtained from 3 healthy blood donors served as a reference group.

the patients in this group with neuritis had NCS evidence of de- ML Induces Transmembrane TNF Protein
myelination in one or more peripheral nerves even though the Expression and TNFR1 Gene Expression in the
frequency of autoantibodies against gangliosides in the serum Human SC Line ST8814
did not correlate with the conduction disturbances detected in We previously found that the human SC line ST8814 is
the NCS. With respect to IgM antibodies, the sample from only nonmyelinating and expresses S-100, CD44, laminin, and
1 patient with neuritis (12.5%) (with associated demyelination) GFAP (35). To evaluate the effects of ML infection on TNF
was immunoreactive to GA1, and only 1 other (12.5%) to GM1. production in human SCs, ST8814 cultures were stimulated
Similarly, elevated IgG antibody levels to GQ1b were found in with irradiated ML at different time points after which TNF
1 patient with neuritis. On the other hand, 2 patients without protein expression was assessed by Western blot, ELISA, and
neuritis (25%) (one with and the other without demyelination) immunofluorescence assays. As indicated by the protein ex-
showed elevated IgM anti-GA1 levels. All healthy control sera pression analysis of whole-cell extracts via Western blot and
exhibited different levels of IgM and IgG GA1 and IgM GQ1b ELISA, membrane-bound TNF (mTNF) was upregulated by
antibodies. ML stimulation at 3 and 7 hours after exposure (Fig. 2A, B).
The evaluation of TNF surface levels in ST8814 cultures after
ML stimulation by immunofluorescence illustrated that the
TNF Is Present in the Serum and Peripheral bacteria not only induced TNF protein expression, but also
Nerves of Leprosy Patients With Neuritis promoted molecular insertion in the cell membrane (Fig. 2C).
Our previous work showed that the peripheral nerves of On the other hand, ST8814 culture supernatants after ML
leprosy patients diagnosed with neuritis had higher TNF gene stimulation did not demonstrate soluble TNF (sTNF) at either
expression than was found in normal nerves (11). Here, we the early or late time points (Fig. 2D, E).
evaluated the protein expression of this cytokine and related Both the soluble and membrane-bound forms of TNF ex-
molecules in nerve lesions of 3 patients by immunofluores- ercise their biological effects through interaction with TNFR1
cence. TNF protein, TNF receptors, and TACE were most of- or TNFR2 (36). The next step involved evaluating ML modu-
ten expressed by SCs (Fig. 1). This led us to investigate the lation of TNF receptors in human SC ST8814. Gene expres-
role of this cytokine in the pathogenesis of leprosy nerve sion analysis of TNFR1 and TNFR2 after stimulation with irra-
damage via SC cultures. diated ML for 24 hours by semiquantitative RT-PCR showed

6
J Neuropathol Exp Neurol  Volume 0, Number 0, Month 2016 Cytokines and Demyelination in Leprous Neuritis

Downloaded from http://jnen.oxfordjournals.org/ by guest on February 21, 2016

FIGURE 1. Immunofluorescence analysis of peripheral nerves of leprosy patients for tumor necrosis factor (TNF), TNF receptors 1
(TNFR1) or 2 (TNFR2), and anti-TNF converting enzyme (TACE) expression in Schwann cells (SCs). Longitudinal cryosections
were stained with anti-S100 (green), a specific phenotypic marker of human SCs, together with anti-TNF, -TNFR1, -TNFR2, or
-TACE (red). Nerve lesions of 3 leprosy patients with neuritis were evaluated. Negative controls were made without the primary
antibodies. Nuclei are stained with 4’,6-diamidino-2-phenylindole (DAPI) (blue). Scale bar ¼ 50 lm.

7
Andrade et al J Neuropathol Exp Neurol  Volume 0, Number 0, Month 2016

Downloaded from http://jnen.oxfordjournals.org/ by guest on February 21, 2016

FIGURE 2. Mycobacterium leprae (ML) induces the expression of membrane-bound tumor necrosis factor (mTNF) but not its
secretion in the human Schwann cell (SC) line ST8814. (A) Evaluation of tumor necrosis factor (TNF) protein expression in
ST8814 SCs after stimulation with irradiated ML for 3, 5, and 7 hours (h) by Western blot. (B) Evaluation of TNF expression in
protein extracts from ST8814 SCs after stimulation with irradiated ML for 1, 3, 5, and 7 hours by enzyme-linked immunosorbent
assay (ELISA). (C) Phenotypic characterization of ST8814 SCs with CD44 (red) and S100 (green) staining (upper panels).
Evaluation of TNF surface levels (red) in ST8814 SCs after stimulation with irradiated ML for 3 and 7 hours by immunofluorescence
(lower panels). Nuclei are stained with 4’,6-diamidino-2-phenylindole (DAPI) (blue). Scale bar ¼ 50 lm. (D) Evaluation of TNF
secretion in supernatants from ST8814 SC cultures after stimulation with irradiated ML for 24 hours by Western blot; 50 ng of RhTNF
were used as a positive control. Protein load was assessed by Ponceau S staining. (E) Evaluation of TNF secretion in supernatants from
ST8814 SC cultures after stimulation with irradiated ML for 3, 6, and 24 hours by ELISA. The ELISA and Western Blot experiments
used 30 lg of total protein. (F, G) Evaluation of TNF receptors 1 (TNFR1) and 2 (TNFR2) gene expression in ST8814 SC cultures after
stimulation with irradiated ML for 24 hours by reverse transcriptase-polymerase chain reaction (RT-PCR). Results are expressed as
mean 6 SE and are representative of 4 or more independent experiments. Comparison between 2 groups was performed using
paired t-test. NS ¼ nonstimulated.

8
J Neuropathol Exp Neurol  Volume 0, Number 0, Month 2016 Cytokines and Demyelination in Leprous Neuritis

whether ML and/or TNF were able to induce secretion of in-

flammatory cytokines in the human SC line ST8814. Concen-
trations of different molecules in the supernatants of ST8814
cultures stimulated with RhTNF and irradiated ML either
alone or together for 24 hours were assessed by ELISA. ML
was not able to induce the secretion of IL-6 or IL-8 (Fig. 3A,
B). Only RhTNF stimulation led to the release of these cyto-
kines in the SC cultures while the combination of TNF and
ML did not differ from the effect caused by TNF alone. Inter-
estingly, IL-23 was detected in the supernatants of SC cul-
tures stimulated with irradiated ML for 24 hours (Fig. 3C).

DISCUSSION
To the best of our knowledge, this is the first report
linking demyelination and acute neuritis in leprosy. In this
study, 17 leprosy patients undergoing reaction with clinical
signs of neurological complications were evaluated and di-

Downloaded from http://jnen.oxfordjournals.org/ by guest on February 21, 2016

vided according to the presence or absence of neuritis.
Although reaction has been considered to be an important
factor underlying nerve damage, we found that only those pa-
tients who had acute neuritis had demyelination in the NCS.
In addition, patients diagnosed with neuritis exhibited more
nerve dysfunction based on the NCS; motor impairment was
more frequent in the neuritis group. Based on the NCS data,
the patients with neuritis had more extensive nerve involve-
ment with a greater number of nerves with demyelination and
absence of nerve conduction than the group without neuritis.
Neuritis patients also had more clinical symptoms than the
non-neuritis group (data not shown).
Demyelination in the group without neuritis could be
due to the development of silent neuritis in which the patient
has no pain symptoms and the diagnosis requires serial neuro-
logical examinations, particularly prior to possible initiation
of steroid treatment.
It is not known whether demyelination is a consequence
of the inflammatory process in neuritis or if it is directly induced
by ML, as reported by Rambukana et al (8). Persistent demye-
FIGURE 3. Mycobacterium leprae (ML) and tumor necrosis factor lination is associated with axonal damage, which progressively
(TNF) induce inflammatory cytokines in the human Schwann compromises large fibers, leading to motor impairment (7).
cell (SC) cell line ST8814. (A–C) Evaluation of interleukin (IL)-6 Damage of the myelin sheath may be a consequence of the in-
(A), IL-8 (B), and IL-23 (C) secretion in ST8814 SC cultures flammatory process resulting either from humoral immunity or
after stimulation with RhTNF and irradiated ML associated the release of immune mediators (37, 38), but the precise mech-
or not for 24 hours by enzyme-linked immunosorbent assay anisms of demyelination in leprosy neuropathy are still unclear.
(ELISA). Results are expressed as mean 6 standard error (SE) It is worth emphasizing that nerve samples taken for di-
and are representative of 4 or more independent experiments. agnosis in pure neural leprosy patients show advanced stages
Comparison between 2 groups was performed using the paired of nerve damage with severe fiber loss, inflammatory infil-
t-test (IL-8) or the Wilcoxon nonparametric test (IL-6). Comparison
among 4 groups was determined by the Kruskal–Wallis test (IL-
trate, and/or endoneurial fibrosis (39); thus, demyelination
23). NS ¼ nonstimulated. *p  0.05. may not be identified in histopathological analyses in earlier
stages. However, the patient nerve samples we analyzed here
showed remyelinated fibers indicating that there had previ-
that ML was able to induce the upregulation of TNFR1 but not ously been demyelination. NCS are clinically more efficient
of TNFR2 (Fig. 2F, G). in detecting demyelination because extensive nerve histo-
pathological analyses are not feasible. It is plausible that de-
myelination occurs in acute neuritis and that it is followed by
ML Induces Secretion of IL-23 in the Human remyelination in some patients who recover nerve function.
Schwann Cell Line ST8814 We found that antiganglioside autoantibodies were not
The capacity to produce inflammatory cytokines allows significantly associated with the detection of demyelination
SCs to modulate immune responses. We next investigated by NCS. Thus, they could not be employed as a predictive
9
Andrade et al J Neuropathol Exp Neurol  Volume 0, Number 0, Month 2016

marker for demyelination. We also evaluated serum levels of In transgenic mice that only express mTNF and not
TNF, IL1-b, and IL-10 but found no significant differences sTNF, the initiation and progression of experimental autoim-
between patients with and without neuritis (data not shown); mune encephalomyelitis (EAE) occurs with the maintenance
this is likely because of the small numbers of samples evalu- of autoimmune properties and resistance to infection similar
ated. On the other hand, evaluation of TNF and related mole- to what is found among wild-type mice (48). On the other
cules in leprosy nerve lesions showed expression of the cyto- hand, Ruuls et al demonstrated that mTNF is incapable of
kine together with both the receptor and shedding enzyme promoting an effective inflammatory response in the brain in
TACE in the SC populations, in agreement with our previous the absence of sTNF (49).
study (22). These data suggest that the possible role of TNF TNFRs seem to play divergent roles in neuroinflamma-
in leprous neuropathy is local and restricted to the site of the tory responses, with TNFR1 being involved in inflammation
injury with little systemic effects. and demyelination and TNFR2 in remyelination and pathol-
In painful neuropathy caused by chronic constriction in- ogy limitation (50). Here, we show that ML induced TNFR1
jury, 2 peaks of TNF production were identified: the first pro- gene expression in ST8814 cultures, a finding also reported
moted by local cells such as SCs and resident macrophages, and by Oliveira et al (32). We also evaluated TNFR1 secretion by
the second most likely caused by recruited macrophages from ML in ST8814 cultures finding that the pathogen was not able
the blood (15). TNF gene expression was also detected in the to induce TNFR1 secretion in the supernatants (data not shown).
sciatic nerves of mice after 1 hour of chronic constriction injury TNF/TNFR1 signaling has also been associated with many hu-
(16), and 24 hours after injury in a model of enhanced axon re- man diseases, including multiple sclerosis; this could be a mech-

Downloaded from http://jnen.oxfordjournals.org/ by guest on February 21, 2016

generation (40). Thus, endogenous TNF released by resident anism by which ML contributes to leprosy neuropathy (51).
populations in the PNS occurs early after nerve damage. Clinical signs of EAE were reduced in single or double
An immediate increase in TNF expression is known to TNFR1 knockout mice compared to wild-type mice with EAE,
induce massive macrophage infiltration (41), which is proba- whereas TNFR2-deficient mice exhibited enhanced disease se-
bly enhanced by metalloproteinases (MMPs) (42). TNF is verity accompanied by higher clinical scores and severe inflam-
also an upstream activator of MMP-9, a molecule involved in mation and demyelination (52, 53). It was further demonstrated
myelin basic protein degradation. Therefore, TNF expression that TNFR2 mediated protective effects, which included oligo-
is likely also related to inflammation-mediated demyelination dendrocyte regeneration and suppression of activated lympho-
(15). Indeed, poor macrophage recruitment and delayed mye- cytes (52).
lin removal in following sciatic nerve transection have been Accumulating evidence indicates that SCs can secrete
reported in TNF-deficient mice (41). inflammatory cytokines, reinforcing their potential role in the
High expression of TNF and related molecules in SCs modulation of the immune response in the PNS (54). Our
in the nerves of leprosy patients led us to speculate about the data show that stimulation with TNF can induce the secretion
effect of TNF on SCs upon exposure to ML in vitro and the of IL-6 and IL-8 in ST8814 cultures, which is in agreement
possibility of TNF contributing to nerve damage. Although with a previous study demonstrating that the ST8814 cell line
ML exposure resulted in upregulated protein expression of is responsive to TNF because the cell line expresses constitu-
mTNF with cell membrane staining, ML did not stimulate the tive levels of TNFR1 and TNFR2 (32). Previous studies have
release of sTNF into the extracellular space at either the early shown that IL-6 plays a role in axon regeneration following
or late time points analyzed. These results are in agreement injury by increasing the expression of regeneration-associated
with a previous study, which found that ML induces TNF genes in neurons (55). On the other hand, IL-8, a proinflam-
gene expression at early time points in ST8814 cultures, de- matory chemokine, has been detected in the CSF of patients
spite no evidence of cytokine detection in the supernatants with optic neuritis, and the presence of IL-8 has been associ-
(32). Conversely, ML activated the expression of a diverse ated with persistent demyelination and final axonal loss (56).
set of innate immunity-related genes in reprogrammed SCs Of the molecules investigated in ST8814 supernatant
after long-term infections; that is, high levels of a large range cultures, ML only induced IL-23. It has been suggested that
of chemokines and cytokines such as TNF were found in cul- IL-23 is critically involved in the pathogenesis of various im-
ture supernatants of SCs after 28 days of ML infection (10). mune-mediated disorders. High IL-23 gene expression has
Even without inducing TNF secretion in SCs, ML been detected at the onset and during the acute phase of ex-
seems to elicit TNF-mediated mechanisms through the trans- perimental autoimmune neuritis, an animal model of Guil-
membrane form of the cytokine. While sTNF exerts its ef- lain-Barré syndrome, which is characterized by multifocal
fects at sites remote from the site where it is produced, the demyelination and mononuclear cellular infiltration of pe-
transmembrane form acts in a cell-to-cell contact fashion, ripheral nerves (57). Macrophages expressing IL-23 have
functioning as a ligand by binding to TNF receptors and, as a also been found in sural nerves and CSF of Guillain-Barré pa-
receptor, transmitting signals back into the mTNF-bearing tients (57). Furthermore, IL-23 is upregulated in EAE, and
cells (43). The importance of mTNF in the immune responses mice lacking IL-23 are resistant to EAE, and the presence of
against several pathogens, including HIV, has been reported this cytokine in the CNS at the effector phase of EAE is re-
(44). In addition, mTNF was shown to provide protection quired for disease induction (58, 59). These data underscore the
against infections by Mycobacterium tuberculosis and the potential pathogenic relevance of IL-23 in the early phases of
less virulent Mycobacterium bovis bacillus Calmette-Guerin; immune-mediated demyelination in leprous neuropathy.
initiating T-cell and macrophage migration as well as granu- Overall, our data indicate that TNF seems to be an im-
loma formation (45–47). portant cytokine that participates in the early stages of SC

10
J Neuropathol Exp Neurol  Volume 0, Number 0, Month 2016 Cytokines and Demyelination in Leprous Neuritis

infection by ML. In addition, we demonstrated that ML is ca- 14. Bonetti B, Valdo P, Stegagno C, et al. Tumor necrosis factor alpha and
pable of contributing to a TNF-mediated response by induc- human Schwann cells: Signalling and phenotype modulation without cell
death. J Neuropathol Exp Neurol 2000;59:74–84
ing mTNF expression and upregulating TNFR1, thus render- 15. Shubayev VI, Myers RR. Upregulation and interaction of TNFalpha and
ing SCs more sensitive to the exogenous TNF levels in the gelatinases A and B in painful peripheral nerve injury. Brain Res
nerve, which likely originates from resident macrophages in 2000;855:83–9
the early stages of injury and, later, from inflammatory cells. 16. Uçeyler N, Tscharke A, Sommer C. Early cytokine expression in mouse
sciatic nerve after chronic constriction nerve injury depends on calpain.
TNF seems to act on SCs by inducing IL-6 and IL-8 produc- Brain Behav Immun 2007;21:553–60
tion, contributing not only directly, but also indirectly to neu- 17. Chattopadhyay S, Myers RR, Janes J, et al. Cytokine regulation of MMP-
roinflammation. Moreover, ML induces IL-23 secretion in 9 in peripheral glia: Implications for pathological processes and pain in
SCs, a cytokine believed to be involved in demyelinating pro- injured nerve. Brain Behav Immun 2007;21:561–8
cesses. Importantly, the type of focal demyelination seen in 18. Teles RMB, Moraes MO, Geraldo NT, et al. Differential TNFalpha
mRNA regulation detected in the epidermis of leprosy patients. Arch
leprosy is most likely caused by the initial inflammatory re- Dermatol Res 2002;294:355–62
sponse triggered by ML. Thus, focal demyelination could po- 19. Sarno EN, Grau GE, Vieira LM, et al. Serum levels of tumour necrosis
tentially become a prime target for therapeutic interventions factor-alpha and interleukin-1 beta during leprosy reactional states. Clin
aiming to improve nerve function in leprosy. Exp Immunol 1991;84:103–8
20. Barnes PF, Chatterjee D, Brennan PJ, et al. Tumor necrosis factor pro-
duction in patients with leprosy. Infect Immun 1992;60:1441–6
21. Andrade PR, Pinheiro RO, Sales AM, et al. Type 1 reaction in leprosy: A
model for a better understanding of tissue immunity under an immuno-

Downloaded from http://jnen.oxfordjournals.org/ by guest on February 21, 2016

pathological condition. Expert Rev Clin Immunol 2015;11:391–407
ACKNOWLEDGMENTS 22. Teles RMB, Antunes SLG, Jardim MR, et al. Expression of metallopro-
We thank Drs Anna Maria Sales and José Augusto teinases (MMP-2, MMP-9, and TACE) and TNF-alpha in the nerves of
leprosy patients. J Peripher Nerv Syst 2007;12:195–204
Nery for their involvement in patient selection; Dr Elizabeth 23. Ridley DS, Jopling WH. Classification of leprosy according to immunity.
Pereira Sampaio for her contribution to the “Discussion” A five-group system. Int J Lepr Other Mycobact Dis 1966;34:255–73
section; Dr Tercia Rodrigues Alves and Helen Ferreira for 24. Garbino JA, Virmond Mda C, Ura S, et al. A randomized clinical trial of
their technical support with the in vitro experiments and im- oral steroids for ulnar neuropathy in type 1 and type 2 leprosy reactions.
munofluorescence staining; and Judy Grevan for editing the Arq Neuropsiquiatr 2008;66:861–7
25. Ministério da Saúde. Guia para controle da Hansenı́ase––Cadernos de
text. Atenção Basica. Brazil: Ministério da Saúde 2002:12–8
26. Ministério da Saúde. Hansenı́ase–Atividades de controle e manual de
REFERENCES procedimentos. Brası́lia, DF: Fundação Nacional de Saúde 2001:15–28
1. Hussain T. Leprosy and tuberculosis: An insight-review. Crit Rev Micro- 27. WHO Expert Committee on Leprosy Sixth Report. Geneva: Technical
biol 2007;33:15–66 Report Series, 1988: 768
2. Stoner GL. Importance of the neural predilection of Mycobacterium lep- 28. Delisa JALH, Baran EM, Lai KS. Manual of Nerve Conduction Velocity
rae in leprosy. Lancet 1979;2:994–6 and Clinical Neurophysiology. Philadelphia, PA: Lippincott Williams &
3. Sibley LD, Franzblau SG, Krahenbuhl JL. Intracellular fate of Mycobac- Wilkins 1994
terium leprae in normal and activated mouse macrophages. Infect Immun 29. Olney RK, Lewis RA, Putnam TD, et al. Consensus criteria for the diag-
1987;55:680–5 nosis of multifocal motor neuropathy. Muscle Nerve 2003;27:117–21
4. Miko TL, Le Maitre C, Kinfu Y. Damage and regeneration of peripheral 30. Glover TW, Stein CK, Legius E, et al. Molecular and cytogenetic analy-
nerves in advanced treated leprosy. Lancet 1993;342:521–5 sis of tumors in von Recklinghausen neurofibromatosis. Genes Chromo-
5. Bahia El Idrissi N, Das PK, Fluiter K, et al. M. leprae components induce somes Cancer 1991;3:62–70
nerve damage by complement activation: Identification of lipoarabino- 31. Scheinman RI, Cogswell PC, Lofquist AK, et al. Role of transcriptional
mannan as the dominant complement activator. Acta Neuropathol 2015; activation of I kappa B alpha in mediation of immunosuppression by glu-
129:653–67 cocorticoids. Science 1995;270:283–6
6. Shetty VP, Mehta LN, Irani PF, et al. Study of the evolution of nerve 32. Oliveira RB, Sampaio EP, Aarestrup F, et al. Cytokines and Mycobacte-
damage in leprosy. Part I–Lesions of the index branch of the radial cuta- rium leprae induce apoptosis in human Schwann cells. J Neuropathol
neous nerve in early leprosy. Lepr India 1980;52:5–18 Exp Neurol 2005;64:882–90
7. Jardim MR, Vital R, Hacker MA, et al. Leprosy neuropathy evaluated by 33. Moraes MO, Sarno EN, Almeida AS, et al. Cytokine mRNA expression
NCS is independent of the patient’s infectious state. Clin Neurol Neuro- in leprosy: A possible role for interferon-gamma and interleukin-12 in re-
surg 2015;131:5–10 actions (RR and ENL). Scand J Immunol 1999;50:541–9
8. Rambukkana A, Zanazzi G, Tapinos N, et al. Contact-dependent demye- 34. World Health Organization. International Classification of Impairments,
lination by Mycobacterium leprae in the absence of immune cells. Sci- Disabilities and Handicaps. Geneva, Switzerland: World Health Organi-
ence 2002;296:927–31 sation 1980
9. Masaki T, Qu J, Cholewa-Waclaw J, et al. Reprogramming adult Schwann 35. Silva TP, Silva AC, Baruque Mda G, et al. Morphological and functional
cells to stem cell-like cells by leprosy bacilli promotes dissemination of in- characterizations of Schwann cells stimulated with Mycobacterium lep-
fection. Cell 2013;152:51–67 rae. Mem Inst Oswaldo Cruz 2008;103:363–9
10. Masaki T, McGlinchey A, Cholewa-Waclaw J, et al. Innate immune re- 36. Ihnatko R, Kubes M. TNF signaling: Early events and phosphorylation.
sponse precedes Mycobacterium leprae-induced reprogramming of adult Gen Physiol Biophys 2007;26:159–67
Schwann cells. Cell Reprogram 2014;16:9–17 37. Bitsch A, Kuhlmann T, Da Costa C, et al. Tumour necrosis factor alpha
11. Oliveira AL, Antunes SLG, Teles RM, et al. Schwann cells producing mRNA expression in early multiple sclerosis lesions: Correlation with de-
matrix metalloproteinases under Mycobacterium leprae stimulation may myelinating activity and oligodendrocyte pathology. Glia 2000;29:366–75
play a role in the outcome of leprous neuropathy. J Neuropathol Exp 38. Genain CP, Cannella B, Hauser SL, et al. Identification of autoantibodies
Neurol 2010;69:27–39 associated with myelin damage in multiple sclerosis. Nat Med 1999;5:
12. Crawford AH, Chambers C, Franklin RJM. Remyelination: The true 170–5
regeneration of the central nervous system. J Comp Pathol 2013;149: 39. Antunes SL, Chimelli L, Jardim MR, et al. Histopathological examina-
242–54 tion of nerve samples from pure neural leprosy patients: Obtaining maxi-
13. Willison HJ, Yuki N. Peripheral neuropathies and anti-glycolipid anti- mum information to improve diagnostic efficiency. Mem Do Inst
bodies. Brain 2002;125:2591–625 Oswaldo Cruz 2012;107:246–53

11
Andrade et al J Neuropathol Exp Neurol  Volume 0, Number 0, Month 2016

40. Mietto BS, Jurgensen S, Alves L, et al. Lack of galectin-3 speeds Waller- 50. Caminero A, Comabella M, Montalban X. Tumor necrosis factor alpha
ian degeneration by altering TLR and pro-inflammatory cytokine expres- (TNF-a), anti-TNF-a and demyelination revisited: An ongoing story. J
sions in injured sciatic nerve. Eur J Neurosci 2013;37:1682–90 Neuroimmunol 2011;234:1–6
41. Liefner M, Siebert H, Sachse T, et al. The role of TNF-alpha during Wal- 51. Mc Guire C, Beyaert R, van Loo G. Death receptor signalling in central
lerian degeneration. J Neuroimmunol 2000;108:147–52 nervous system inflammation and demyelination. Trends Neurosci
42. Shubayev VI, Angert M, Dolkas J, et al. TNFa-induced MMP-9 pro- 2011;34:619–28
motes macrophage recruitment into injured peripheral nerve. Mol Cell 52. Suvannavejh GC, Lee HO, Padilla J, et al. Divergent roles for p55 and
Neurosci 2006;31:407–15 p75 tumor necrosis factor receptors in the pathogenesis of MOG(35-55)-
43. Eissner G, Kolch W, Scheurich P. Ligands working as receptors: Reverse induced experimental autoimmune encephalomyelitis. Cell Immunol
signaling by members of the TNF superfamily enhance the plasticity of 2000;205:24–33
the immune system. Cytokine Growth Factor Rev 2004;15:353–66 53. Kassiotis G, Kollias G. Uncoupling the proinflammatory from the immu-
44. Lazdins JK, Grell M, Walker MR, et al. Membrane tumor necrosis factor nosuppressive properties of tumor necrosis factor (TNF) at the p55 TNF
(TNF) induced cooperative signaling of TNFR60 and TNFR80 favors in- receptor level: Implications for pathogenesis and therapy of autoimmune
duction of cell death rather than virus production in HIV-infected T cells. demyelination. J Exp Med 2001;193:427–34
J Exp Med 1997;185:81–90 54. Rutkowski JL, Tuite GF, Lincoln PM, et al. Signals for proinflammatory
45. Olleros ML, Guler R, Corazza N, et al. Transmembrane TNF induces an cytokine secretion by human Schwann cells. J Neuroimmunol 1999;101:
efficient cell-mediated immunity and resistance to Mycobacterium bovis 47–60
bacillus Calmette-Guérin infection in the absence of secreted TNF and 55. Cafferty WB, Gardiner NJ, Das P, et al. Conditioning injury-induced spi-
lymphotoxin-alpha. J Immunol 2002;168:3394–401 nal axon regeneration fails in interleukin-6 knock-out mice. J Neurosci
46. Olleros ML, Guler R, Vesin D, et al. Contribution of transmembrane tu- 2004;24:4432–43
mor necrosis factor to host defense against Mycobacterium bovis bacillus 56. Rossi S, Motta C, Studer V, et al. Interleukin-8 is associated with acute and
Calmette-Guerin and Mycobacterium tuberculosis infections. Am J Pathol persistent dysfunction after optic neuritis. Mult Scler 2014;20:1841–50

Downloaded from http://jnen.oxfordjournals.org/ by guest on February 21, 2016

2005;166:1109–20 57. Hu W, Dehmel T, Pirhonen J, et al. Interleukin 23 in acute inflammatory
47. Saunders BM, Tran S, Ruuls S, et al. Transmembrane TNF is sufficient demyelination of the peripheral nerve. Arch Neurol 2006;63:858–64
to initiate cell migration and granuloma formation and provide acute, but 58. Cua DJ, Sherlock J, Chen Y, et al. Interleukin-23 rather than interleukin-
not long-term, control of Mycobacterium tuberculosis infection. J Immu- 12 is the critical cytokine for autoimmune inflammation of the brain.
nol 2005;174:4852–9 Nature 2003;421:744–8
48. Alexopoulou L, Kranidioti K, Xanthoulea S, et al. Transmembrane TNF 59. Li J, Gran B, Zhang GX, et al. Differential expression and regulation of
protects mutant mice against intracellular bacterial infections, chronic in- IL-23 and IL-12 subunits and receptors in adult mouse microglia. J Neu-
flammation and autoimmunity. Eur J Immunol 2006;36:2768–80 rol Sci 2003;215:95–103
49. Ruuls SR, Hoek RM, Ngo VN, et al. Membrane-bound TNF supports
secondary lymphoid organ structure but is subservient to secreted TNF in
driving autoimmune inflammation. Immunity 2001;15:533–43

12