Parasite Immunology, 2010, 32, 285–295 DOI: 10.1111/j.1365-3024.2009.01188.

Signs of an in situ inflammatory reaction in scars of human American

tegumentary leishmaniasis

F. N. MORGADO,1,2 A. SCHUBACH,2 E. VASCONCELLOS,2 R. B. AZEREDO-COUTINHO,2 C. M. VALETE-ROSALINO,2,3

L. P. QUINTELLA,2 G. SANTOS,2 M. SALGUEIRO,2 M. R. PALMEIRO1,2 & F. CONCEIffiO-SILVA1

1
Laboratory of Immunoparasitology, Oswaldo Cruz Institute, IOC ⁄ FIOCRUZ, 2Department of Pathology and VigiLeish, Outpatient
Clinics of Otolaryngology and Leishmaniasis, IPEC ⁄ FIOCRUZ, 3Department of Otolaryngology and Ophthalmology, Faculty of Medicine,
UFRJ, Rio de Janeiro, Brazil

SUMMARY
Skin inflammation plays an important role during the heal-
ing of American tegumentary leishmaniasis (ATL), the dis-
tribution of cells in active lesions may vary according to
disease outcome and parasite antigens in ATL scars have
already been shown. We evaluated by immunohistochemistry,
18 patients with 1- or 3-year-old scars and the correspond-
ing active lesions and compared them with healthy skin.
Small cell clusters in scars organized as in the active lesions
spreaded over the fibrotic tissue were detected, as well as
close to vessels and cutaneous glands, despite a reduction in
the inflammatory process. Analysis of 1-year-old scar tissue
showed reduction of NOS2, E-selectin, Ki67, Bcl-2 and Fas
expression. However, similar percentages of lymphocytes and
macrophages were detected when compared to active lesions.
Only 3-year-old scars showed reduction of CD3+, CD4+ and
CD8+T cells, in addition to reduced expression of NOS2,
E-selectin, Ki67 and BCl-2. These results suggest that the
pattern of cellularity of the inflammatory reaction observed
in active lesions changes slowly even after clinical healing.
Analysis of 3-year-old scars showed reduction of the inflam-
matory reaction as demonstrated by decrease in inflamma-
tory cells and in the expression of cell-activity markers,
suggesting that the host–parasite balance was only estab-
lished after that period.
Keywords ATL, cellular immune response, immunohisto-
chemistry, in situ, scars
Correspondence: F. Conceiżo-Silva, Lab Imunoparasitologia,
IOC ⁄ FIOCRUZ ⁄ Pavilh¼o 26 – 4 andar sala 406C, Av. Brasil
4365, Manguinhos – 21045-900 ⁄ Rio de Janeiro – RJ, Brazil
(e-mail: fconcei@ioc.fiocruz.br).
Disclosures: The authors state no conflict of interest.
Received: 12 August 2009
Accepted for publication: 26 November 2009
INTRODUCTION
The most frequent clinical feature of American tegumentary
leishmaniasis (ATL) produced by Leishmania braziliensis
in Rio de Janeiro, Brazil, are skin ulcers, (one or more)
containing few parasites and cases of mucosal involvement
are rare (1). Clinical healing usually occurs after antimo-
nial therapy, although partial resistance and even relapse
several years after clinical cure may also occur (2).
The skin is the major interface between the body and
the environment and it is responsible for the first defense
against pathogen invasion or trauma (3). In this respect,
the skin immune response represents a system organized
to elicit a rapid and sustained response to aggressions
(4,5). Thus, the healing of skin lesions involves a complex
process, including three phases: inflammatory, proliferative
or fibroblastic and the tissue remodelling (6–8).
In ATL some factors controlling the inflammatory pro-
cess like apoptosis, the cytokine milieu and nitric oxide
production in the skin wound after healing have been
described (9–15). In a recent study, we demonstrated an
association between high expression of NOS2 in situ and
low parasite numbers, suggesting the importance of that
enzyme for the control of local parasite burden (16).
Although various studies have investigated the factors
influencing the course of ATL, the process of mainte-
nance, cure or reactivation are not completely understood
(17–20). In this context, a recent review (21) on infectious
diseases points to the persistence of parasites, bacteria or
viruses and their influence on the processes of fibrosis and
healing.
In view of the possibility of parasite persistence (11–15),
the study of the in vitro and in situ cellular immune
responses in patients clinically cured of ATL and the com-
parison with findings obtained by biopsy of active skin
lesions may contribute to understanding of the dynamics
of healing or development of late lesions in ATL. This is

 2010 Blackwell Publishing Ltd 285

F. N. Morgado et al.
of a particular interest to understand how reactivation of
lesions in immunosuppressed patients can be prevented or
better treated. Therefore, we evaluated the inflammatory
reaction in the healed ulcers (scars) of ATL patients at
variable times after clinical cure.
MATERIALS AND METHODS
Patients
Tissue fragments from skin ulcers of eighteen ATL
patients diagnosed by isolation and identification of
L. (Viannia) braziliensis as previously described (22) were
examined and compared with fragments obtained from the
respective ATL scar 1 year (n = 9) or 3 years (n = 9) after
healing. The 1-year-old scar group (n = 9) was composed
of patients presenting 1–3 skin ulcers during active disease
(median 1Æ0), six out of nine the patients presented one
lesion, two patients presented two lesions and only one
patient presented three lesions. The duration of active
lesions was 1Æ5–4Æ0 months (median 2Æ0). The 3-year-old
scar group (n = 9) was composed of patients, presented
1–6 skin ulcers during active disease (median 1Æ0), six
patients presented one lesion, one patient presented two
lesions, one patient presented three lesions, and one
patient presented six lesions. The duration of active lesions
was 1Æ0–12Æ0 months (median 2Æ0). No difference was
observed between the two groups of active lesions sug-
gesting that the number of lesions had no influence in
the healing process. This data led us to imply that the
comparison between the two groups of scars was not influ-
enced by the profile of their active lesions. In case of mul-
tiple lesions, the same area (scar and active lesion) was
evaluated at different time. As a control, five healthy skin
fragments (HS) obtained during plastic surgery of trunk
from subjects without a history of ATL were analysed.
The tissue fragments were taken from the border of scar
or skin ulcer. All tissue fragments, including controls,
lesions and scars, were divided into three parts: (i) histo-
pathology (fixed in 10% buffered formalin), (ii) immuno-
histochemistry [cryopreserved at )196C in optimal
cutting temperature (OCT) compound (Sakura Finetek.,
Torrance, CA, USA)] and (iii) isolation of Leishmania sp.
(sterile saline) (22). The patients were included in the
study after formal consent approved by the Institutional
Ethics Committee on Human Research (protocols
0047Æ0Æ011Æ009–06 and 014 ⁄ 2001). All patients were treated
with glucantime at the time of active lesions and followed
by clinical and laboratorial examination for 48 months
after healing.
When secondary infection was detected the patients
were submitted to a cycle of antibiotics before the onset
286
Parasite Immunology
of specific treatment with glucantime. After the beginning
of glucantime therapy, no sign of secondary infection was
observed during the follow-up.
Histopathology
The formalin-fixed fragments were stained with
haematoxylin-eosin and examined by light microscopy
(Zeiss, Jena, Germany). The intensity of the inflammatory
infiltrate was quantified in at least 12 microscopic fields
under high magnification (200·) using a grid-scale, with
20 · 20 subdivisions in an area of 10 mm2. In addition,
the intensity of the inflammatory infiltrate was scored as
discrete (until 300 cells ⁄ mm2), moderate (more than
300 cells ⁄ mm2 until 1000 cells ⁄ mm2) or intense (more than
1000 cells ⁄ mm2).
Immunohistochemistry
The immunohistochemistry procedure was performed as
previously described (16). Briefly, 3-lm sections of the fro-
zen fragments mounted on microscope silanized slides
(DakoCytomation, Carpinteria, CA, USA) were fixed in
acetone PA (pure acetone Pro Analyse) (Merck, Darms-
tadt, Germany). After hydration in PBS, pH 7Æ4 and block-
age of unspecific staining (peroxidase blocking reagent,
Dako and normal goat serum, Zymed Laboratories Inc.,
South San Francisco, CA, USA) the specific antibodies
were applied. The following primary antibodies were used:
CD3+ (clone UCHT1), CD4+ (clone MT310) and CD8+
(clone DK25) T lymphocytes, CD22+ (clone 4KB128) B
lymphocytes, CD1a+ (clone NA1 ⁄ 34) langerhans cells,
CD68+ (clone KP1) macrophages, CD62E+ (clone 1Æ2B6)
E-selectin (endothelial activation), BCl-2+ (clone 124 – apop-
tosis inhibitor), Ki67+ (clone Ki-S5 – proliferative cells) and
anti-neutrophilic elastase (clone NP57 – neutrophils) (Dako-
Cytomation), NOS2 (clone 6 – nitric oxide synthase 2) (BD
Transduction Laboratories, Lexington, KY, USA), cutane-
ous lymphocyte antigen (CLA) (clone HECA-452), Fas
(clone DX2) and FasL (clone G247-4) (apoptosis inducers)
(BD Biosciences Pharmingen, San Diego, CA, USA), as
well as a polyclonal rabbit anti-Leishmania sp. serum pro-
vided by Dr Madeira (IPEC-FIOCRUZ). The specimens
were then incubated in sequence with: biotinylated second-
ary antibody, streptavidin-biotin-peroxidase complex (ABC
kit, DakoCytomation) and aminoethyl carbazole (AEC kit,
Zymed). The slides were counterstained with Mayer’s hae-
matoxylin (Dako) and examined under a light microscope
(Zeiss). The percentage of stained cells was determined by
counting 500 mononuclear cells as standard. The intensity
of NOS2 and E-selectin staining was scored in ten micro-
scope fields (20· magnification) as discrete (at least one
 2010 Blackwell Publishing Ltd, Parasite Immunology, 32, 285–295
Volume 32, Number 4, April 2010 Signs of inflammation in ATL scars

positive site per field), moderate (two or three positive sites),
intense (four or five positive sites), and very intense (more
than five positive sites) as previously described (16). For
each marker, we examined three sections of the fragment
tissues in different experiments (times).
Statistical analysis
The SPSS11 for Windows (SPSS Inc., Chicago, Illinois,
USA) was used for statistical analysis. The data were anal-
ysed using the nonparametric Mann–Whitney to compare
data obtained for 1-year-old vs. 3-year-old scars and the
Willcoxon test to compare data from 1- or 3-year-old scars
vs. their respective active lesions. Non-numerical data were
analysed in 2 · 2 contingency tables by Fisher’s exact test
using the INSTAT program (Graphpad Software V2-04,
Graphpad InstatTM Graphpad, San Diego, CA, USA).
Age is reported as mean € SD. The other data are
reported as mean € SEM.
fibrosis or around cutaneous adnexa with these cell clus-
ters being more concentrated in the papillary dermis. The
histopathological aspects of 1-year-old and 3-year-old
scars were similar. When compared to active lesions, the
ATL scars showed structural differences such as: presence
of cicatricial fibrosis, increased number of fibroblasts and,
in some cases, hyperkeratosis and hyperpigmentation. Cel-
lularity per mm2 showed differences between active lesions
and 1-year-old scars (P = 0Æ021) and 3-year-old scars
(P = 0Æ028) (Figure 1 and Table 1).
The presence of discrete to moderate inflammatory
infiltrate was detected in three cases (two patients with
1-year-old and one with 3-year-old scars, Table 1). Clinical
examination of these two patients showed no macroscopic
signs of ATL activity neither amastigote forms were
microscopically detected.
Healthy skin samples showed no major histopathologi-
cal alterations (Table 1).

RESULTS
P = 0.001
Characteristics of the patients 3000
P = 0.028
Five (27Æ7%) of the 18 patients were females and 13 P = 0.021
(72Æ3%) were males, ranging in age from 18 to 69 years
Cellularity/mm2

(41Æ94 € 15Æ20). The scars and active lesions analysed were 2000
located on the upper limbs (seven in 1-year group and two
in 3-years groups), lower limbs (two in 1-year and five in
3-years groups) and trunk (two in 3-years group). No 1000
significant difference was observed between the groups
with regard to scar distribution (OR = 8Æ75; P = 0Æ13).

0
Histopathology Active Scar 1 Scar 3 HS
2
Figure 1 Number of cells per mm active lesions, 1-year-old scars,
An intense and diffuse infiltrate predominated in active 3-year-old scars and healthy skin. Each point corresponds to one
lesions (Table 1). In contrast, scars were generally charac- patient and line corresponds to median. Active: active lesions;
terized by the presence of cell nests amidst intense dermal scar 1: 1-year-old scars, scar 3: 3-year-old scars; HS: healthy skin.

Table 1 Quantitative and qualitative analysis of the inflammatory infiltrate in ATL lesions and scars

Intensity of the inflammatory infiltrate upon histopathology

Cellularity ⁄ mm2 (mean € SEM) Absent Discrete Moderate Intense

Active lesion 1394Æ89 € 206Æ68 0 2 3 13

1-year-old scar 254Æ52 € 55Æ92 7 1 1 0
3-year-old scar 143Æ66 € 26Æ29 8 1 0 0
Healthy skin 170Æ76 € 30Æ90 5 0 0 0

Active lesion vs. 1-year-old scar (P = 0Æ021, Willcoxon test); active lesion vs. 3-year-old scar (P = 0Æ028, Willcoxon test); 1-year-old scar vs.
3-year-old scar (P = 0Æ171, Mann–Whitney test); 1-year-old scar vs. healthy skin (P = 0Æ641, Mann–Whitney test); 3-year-old scar vs.
healthy skin (P = 0Æ641, Mann–Whitney test).

 2010 Blackwell Publishing Ltd, Parasite Immunology, 32, 285–295 287

F. N. Morgado et al. Parasite Immunology

Langerhans cells (CD1a+) were found concentrated in

Immunohistochemistry
the epidermis of scars in all cases, and in the papillary der-
Figures 2–5 show the percentage of each cell marker mis and close to blood vessels in small number (Fig-
observed in the active lesions, and their respective 1-year- ure 6a.5 ⁄ a.6). No significant difference was observed when
old or 3-year-old scars. scars were compared with HS (data not shown).

Macrophages, lymphocytes, neutrophils and langerhans cells Markers of inflammation

In the scars, CD68+ macrophages were observed in the cell
clusters with variable percentages. No significant differ-
ences in the percentages of macrophages were observed
between scars, active lesions or HS samples (Figure 2a,b
and Table 2). Similar results were observed in terms of
CD22+ B-cells, in despite of their scarcity in the lesions,
scars and healthy skin (Figure 3a,b). CD22+ B-cells were
detected in eight of the 18 scars examined (four 1-year-old
scars and four 3-year-old scars).
With regard to CD3+, CD4+ and CD8+ T cells, a sta-
tistically significant reduction of the percentages of these
T-cell types were observed in 3-year-old scars as com-
pared with active lesions (P = 0Æ043, 0Æ015 and 0Æ011
respectively, Figures 3d, f and h and 6a.1–a.4). When
compared 1-year-old scars vs. active lesions the results
were not statistically different (Figure 3c, e and g). In
addition, CD4+ T cells showed significant differences
between healthy skin and 1-year-old scars (P = 0Æ005)
and 3-year-old scars (P = 0Æ02), whereas CD3+ T cells
showed such difference only for HS vs. 1-year-old scars
(P = 0Æ005) (Table 2).
The frequency of neutrophils in cicatricial tissues was
lower than in active lesions (P = 0Æ043 and 0Æ051 respec-
tively – Figure 2c,d) and no significant difference was
observed between scars and HS.
Activated endothelial cells (CD62E+) were often co-local-
ized with CLA+ cells, a fact commonly observed in active
lesions (Figure 6b.1 ⁄ b.2). The percentage of CLA+ cells
showed no significant differences between active lesions
and their 1-year-old or 3-year-old scars (Figure 4a,b). Sig-
nificant differences were however observed when compar-
ing HS samples and 1-year-old (P = 0Æ013) and 3-year-old
scars (P = 0Æ019) (Table 2).
Analysis of Ki67+ cells in clusters areas showed signifi-
cant reduction between active lesions and their 1-year-old
or 3-year-old scars (P = 0Æ028 and P = 0Æ011, respectively)
(Figure 5a,b). Ki67+ cells were not found in the dermis of
five of the 1-year-old scars and in four of the 3-year-old
scars. No significant difference was observed between scars
and HS.
Analysis of the concentration of BCl-2+ cells showed a
significant reduction in 1-year-old and 3-year-old scars
when compared to active lesions (P = 0Æ036 and
P = 0Æ021, respectively) (Figure 5c,d). In addition, a signif-
icant difference was observed between 3-year-old scars
and HS (P = 0Æ028), but not between 1-year-old scars and
HS (Table 2).
No significant differences were observed in the expres-
sion of FasL between active lesions and their respective
1-year-old and 3-year-old scars (Figure 5e,f). Expression of

Active lesion 1-year-scar Active lesion 3-year-scar

80 80
P = 0.11
70 (a) 70 (b) P = 0.44
expressing cells
Percentage of

60 60
50 50
CD68 40 40
30 30
20 20
10 10
0 0
Active 1 Scar 1 Active 3 Scar 3
MO MO

80 80
expressing cells

70 (c) P = 0.043 P = 0.051

Percentage of

70 (d)
60 60
50 50
NELA 40 60
30 30
20 20
10 10
0 0
Active 1 Scar 1 Active 3 Scar 3
NEU NEU
Figure 2 Percentage of myeloid cells (macrophages and neutrophils) (a, c) in active lesions vs. 1-year-old scars and (b, d) in active lesions
vs. 3-year-old scars. Each line corresponds to one patient. Willcoxon test. MO: macrophages; NEU: neutrophils; NELA: neutrophilic
elastase; active 1: active lesions 1-year-old group; active 3: active lesion 3-year-old group; scar 1: 1-year-old scars; scar 3: 3-year-old scars.

288  2010 Blackwell Publishing Ltd, Parasite Immunology, 32, 285–295

Volume 32, Number 4, April 2010 Signs of inflammation in ATL scars

Active lesion 1-year scar Active lesion 3-year scar

80 80
(a) (b)

expressing cells
70 P = 0.13 70 P = 0.068

Percentage of
60 60
50 50
CD22 40 40
30 30
20 20
10 10
0 0
Active 1 Scar 1 Active 3 Scar 3
CD22 CD22

80 (c) 80 (d)
70 P = 0.26 70 P = 0.043
expressing cells
Percentage of

60 60
50 50
CD3 40 40
30 30
20 20
10 10
0 0
Active 1 Scar 1 Active 3 Scar 3
CD3 CD3

80 80
70
(e) P = 0.67 (f) P = 0.015
70
expressing cells
Percentage of

60 60
50 50
CD4 40 40
30 30
20 20
10 10
0 0
Active 1 Scar 1 Active 3 Scar 3
CD4 CD4

80 80
(g) P = 0.31 (h) P = 0.011
70 70
expressing cells
Percentage of

60 60
50 50
CD8
40 40
30 30
20 20
10 10
0 0
Active 1 Scar 1 Active 3 Scar 3
CD8 CD8
Figure 3 Percentage of lymphocyte subsets (CD22, CD3, CD4 and CD8) (a, c, e, g) in active lesions vs. 1-year-old scars and (b, d, f, h) in
active lesions vs. 3-year-old scars. Each line corresponds to one patient. Willcoxon test. Active 1: active lesions 1-year-old group; active 3:
active lesion 3-year-old group; scar 1: 1-year-old scars; scar 3: 3-year-old scars.

the Fas showed significant difference between active lesions
and 1-year scars (P = 0Æ028) but not with 3-year-old scars
(Figure 5g,h).
The intensity of NOS2 expression was generally low in
scars. The intensity of NOS2 was significantly reduced in
3-year-old when compared to 1-year-old scars (P = 0Æ03).
Comparing active lesions with 1-year-old or 3-year-old
scars (P = 0Æ023 and P = 0Æ006, respectively) (Table 3 ⁄
Figure 4c,d) significant reduction of NOS2 was observed.
Areas of inflammatory activity persisted amidst cicatricial
tissue, although markedly reduced.
Low expression of E-selectin ranging from discrete to
moderate was observed in scars. Activated vessels were
observed in the centre of the cell clusters present in the
dermis amidst intense fibrosis. When compared to active
lesions, a significant reduction in E-selectin expression
was observed in 1-year-old and 3-year-old scars
(P = 0Æ024 and P = 0Æ026, respectively) (Table 3 and
Figure 4e,f). There was no difference between scars and
HS.
Parasites
Parasites were detected in 10 active lesions and in two
3-year-old scars. However, parasites were not detected in
active lesions of these two patients. Regardless to the
inflammatory markers these two patients presented
similar percentages when compared with the other
patients.

 2010 Blackwell Publishing Ltd, Parasite Immunology, 32, 285–295 289

F. N. Morgado et al. Parasite Immunology

Active lesion 1-year scar Active lesion 3-year scar

80 80 P = 0.21

expressing cells
(a) (b)

Percentage of
70 P = 0.08 70
60 60
50 50
CLA 40 40
30 30
20 20
10 10
0 0
Active 1 Scar 1 Active 3 Scar 3
CLA CLA

4.5 4.5
NOS2 expression

4 (c) P = 0.023 4 (d) P = 0.006

Intensity of

3.5 3.5
3 3
2.5 2.5
NOS2 2 2
1.5 1.5
1 1
0.5 0.5
0 0
Active 1 Scar 1 Active 3 Scar 3
NOS2 NOS2
e-selectin expression

4.5 (e) 4.5

4
P = 0.024
4
(f) P = 0.026
Intensity of

3.5 3.5
3 3
CD62e 2.5 2.5
2 2
1.5 1.5
1 1
0.5 0.5
0 0
Active 1 Scar 1 Active 3 Scar 3
E-selectin E-selectin
Figure 4 Percentage of inflammation markers (cutaneous lymphocyte antigen) (a) in active lesions vs. 1-year-old scars and (b) in active
lesions vs. 3-year-old scars. Intensity of NOS2 and E-selectin in (c, e) active lesions vs. 1-year-old and (d, f) active lesions and 3-year-old
scars. Each line corresponds to one patient. Willcoxon test. CLA: cutaneous lymphocyte antigen; active 1: active lesions 1-year-old group;
active 3: active lesion 3-year-old group; scar 1: 1-year-old scars; scar 3: 3-year-old scars.

residual inflammatory activity in scars, which was always

DISCUSSION
In this study, we evaluated scars and active lesions of 18
patients with ATL. Two groups were formed. The 1- and
3-year-old scars were not from the same patients because
of ethical conflicts. However the two groups were consid-
ered homogeneous as regard to both clinical and laborato-
rial parameters. Furthermore, several authors have
compared different patients with different clinical courses
to investigate different time points of ATL follow-up
(16,23–25).
Complete re-epithelization of the lesion, absence of
hyperaemia or oedema was used as criteria of cure. Com-
parison of lesions and scars in each patient showed a sig-
nificant reduction in the intensity of the inflammatory
infiltrate in cicatricial tissues. However, this reduction
occurred in a gradual and slow manner between the first
and third year after clinical cure. The histopathology of
some cases showing a discrete to moderate inflammatory
infiltrate in 3-year-old scars suggests that this process
might be even lengthier. Using histopathological examina-
tion, the presence of residual cell clusters in scars was
already detected (15). However, no immunological study
was performed. Our results indicate the persistence of
delimited and restricted to cell clusters and around blood
vessels resembling the dermal perivascular units (26).
Intense replacement of the skin architecture with cicatri-
cial tissue was observed in 1-year-old scars. However, cell
clusters were still observed. The concentration of cells
present at these sites was similar to those observed in
active lesions despite of a marked reduction in the inten-
sity of the inflammatory infiltrate per tissue area. Other
authors (19), studying mucosal lesions suggested that the
persistence of parasite antigens might be related to the
maintenance of the local inflammatory response. The par-
asite persistence was already described in scars (11,13–15).
Thus, the demonstration in this study of significant differ-
ences in the percentage of CD3+, CD4+ and CLA+ cells in
scars when compared to healthy skin samples agrees with
this hypothesis.
T lymphocytes were significantly reduced in 3-year-old
scars. However, the concentration of macrophages was
similar to that observed in active lesions despite a more
restricted distribution since areas of fibrosis predominated
in the tissue. These results indirectly indicate that the
scars’ areas where macrophages are concentrated remain
with characteristics of inflammatory activity. As parasites

290  2010 Blackwell Publishing Ltd, Parasite Immunology, 32, 285–295

Volume 32, Number 4, April 2010 Signs of inflammation in ATL scars

Active lesion 1-year scar Active lesion 3-year scar

80 80
(a) (b)

expressing cells
70 P = 0.028 70 P = 0.011

Percentage of
60 60
50 50
Ki-67 40 40
30 30
20 20
10 10
0 0

Active 1 Scar 1 Active 3 Scar 3

Ki-67 Ki-67

80 80
(c) (d)
expressing cells

70 P = 0.036 70 P = 0.021
Percentage of

60 60
50 50
40 40
BCl-2 30
30
20 20
10 10
0 0

Active 1 Scar 1 Active 3 Scar 3

BCI-2 BCI-2

80 (e) 80
(f)
P = 0.44
expressing cells

70 70 P = 0.24
Percentage of

60 60

FasL 50 50
40 40
30 30
20 20
10 10
0 0

Active 1 Scar 1 Active 3 Scar 3

FasL FasL

80 80
(g) (h)
expressing cells

70 P = 0.028 70 P = 1.0
Percentage of

60 60
50 50
Fas 40
40
30 30
20 20
10 10
0 0

Active 1 Scar 1 Active 3 Scar 3

Fas Fas
Figure 5 Percentage of proliferative markers (Ki-67, BCl-2) and apoptosis inducers (FasL ⁄ Fas) (a, c, e, g) in active lesions vs. 1-year-old
scars and (b, d, f, h) in active lesions vs. 3-year-old scars. Willcoxon test. Active 1: active lesions 1-year-old group; active 3: active lesion
3-year-old group; scar 1: 1-year-old scars; scar 3: 3-year-old scars.

in the two positive samples were also observed in these
areas, the presence of macrophages and NOS2 expression
could maintain the active control of parasite burden. The
activation of macrophages depends on the cytokine profile
to which they are exposed (21). Macrophages exposed to
Th1 cytokines produce nitric oxide (NO) which is respon-
sible for the elimination of amastigotes (16,21). However,
in the presence of type-2 response, macrophages can par-
ticipate in the processes of cell debris removal and tissue
repair, which cause tissue fibrosis (21,27). Therefore, mac-
rophages participate in all steps of the inflammatory pro-
cess, including the onset of inflammation, maintenance
and resolution, as well as tissue repair (27). Some factors
controlling the inflammatory process and establishing the
parasite–host balance have been described for other infec-
tious diseases (28–30). The course of cutaneous leishmani-
asis has been shown to depend on the Th1 ⁄ Th2 balance,
with a predominantly Th1 response leading to resistance
and cure, whereas a predominantly Th2 response results in
susceptibility and progressive disease (31). Thus, the envi-
ronment generated by these cytokines may permit the per-
sistence of pathogens if they are not eliminated during the
initial phases of infection. Macrophages are activated by
the cytokine milieu. Very few data is available regarding
ATL scars, but some results have pointed the importance
of cytokine balance during the healing of ATL lesions
(23,32,33). In addition, recently interesting results in clini-
cal presentation of visceral leishmaniasis (VL) were pub-
lished showing the importance of cytokines such TGF-b
as well as mast cells in the healing of lesions and inflam-
matory reaction control. Other authors (34) suggested the
role of TGF-b produced by Kupffer cells in the perihepato-
cytic fibrosis in residual hepatomegaly after healing of VL.
In the same manner, the role of mast cells during the active

 2010 Blackwell Publishing Ltd, Parasite Immunology, 32, 285–295 291

F. N. Morgado et al. Parasite Immunology

Table 2 Differences in the percentage of cell markers observed when scars and healthy skin were compared

Cells or cellular 1-year-old scar % 3-Year-old scar % Healthy skin % 1-year-old scars vs. 3-year-old scars vs.
markers mean € SEM (range) mean € SEM (range) mean € SEM (range) healthy skin P healthy skin P

CD3 45Æ12 € 2Æ6 (37–58Æ7) 44Æ01 € 4Æ14 (30–56Æ1) 31Æ95 € 2Æ13 (26Æ1–36Æ2) 0Æ005 0Æ11
CD4 33Æ19 € 3Æ38 (15Æ7–51Æ7) 31Æ83 € 2Æ74 (18Æ7–45Æ7) 21Æ36 € 0Æ79 (19Æ7–24Æ1) 0Æ005 0Æ02
CLA 24Æ06 € 3Æ7 (7–36Æ3) 27Æ16 € 3Æ43 (8Æ6–38Æ9) 37Æ06 € 1Æ87 (30–40Æ1) 0Æ013 0Æ019
BCl-2 15Æ85 € 4Æ48 (0–35Æ8) 26Æ92 € 4Æ2 (12Æ6–45) 13Æ38 € 1Æ31 (10Æ1–16Æ3) 1Æ00 0Æ028

as well as the healing phase of ATL has been pointed by sev- activated macrophage can be verified by the expression of
eral authors (19,23,35). In this context, the cytokines pres- NOS2 which is responsible to convert arginine in NO, which
ent in cicatricial tissue are currently being studied. But has an important role during the Leishmania clearance in

(a) (a1) (a2)

(a3) (a4)

(a5) (a6)

(b) (b1) (b2)

Figure 6 Immunohistochemical detection

of CD4+ (a.1, a.2), CD8+ (a.3, a.4),
CD1a+ (a.5, a.6), expression of E-selectin
(b.1) and CLA+ (b.2) cells prepared as
described in Materials and Methods.
Active lesion (a.1, a.3, a.5); scar (a.2, a.4,
a.6, a.1, a.2). The arrows indicate positive
cells. Original magnification 200· (scale
bar = 50 lm).

292  2010 Blackwell Publishing Ltd, Parasite Immunology, 32, 285–295

Volume 32, Number 4, April 2010
Table 3 Distribution of the patients according to the intensity of nitric oxide synthase type 2 (NOS2) and E-selectin expressions as well as
the duration of scar and active lesion
Active lesion or scar Absent Discrete
NOS2 Active lesion 0 2
1-year-old scar 1 3
Active lesion 0 0
3-year-old scar 3 6
E-selectin Active lesion 0 2
1-year-old scar 0 7
Active lesion 0 2
3-year-old scar 0 6
Discrete, at least one positive site per field; moderate, two or three positive sites; intense, four or five positive sites; very intense, more than
five positive sites.
the active lesions (16). We observed a significant reduction
in the expression of NOS2, which was even more marked
3 years after cure. It should be emphasized that despite this
reduction, expression of NOS2 could still be detected, a
finding that might be related to the mechanisms controlling
residual parasite burden. Other authors (13) demonstrated
the persistence of parasites in fibroblasts that presented a
reduced ability to produce NOS2. In addition, the action of
this enzyme on the control of parasite burden has been dem-
onstrated in a murine model (9). Within this context, the
detection of areas positive for NOS2 in 3-year-old scars sug-
gests the maintenance of the local activity.
Other cells that play a relevant role in the control of
parasite burden in leishmaniasis are CD8+ T lymphocytes
(18,36). In this study, the percentage of CD8+ cells
remained constant in 1-year-old scars, whereas a reduction
in the number of these cells was observed in 3-year-old
scars when compared to the respective active lesions. CD8
cells have been implicated in the mechanisms of healing in
cutaneous ATL (18). However, the unbalance of
CD8 ⁄ CD4 has been related with treatment fail in the
other clinical presentation. In a recent study, other authors
(23) have related increasing of CD8+ T cells and NK cells,
as well as, the reduction of IL-10 with relapse in mucosal
form of ATL. In this case, the patients would present high
parasite burden, consequently relapses are more frequent
(37).
In this study, the presence of the parasite could only be
detected in two 3-year-old scars. This fact might be related
to the small number of microorganisms associated with
their heterogeneous distribution at the affected site. In the
other hand, inflammatory markers observed in the two
cases are similar to the other scars examined. Reactivation
after clinical cure has already been demonstrated and can
occur as a result of trauma or immunosuppression (38,39).
The factors that cause reactivation have not been com-
pletely demonstrated; furthermore this data highlights the
 2010 Blackwell Publishing Ltd, Parasite Immunology, 32, 285–295
Signs of inflammation in ATL scars
Moderate Intense Very intense
1 2 4
5 0 0
2 5 2
0 0 0
2 2 3
2 0 0
2 1 4
3 0 0
hypothesis that even patients with spontaneous healing
needs the specific antimonial treatment since it could
reduce the parasite burden. In addition, patients who pres-
ent therapeutic failure should be followed-up after clinical
cure.
A reduction of inflammatory activity characterized by a
decrease in the expression of NOS2, E-selectin, Ki67,
BCl2 and Fas was observed in recent scars and was even
more intense in old scars. This finding, together with the
maintenance of expression of apoptosis inducers (FasL+),
suggests that the healing process is a slow phenomenon at
the microscopic level despite clinical cure of the lesion.
Reactivation after clinical cure may occur as a result of
exacerbation of residual inflammatory activity both in a
specific (increased parasite replication) and in a non-
specific manner (i.e. local trauma).
Other authors (40) studying the reaction caused by the
intradermal injection of purified protein derivative of
tuberculin (PPD) observed that the inflammatory process
starts with a proliferative and recruitment phase character-
ized by the expression of CLA, Ki67 and BCl-2. After
elimination of the antigen, there is the resolution phase
characterized by a reduction in BCl-2+ cells and an
increase in FasL+ cells, permitting healing of the lesion.
Furthermore, the authors suggested that in chronic lesions,
such as those observed in ATL, the control of these
phases of the inflammatory processes is abnormal. This
data indicate that in ATL these phases of the inflamma-
tory process overlap. Greenhalgh (6) discussed that resolu-
tion of the inflammatory phase requires the elimination of
the initial antigen stimulus. As the parasite may persist in
cured lesions (9,11–15), this process may become disorga-
nized in the presence of antigen. These data may explain
the observation of an inflammatory infiltrate, although
discrete, amidst intense cicatricial fibrosis in three of the
18 scars studied. Our results and those reported by others
(15) suggest that clinical cure does not always coincide
293
F. N. Morgado et al.
with histological cure. The finding of the persistence of
inflammatory activity in healed lesions and its gradual
decline with increasing healing duration supports this con-
clusion.
Taken together, these results indicate the maintenance of
areas with potential inflammatory activity along the scar-
forming fibrotic tissue, especially in more recent scars. The
observation that the number of lymphocytes starts to
decline only after a minimum period after clinical cure
supports this hypothesis.
On the basis of this data we suggest that: (i) the dynam-
ics of the inflammatory process in the cutaneous forms of
ATL maintains a pattern of cell distribution that changes
slowly even after apparent healing of the lesions, (ii) anal-
ysis of 3-year-old scars showed a marked reduction of
inflammatory activity markers such as NOS2, Ki67 and
BCl-2 and of lymphocytes, suggesting that the in situ para-
site–host balance starts to be established around this per-
iod, (iii) these results could suggest that the parasite–host
balance is established gradually even after clinical cure of
active lesions and it might be disrupted in situations of
immunosuppression or trauma with consequent reactiva-
tion of the disease.
ACKNOWLEDGEMENTS
This work was supported by IOC and IPEC-FIOCRUZ,
PAPES 4 and 5, FUNASA ⁄ MS, CNPq and FAPERJ,
Brazil. The authors thank Dr Sergio Coutinho and Dr
Joseli Oliveira-Ferreira (IOC-FIOCRUZ) for the critical
review of the manuscript, Mr Genilton Jos Vieira and Mr
Rodrigo Mexas for the figure artwork as well as Dr Maria
de Ftima Madeira for the anti-Leishmania antibody.
FNM and LPQ are PhD students from IPEC-FIOCRUZ.
REFERENCES
1 Vieira-GonÅalves R, Pirmez C, Jorge ME, et al. Clinical fea-
tures of cutaneous and disseminated cutaneous leishmaniasis
caused by Leishmania (Viannia) braziliensis in Paraty, Rio de
Janeiro. Int J Dermatol 2008; 47: 926–932.
2 Oliveira-Neto MP, Mattos M, Souza CS, Fernandes O &
Pirmez C. Leishmaniasis recidiva cutis in New World cutane-
ous leishmaniasis. Int J Dermatol 1998; 37: 846–849.
3 Kupper TS & Fuhlbrigge RC. Immune surveillance in the skin:
mechanisms and clinical consequences. Nat Rev Immunol 2004;
4: 211–222.
4 Streilein JW. Skin-associated lymphoid tissues (SALT): origins
and functions. J Invest Dermatol 1983; 80: 12s–16s.
5 Bos JD & Kapsenberg ML. The skin immune system: progress
in cutaneous biology. Immunol Today 1993; 14: 75–78.
6 Greenhalgh DG. The role of apoptosis in wound healing. Int J
Biochem Cell Biol 1998; 30: 1019–1030.
7 Hunt TK, Hopf H & Hussain Z. Physiology of wound healing.
Adv Skin Wound Care 2000; 13: 6–11.
294
Parasite Immunology
8 Henry G & Garner WL. Inflammatory mediators in wound
healing. Surg Clin North Am 2003; 83: 483–507.
9 Stenger S, Donhauser N, Thring H, Rçllinghoff M & Bogdan
C. Reactivation of latent leishmaniasis by inhibition of induc-
ible nitric oxide synthase. J Exp Med 1996; 183: 1501–1514.
10 Conceiżo-Silva F, Hahne M, Schrçter M, Louis J & Tschopp
J. The resolution of lesions induced by Leishmania major in
mice requires a functional Fas (APO-1, CD95) pathway of
cytotoxicity. Eur J Immunol 1998; 28: 237–245.
11 Schubach A, Marzochi MC, Cuzzi-Maya T, et al. Cutaneous
scars in American tegumentary leishmaniasis patients: a site of
Leishmania (Viannia) braziliensis persistence and viability ele-
ven years after antimonial therapy and clinical cure. Am J Trop
Med Hyg 1998; 58: 824–827.
12 Bogdan C & Rçllinghoff M. How do protozoan parasites sur-
vive inside macrophages? Parasitol Today (Regul Ed) 1999; 15:
22–28.
13 Bogdan C, Donhauser N, Dçring R, Rçllinghoff M, Diefen-
bach A & Rittig MG. Fibroblasts as host cells in latent leish-
maniosis. J Exp Med 2000; 191: 2121–2130.
14 Schubach A, Cuzzi-Maya T, Oliveira AV, et al. Leishmanial
antigens in the diagnosis of active lesions and ancient scars of
American tegumentary leishmaniasis patients. Mem Inst
Oswaldo Cruz 2001; 96: 987–996.
15 MendonÅa MG, de Brito MEF, Rodrigues EHG, Bandeira V,
Jardim ML & Abath FGC. Persistence of leishmania
parasites in scars after clinical cure of American cutaneous
leishmaniasis: is there a sterile cure? J Infect Dis 2004; 189:
1018–1023.
16 Morgado FN, Schubach A, Rosalino CMV, et al. Is the in situ
inflammatory reaction an important tool to understand the
cellular immune response in American tegumentary leishmani-
asis? Br J Dermatol 2008; 158: 50–58.
17 Convit J, Ulrich M, Fernndez CT, et al. The clinical and
immunological spectrum of American cutaneous leishmaniasis.
Trans R Soc Trop Med Hyg 1993; 87: 444–448.
18 Da-Cruz AM, Conceiżo-Silva F, Bertho AL & Coutinho SG.
Leishmania-reactive CD4+ and CD8+ T cells associated with
cure of human cutaneous leishmaniasis. Infect Immun 1994; 62:
2614–2618.
19 Amato VS, de Andrade HF & Duarte MIS. Mucosal leishmani-
asis: in situ characterization of the host inflammatory response,
before and after treatment. Acta Trop 2003; 85: 39–49.
20 Gaze ST, Dutra WO, Lessa M, et al. Mucosal leishmaniasis
patients display an activated inflammatory T-cell phenotype
associated with a nonbalanced monocyte population. Scand J
Immunol 2006; 63: 70–78.
21 Meneghin A & Hogaboam CM. Infectious disease, the innate
immune response, and fibrosis. J Clin Invest 2007; 117: 530–
538.
22 Madeira MF, Schubach A, Schubach TMP, et al. Mixed infec-
tion with Leishmania (Viannia) braziliensis and Leishmania
(Leishmania) chagasi in a naturally infected dog from Rio de
Janeiro, Brazil. Trans R Soc Trop Med Hyg 2006; 100: 442–
445.
23 Tuon FF, Duarte MIS & Amato VS. A quantitative and mor-
phometric study of mast cells in cutaneous leishmaniasis. Para-
site Immunol 2008; 30: 641–645.
24 Da-Cruz AM, Bittar R, Mattos M, et al. T-cell-mediated
immune responses in patients with cutaneous or mucosal leish-
maniasis: long-term evaluation after therapy. Clin Diagn Lab
Immunol 2002; 9: 251–256.
 2010 Blackwell Publishing Ltd, Parasite Immunology, 32, 285–295
Volume 32, Number 4, April 2010
25 Da-Cruz AM, Bertho AL, Oliveira-Neto MP & Coutinho SG.
Flow cytometric analysis of cellular infiltrate from American
tegumentary leishmaniasis lesions. Br J Dermatol 2005; 153:
537–543.
26 Bos JD, Zonneveld I, Das PK, Krieg SR, van der Loos CM &
Kapsenberg ML. The skin immune system (SIS): distribution
and immunophenotype of lymphocyte subpopulations in nor-
mal human skin. J Invest Dermatol 1987; 88: 569–573.
27 Fujiwara N & Kobayashi K. Macrophages in inflammation.
Curr Drug Targets Inflamm Allergy 2005; 4: 281–286.
28 Murray PJ. Defining the requirements for immunological con-
trol of mycobacterial infections. Trends Microbiol 1999; 7: 366–
372.
29 Lder CG, Gross U & Lopes MF. Intracellular protozoan par-
asites and apoptosis: diverse strategies to modulate parasite-
host interactions. Trends Parasitol 2001; 17: 480–486.
30 Lder CGK & Gross U. Apoptosis and its modulation during
infection with Toxoplasma gondii: molecular mechanisms and
role in pathogenesis. Curr Top Microbiol Immunol 2005; 289:
219–237.
31 Uzonna JE & Bretscher PA. Anti-IL-4 antibody therapy causes
regression of chronic lesions caused by medium-dose Leish-
mania major infection in BALB ⁄ c mice. Eur J Immunol 2001;
31: 3175–3184.
32 Osorio Y, Melby PC, Pirmez C, Chandrasekar B, Guarn N &
Travi BL. The site of cutaneous infection influences the immuno-
logical response and clinical outcome of hamsters infected with
Leishmania panamensis. Parasite Immunol 2003; 25: 139–148.
33 Gomes-Silva A, de Cssia Bittar R, Dos Santos Nogueira
R, et al. Can interferon-gamma and interleukin-10 balance
 2010 Blackwell Publishing Ltd, Parasite Immunology, 32, 285–295
Signs of inflammation in ATL scars
be associated with severity of human Leishmania (Viannia)
braziliensis infection? Clin Exp Immunol 2007; 149: 440–
444.
34 Duarte MIS, de Andrade HF, Takamura CFH, Sesso A &
Tuon FF. TGF-beta and mesenchymal hepatic involvement
after visceral leishmaniasis. Parasitol Res 2009; 104: 1129–
1136.
35 de Oliveira MP, Lima MCR, Calheiros AS, et al. Leishmania
(Viannia) braziliensis: human mast cell line activation induced
by logarithmic and stationary promastigote derived-lysates.
Exp Parasitol 2005; 109: 72–79.
36 Belkaid Y, Piccirillo CA, Mendez S, Shevach EM & Sacks DL.
CD4+CD25+ regulatory T cells control Leishmania major per-
sistence and immunity. Nature 2002; 420: 502–507.
37 van der Meide WF, Peekel I, van Thiel PPAM, et al. Treat-
ment assessment by monitoring parasite load in skin biopsies
from patients with cutaneous leishmaniasis, using quantitative
nucleic acid sequence-based amplification. Clin Exp Dermatol
2008; 33: 394–399.
38 Kubar J, Marty P, Lelivre A, et al. Visceral leishmaniosis in
HIV-positive patients: primary infection, reactivation and
latent infection. Impact of the CD4+ T-lymphocyte counts.
AIDS 1998; 12: 2147–2153.
39 Czechowicz RT, Millard TP, Smith HR, Ashton RE, Lucas SB
& Hay RJ. Reactivation of cutaneous leishmaniasis after sur-
gery. Br J Dermatol 1999; 141: 1113–1116.
40 Orteu CH, Poulter LW, Rustin MH, Sabin CA, Salmon M &
Akbar AN. The role of apoptosis in the resolution of T cell-
mediated cutaneous inflammation. J Immunol 1998; 161: 1619–
1629.
295
Copyright of Parasite Immunology is the property of Blackwell Publishing Limited and its content may not be
copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written
permission. However, users may print, download, or email articles for individual use.