Anda di halaman 1dari 11

Parasite Immunology, 2002, 24, 499 – 509

Cellular responses and cytokine profiles in Ascaris lumbricoides and


Immune
ORIGINALresponse
Blackwell to intestinal
Publishing
ARTICLE Ltd. helminthic co-infections

Trichuris trichiura infected patients

STEFAN M. GEIGER1,2, CRISTIANO L. MASSARA1, JEFFREY BETHONY1,3, PETER T. SOBOSLAY2, OMAR S. CARVALHO1
& RODRIGO CORRÊA-OLIVEIRA1

1
Fundação Oswaldo Cruz, Centro de Pesquisas René Rachou, Laboratório de Imunologia, Avenida Augusto de Lima 1715, 30190-002 Belo
Horizonte-MG, Brazil, 2Institute for Tropical Medicine, University of Tübingen, Wilhelmstr. 27, 72074 Tübingen, Germany and 3Southwest
Foundation of Biomedical Research, San Antonio, Texas, USA

SUMMARY Keywords Ascaris lumbricoides, cytokines, immunity, intestinal


helminths, Trichuris trichiura
The impact of intestinal helminth infection, i.e. Ascaris lum-
bricoides and Trichuris trichiura, on cellular responsiveness and
cytokine production was investigated in young adults. Ascaris- INTRODUCTION
specific cellular responsiveness was higher in parasite-free
Intestinal nematode infections are among the most pre-
endemic controls than in patients infected with T. trichiura, or
valent diseases world-wide, affecting at least one-quarter of
A. lumbricoides, or patients co-infected with both parasites. Also,
the world’s population (1). The recent World Health Organ-
mitogen-induced tumour necrosis factor (TNF)-α, interleukin
ization (WHO) evaluation estimates are that 1·4 billion
(IL)-12 and interferon (IFN)-γ secretion by peripheral blood
individuals are infected with Ascaris lumbricoides, 1 billion
mononuclear cells (PBMC) was higher in negative endemic
with Trichuris trichiura and 1·3 billion with hookworms (2).
controls than in infected individuals. Ascaris antigen-specific
Although such chronic infections are normally not fatal, they
production of TNF-α, IL-12 and IFN-γ was low in singly
cause a high degree of morbidity (3). However, in heavily
Ascaris as well as in co-infected patients, whereas secretion of
infected individuals thousands of deaths per year occur
IL-10 and IL-13 was elevated and similarly high in all patient
due to intestinal obstruction, massive dysentery syndrome
groups. The detection of Trichuris-specific and Ascaris-specific
or severe iron-deficiency anaemia (2).
IgG4 revealed significantly higher serum antibody levels in
In experimental infections it was shown that intestinal
Trichuris or Ascaris patients when compared to endemic
helminth infections will skew host immune responses
controls (P < 0·05), whereas parasite-specific IgE antibody
towards a type 2 cytokine profile, which leads to IgE pro-
levels were similarly high in infected individuals and in endemic
duction, mast cell activation, mucus secretion and will
controls. In summary, chronically infected Ascaris and
ultimately lead to parasite elimination (4,5). In accordance
Trichuris patients with a high parasite load presented reduced
with these experimental infections, patients with an Ascaris
cellular reactivity and lower type 1 TNF-α, IFN-γ and IL-12
lumbricoides infection have highly polarized type 2 cytokine
responses when compared with endemic controls, whereas
responses (6) and children, putatively immune against A.
type 2 IL-10 and IL-13 productions were similar in all groups
lumbricoides, have significantly elevated IgE levels specific for
from the endemic area. The former may support parasite
a recombinant Ascaris allergen (7). Furthermore, Needham
persistence, whereas substantial type 2 cytokine release may
et al. (8) showed a protective role for IgA in immunity to
promote protective immunity, suggesting an adaptation of
T. trichiura in areas of intense transmission. However, apart
the host to control the parasite burden while minimizing
from these few reports, knowledge on the immune responses
immune-mediated host self-damage.
induced in humans infected with one or more intestinal
helminth species remains scarce.
In endemic tropical areas polyparasitism is commonly
Correspondence: Present address: Dr Stefan Geiger, Institute found, and such coexistence of several parasite species may
for Comparative Tropical Medicine and Parasitology, University
not only profoundly influence the host–parasite interaction
of Munich, Leopoldstraße 5, 80802 München, Germany
(e-mail: stefan.geiger@tropa.vetmed.uni-muenchen.de). (9,10), but also confound the development of acquired
Received: 15 February 2002 immunity against individual parasites. Experimental co-
Accepted for publication: 30 October 2002 infections with intestinal helminth parasites showed that a

© 2002 Blackwell Science Ltd 499


S. M. Geiger et al. Parasite Immunology

primary helminth-induced Th 2-type immune response (NEG) with no recent reports of helminth infections were
alters the outcome of a concurrent infection with a second from the urban area of Belo Horizonte. Twenty millilitres of
helminth parasite (11,12). In humans, however, the impact blood were collected, from randomly selected individuals, in
of multiple parasite infections on expression of immunity is a heparinized Vacutainer tube (Becton Dickinson, Franklin
controversial and remains poorly studied (13,14). In the Lakes, NJ). For differential blood counts and for a haemo-
present study, we investigated the humoral and cellular gram, 5 mL were collected in a Vacutainer tube containing
immune responses in children and young adults infected with EDTA (Becton Dickinson) and analysed in a Coulter counter
one or two of the common geohelminths, A. lumbricoides (Coulter Corporation, Miami, FL). For differential blood
and /or T. trichiura, and we observed that co-infection and a counts, 5 µL of blood were put on a slide, dried, and stained
high parasite load will not only exhaust parasite-specific with a fast-staining solution (Laborclin, Pinhais-Paraná,
cellular responsiveness but will also reduce type 1 cytokine Brazil) and 100 white blood cells were counted under a light
responses. microscope (400 × magnification) to determine eosinophilia
in the peripheral blood. Furthermore, blood was collected
in a non-heparinized tube and, after centrifugation, the serum
MATERIALS AND METHODS
samples were stored at −70°C until further use.
Study population
Mitogen and antigen preparation
This study was conducted in two rural communities, Motas
and Pires, about 70 km south of Belo Horizonte, the capital For polyclonal stimulation of PBMC phytohaemagglutinin-
of the state of Minas Gerais, Brazil. Both communities L (PHA-L, Difco Laboratories, Detroit, MI) was used. Ascaris
were situated close to each other and had the same socio- lumbricoides (Ascaris antigen: ASC-Ag) and T. trichiura adult
economic conditions. The region around Motas and Pires worms (Trichuris antigen: TRIC-Ag) were collected from
is characterized by small agricultural units and a big mining treated individuals from the same endemic area. Worms
area. Most of the households in the two communities have were extensively washed in cold phosphate buffered saline
no access to clean water supplies and appropriate sanita- (PBS), cut in pieces and transferred into a tissue grinder in
tion. In Motas a total of 378 subjects were examined for a small volume of PBS. The worms were homogenized on
gastrointestinal helminth infections, which covered over ice and subsequently centrifuged at 20 000 g for 40 min at
95% of the population. In Pires 201 children from the local 4°C. Schistosoma mansoni parasitic stages were obtained
primary and secondary school were examined. All Ascaris/ from Swiss mice experimentally infected at the animal facil-
Trichuris positive subjects received a standard treatment of ities of the Centro de Pesquisas René Rachou-FIOCRUZ in
6 × 100 mg mebendazole during three consecutive days. Belo Horizonte. Schistosoma mansoni soluble adult worm
Those individuals with Taenia sp. or with hookworm infec- (SWAP) and egg antigens (SEA) were prepared as previ-
tion received a 6 × 200 mg mebendazole or a single 400 mg ously described (16). All PBS-soluble antigen fractions were
dose of albendazole, respectively. Subjects with a Schisto- passed through a low protein binding 0·2 µm filter (Millipore,
soma mansoni infection received praziquantel treatment Bedford, MA) and the protein concentration was determined
from the local health authority. This study was approved by according to the method of Lowry et al. (17).
the Ethical Committee of the Centro de Pesquisas René
Rachou. Participation in the study was dependent on written
Peripheral blood mononuclear cell (PBMC) proliferation
consent, and in the case of children, written consent was
assays
given by their parents or guardians.
PBMC were isolated on a ficoll–diatriozoate cushion
(1·077 g/L, lymphocyte separation medium; Organon
Subject sampling and parasitology
Teknika, Charleston, SC) by centrifugation at 400 g for
Individuals were examined for intestinal helminth infections 40 min at room temperature (RT). PBMCs were collected,
using the Kato–Katz technique (15). Mean egg counts per washed three times with minimal essential medium (Gibco,
gram of faeces (epg) were calculated from at least two slides Grand Island, NY), and resuspended in RPMI 1640
per sample and up to 5000 eggs per slide were counted medium (Gibco), containing 5% heat-inactivated normal
under a microscope. Egg-negative endemic controls (EC) human AB serum, antibiotic/antimycotic solution (penicil-
were examined twice with two slides each time and within lin: 100 U/mL, streptomycin: 100 µg/mL, amphotericin-B:
1 week before enrolment in the study. Participants had not 0·25 µg /mL; Sigma, St. Louis, MO), and 2 m -glutamine
received chemotherapy during the last 6 months before (Winlab, Leicestershire, United Kingdom). The cell suspensions
enrolment into the study. Negative non-endemic controls were adjusted to 1·25 × 106 cells /mL and 200 µL per well were

500 © 2002 Blackwell Science Ltd, Parasite Immunology, 24, 499–509


Volume 24, Number 11/12, November/December 2002 Immune response to intestinal helminthic co-infections

seeded in 96-well tissue culture plates (Costar, Corning, NY). (3·1 pg /mL), IFN-γ (9·4 pg /mL), and TNF-α (15·6 pg/mL).
All cell culture experiments were performed in a humidified For IL-12, the detection limit was assessed to be at around
incubator at 5% CO2 and at 37°C. Stimulation assays were 40 pg/mL.
performed in triplicate and mitogen and antigens were
added at previously determined concentrations, known to
Parasite-specific antibody reactivity in sera
result in optimal proliferation, i.e. PHA (4 µg /mL), ASC-Ag
(20 µg /mL); TRIC-Ag (20 µg/mL), SEA (25 µg /mL), SWAP Sera from infected patients, endemic controls and negative
(25 µg /mL). For mitogen stimulation, the cells were pulsed controls were screened for Ascaris and Trichuris antigen-
after 48 h of culture with tritiated thymidine (Amersham specific antibodies in an indirect ELISA. Briefly, ELISA
Pharmacia, Sao Paulo, Brazil; 0·5 µCi /well, specific activity plates (Nunc, Maxisorp, Naperville, IL) were coated over-
6·7 Ci/m) and harvested after further 18 h of culture. night at 4°C with 100 µL of ASC-Ag or TRIC-Ag (5 µg/mL)
Antigen-stimulated cell cultures were pulsed after 6 days diluted in carbonate buffer (15 m Na2CO3, 35 m NaHCO3,
with tritiated thymidine and harvested after an additional 6 h 0·02% NaN3, pH 9·6). The plates were washed twice with
of culture. Incorporated activity was determined in a liquid PBS/0·05% Tween-20 and incubated with the individual
scintillation counter and the data expressed as stimulation sera diluted in PBS/10% FCS (1 : 100 for IgG and IgA, and
indices SI (SI = mean proliferation of stimulated culture / 1 : 50 for IgG1, IgG4, and IgE) at 4°C overnight. On the
mean proliferation of unstimulated culture). following day, the plates were washed (5×) and 100 µL
per well of alkaline phosphatase-conjugated detection
antibody in PBS/10% FCS was added at the following
Cytokine production in PBMC supernatants
dilutions: IgG 1 : 1000 (Sigma-Aldrich, Steinheim, Germany);
PBMCs were isolated and adjusted as described above and IgG1, IgG4, IgA 1 : 1000 (Pharmingen); IgE 1 : 2000
600 µL per well were seeded in 48-well tissue culture plates (Pharmingen). The plates were incubated for 2 h at RT and
(Costar), and then stimulated with mitogen or worm anti- subsequently washed 7 times with PBS/0·05% Tween-20.
gens in the same concentrations as above. The cultures were In a final step, 100 µL of p-nitrophenylphosphate substrate
harvested at 48, 72, and 96 h, centrifuged, and the cell-free (Sigma) diluted in substrate buffer (0·5 m MgCl2, 10%
supernatants stored at −70°C until cytokine quantification. diethanolamine, pH 9·8) were added to the plates and the
Cytokine-specific enzyme-linked immunosorbent assays colorimetric reaction measured in an automated ELISA
(ELISA) were performed according to the manufacturer’s reader at 405 nm. Serum antibody reactivity for individual
instructions (Opti-EIA Sets; Pharmingen, Heidelberg, samples was performed in duplicate. Values from different
Germany) for the following cytokines: interleukin (IL)-10, plates were compared and standardized according to standard
IL-12, IL-13, interferon-γ (IFN-γ), and tumour necrosis reference sera from Ascaris- and Trichuris-infected patients
factor-α (TNF-α). Briefly, the plates were sensitized with the and endemic control sera. Values are expressed as mean optical
capture-antibody and incubated at 4°C overnight. After a densities and compared between the different patient groups.
blocking step with PBS/10% foetal calf serum (FCS) (Biomast,
Rio de Janeiro, Brazil) for 1 h at room temperature (RT),
Statistical analyses
the plates were washed twice with PBS/ Tween-20 (0·05%)
and subsequently incubated for 2 h at RT with the cell All values were checked for normal distribution by the
culture supernatants and the diluted cytokine standards. Shapiro–Wilk W-test. Differences between the five patient
After another washing step with PBS/ Tween-20, detection groups were checked for significance by using the nonpara-
antibody plus avidin–horseradish peroxidase were added to metric Wilcoxon test. Bivariate analysis was done by perform-
the wells and incubated for 1 h at RT. The plates were ing the nonlinear Spearman’s rank test. Significance levels
washed again (7×) and finally the substrate solution (tetram- were taken at the 5% level unless otherwise indicated. Due
ethylbenzidine, TMB) was added. Colorimetric reaction was to multiple testing, a Bonferroni–Holm test was performed
stopped with 2  H2SO4 and absorbance measured at in order to correct significance values in each experiment,
450 nm in an automated ELISA reader. Cytokine concen- and adjusted P-values are indicated in the figures and tables.
trations in supernatants were measured in duplicate and
were calculated according to standard curves. Mean cytokine
RESULTS
concentrations are expressed in pg /mL and are shown as
net production from which background cytokine secretion
Study group characteristics
has been subtracted. The detection limits of the different
cytokine assays were below the standard curves as recom- The prevalence of A. lumbricoides in Motas and Pires was
mended by the manufacturers for IL-10 (7·8 pg /mL), IL-13 43·7% and 53·7%, respectively. For T. trichiura infection the

© 2002 Blackwell Science Ltd, Parasite Immunology, 24, 499 –509 501
S. M. Geiger et al. Parasite Immunology

Table 1 Study characteristics in patients infected with Ascaris lumbricoides or Trichuris trichiura, and in uninfected control groups

ASC + TRIC ASC TRIC EC NEG


n = 16 n = 14 n = 14 n=9 n=8

Sex: M/F 10/6 5/9 7/7 3/6 3/5


Mean age, years 11·5 14·6 13·9 13·6 29·4
(7–19) (8–27) (9–25) (11–17) (25–32)
Ascaris lumbricoides, epg 10 944 12 454 0 0 0
(12–120 000) (12–120 000)
Trichuris trichiura, epg 227 0 171 0 0
(12–972) (24–1044)
Eosinophilia percentage 5·7 6·7 4·2 3·7 1·1

Eggs per gram stool (epg) are indicated as geometric mean. Numbers in parentheses indicate the range. A. lumbricoides eggs are counted for
a maximum of 5000 eggs per slide and set as 120 000 eggs per gram stool. M, male; F, female. Study groups: ASC + TRIC, A. lumbricoides
and T. trichiura co-infected; ASC, A. lumbricoides mono-infected; TRIC, T. trichiura mono-infected; EC, parasite-free endemic controls; NEG,
parasite-free non-endemic controls.

prevalence was 20·6% (Motas) and 37·8% (Pires). Individu- were not statistically significant. The response to S. mansoni
als with S. mansoni infection (prevalence: Motas: 0·5%; soluble egg antigens (SEA) was low in all groups. Interest-
Pires 1·5%) or hookworm infection (prevalence: Motas: ingly, an elevated cellular reactivity was observed in PBMC
0·8%; Pires 0·5%) had migrated into the study area and sup- from endemic and non-endemic controls after stimulation
posedly got this infection in other localities where these with SWAP. Here, mono-infected Ascaris and co-infected
helminth parasites are endemic. Demographic and parasito- patients showed the lowest proliferative responses, and dif-
logical data for the study groups are shown in Table 1. The ferences between co-infected patients and non-endemic
patient groups from the endemic area had no statistically controls were statistically significant (P < 0·05) (Figure 1). The
significant differences in their age distribution. The mean reactivity of PBMC in response to the mitogen PHA did not
infection intensity for A. lumbricoides in the mono- and co- differ between the patient groups (data not shown).
infected groups was moderate (A. lumbricoides mono- The in vitro production of cytokines by mitogen- and
infected: 12 454 epg; co-infected: 10 944), whereas patients antigen-stimulated PBMC is shown in Table 2a for PHA
with T. trichiura infection had light infections (T. trichiura and in Table 2b for Ascaris antigen stimulation. After
mono-infected: 171 epg; co-infected: 227 epg). Analysis of mitogen stimulation with PHA, PBMC from endemic and
variance () did not reveal significant differences in non-endemic controls showed an elevated production of
intensities of infection between the mono- and co-infected TNF-α. The differences in TNF-α secretion between
groups for both parasite species. Ascaris lumbricoides and T. endemic controls and non-endemic controls and mono-
trichiura patients presented low to moderate eosinophilia infected Ascaris patients were statistically significant
(Table 1), as determined by differential blood counts, and (P < 0·05). The PHA-induced IL-12 production was highest
none of the individuals showed signs of severe anaemia or in endemic controls and mono-infected Trichuris patients;
malnutrition. For parasite-specific ELISA, the number of those groups also presented significantly higher levels of
patients enrolled in each study group was increased as indi- IL-12 production than mono-infected Ascaris patients
cated below ( Table 3a,b). (P < 0·05). Similarly, IFN-γ production was elevated in
mitogen-stimulated PBMC from endemic controls, and was
significantly higher (P < 0·05) than that of mono-infected
Lymphocyte proliferation and cytokine production
Ascaris, mono-infected Trichuris patients or negative
The proliferation of PBMC after stimulation with parasite non-endemic controls (Table 2a). For PHA-induced IL-10
antigens is indicated as stimulation indices and presented in production, no significant differences between the patient
Figure 1. In response to Ascaris antigen, endemic controls groups were observed (Table 2a). The PHA-induced IL-
showed the highest cellular responses, whereas cellular reac- 13 production of PBMC was lowest in non-endemic con-
tivity in egg-positive Ascaris patients was in the range of trols when compared to the groups from the endemic area,
non-endemic negative controls. After stimulation with but these differences were not statistically significant
Trichuris antigens, PBMC from T. trichiura mono-infected (Table 2a).
patients and endemic controls showed the highest prolifera- For the cytokine production after antigenic stimulation,
tion when compared with the other groups, but the differences all patients responded most strongly to the A. lumbricoides

502 © 2002 Blackwell Science Ltd, Parasite Immunology, 24, 499–509


Volume 24, Number 11/12, November/December 2002 Immune response to intestinal helminthic co-infections

Figure 1 In vitro proliferation of peripheral blood mononuclear cells (PBMC), shown as stimulation indices (SI), in different patient groups
after stimulation with several helminth antigens. PBMC were stimulated with A. lumbricoides adult worm antigens (ASC-Ag, 20 µg /mL),
T. trichiura adult worm antigens (TRIC-Ag, 20 µg /mL), S. mansoni egg (SEA, 25 µg /mL) and adult worm antigens (SWAP, 25 µg /mL). Data
are shown as mean values ± standard error of the mean (SEM). Because of multiple testing a Bonferroni adjustment was performed and
P-values with P < 0·05 were considered significant and are indicated with an asterisk. Patient groups: ASC + TRIC, Ascaris and Trichuris
co-infected; ASC, Ascaris mono-infected; TRIC, Trichuris mono-infected; EC, endemic controls; NEG, non-endemic controls.

antigen preparation (Table 2b), whereas individually secreted IL-10 secretion and Ascaris egg counts reached statistical
cytokine levels were low for the T. trichiura, SEA, or SWAP significance (P < 0·01).
antigens, and frequently were at or below the detection All patients from the endemic region were tested for any
limits of the cytokine assays (data not shown). Following correlation with antigen- or mitogen-induced cytokine
A. lumbricoides antigenic stimulation, an elevated production production and age. Here, PHA-induced IL-13 secretion
of TNF-α and IL-12 was detected in culture supernatants resulted in a significantly negative correlation with age (rho =
from mono-infected Trichuris patients and from endemic −0·3170; P = 0·0207). The negative correlation of PHA-
and non-endemic controls, whereas their concentrations induced IL-13 production and age was even stronger when
were lowest in mono-infected Ascaris and co-infected only infected patients were considered (rho = −0·3803; P =
patients (Table 2b), albeit these differences were not statistically 0·0109). For the other cytokines no correlation with age
significant. Also, antigen-induced IFN-γ production was was found.
highest in endemic controls (Table 2b). For type 2 cytokines, When mitogen- and Ascaris antigen-induced cytokine
PBMC from non-endemic controls produced the lowest secretions in endemic individuals were checked for sex dif-
amounts of IL-10 and IL-13 in response to Ascaris antigens ferences, no statistically significant differences were found
when compared to infected patients and endemic controls for IL-10, IL-12, IL-13, and IFN-γ. However, when TNF-α
(Table 2b). Furthermore, mitogen- and Ascaris antigen- secretion was checked for sex differences in all individuals
induced IL-10 production was positively correlated with from the endemic area, the mean TNF-α secretion after
Ascaris egg counts both in mono-infected Ascaris (rho = stimulation with Ascaris antigen was significantly lower
0·4967 and rho = 0·3458) and in co-infected Ascaris and (P < 0·01) in female (107 ± 25 pg/mL) than in male individ-
Trichuris patients (rho = 0·6752 and rho = 0·2089). For the uals (230 ± 39 pg/mL). If TNF-α secretion was checked
co-infected patients the correlation between mitogen-induced only in infected patients, again, the mean TNF-α production

© 2002 Blackwell Science Ltd, Parasite Immunology, 24, 499 –509 503
S. M. Geiger et al. Parasite Immunology

Table 2a–b Mean cytokine production by PBMC cultures from patients with intestinal helminth infections and from negative controls.
(a) Cultures stimulated by PHA. (b) Cultures stimulated by Ascaris antigen

(a) PHA stimulation

ASC + TRIC ASC TRIC EC NEG


Cytokine Mean Mean Mean Mean Mean
pg/mL (± SEM) (± SEM) (± SEM) (± SEM) (± SEM)

TNF-α 407 211a*,b* 378 610a 592b


(± 68) (± 54) (± 57) (± 77) (± 134)
IL-12 52 22c*,d* 78c 100d 50
(± 12) (± 8) (± 14) (± 17) (± 12)
IFN-γ 2528 1015e* 1594e* 6511e 1741e*
(± 659) (± 184) (± 611) (± 1278) (± 183)
IL-10 325 452 296 302 513
(± 38) (± 99) (± 38) (± 60) (± 45)
IL-13 1097 960 1024 1054 826
(± 38) (± 75) (± 60) (± 77) (± 109)

(b) Ascaris-AG

ASC + TRIC ASC TRIC EC NEG


Cytokine Mean Mean Mean Mean Mean
pg/mL (± SEM) (± SEM) (± SEM) (± SEM) (± SEM)

TNF-α 119 123 253 176 192


(± 37) (± 37) (± 58) (± 43) (± 78)
IL-12 43 52 60 88 60
(± 16) (± 25) (± 14) (± 41) (± 21)
IFN-γ 52 17 81 95 69
(± 26) (± 6) (± 41) (± 35) (± 54)
IL-10 86 168 115 96 63
(± 15) (± 32) (± 29) (± 28) (± 13)
IL-13 75 73 111 125 50
(± 12) (± 14) (± 20) (± 36) (± 18)

Significant differences between the patient groups are indicated by letters a– e. P-values are corrected for multiple testing (*P < 0·05). Patient
groups: ASC + TRIC, Ascaris and Trichuris co-infected; ASC, Ascaris mono-infected; TRIC, Trichuris mono-infected; EC, endemic controls;
NEG, non-endemic controls.

was significantly lower in females than in males, both between the patient groups for the IgG1 reactivity against
after PHA stimulation (females: 254 ± 50 pg /mL; males: A. lumbricoides adult worm antigens (Table 3a). Compared
416 ± 53 pg/mL; P < 0·05) and after stimulation with Ascaris to endemic and non-endemic controls, significantly
antigen (females: 93 ± 27 pg /mL; males: 232 ± 44 pg /mL; increased levels of Ascaris-specific IgG4 were detected in
P < 0·05). co-infected patients (P < 0·05); furthermore, mono-infected
Ascaris patients showed significantly (P < 0·05) higher IgG4
levels than endemic controls (Table 3a). Specific IgE was
Parasite-specific antibody responses
elevated in sera of all patient groups from the endemic area
Parasite-specific antibody reactivities against A. lumbri- when compared to the non-endemic controls. For mono-
coides and T. trichiura adult worm antigens are presented infected Trichuris patients and co-infected patients these
in Table 3a and Table 3b, respectively. For A. lumbricoides- differences were statistically significant (P < 0·05) (Table 3a).
specific reactivity, total IgG was significantly elevated With respect to parasite-specific IgA, no significant differ-
(P < 0·05) in all infected patient groups when compared to ences were detected between the groups, although reactivity
the non-endemic negative controls (Table 3a). Higher levels was higher in the infected than in the negative patient
of parasite-specific total IgG were also detected in sera from groups (data not shown).
endemic controls when compared to the non-endemic con- The T. trichiura-specific reactivity of total IgG in the
trols (Table 3a). No significant differences were observed patient groups is shown in Table 3b. In mono-infected T.

504 © 2002 Blackwell Science Ltd, Parasite Immunology, 24, 499–509


Volume 24, Number 11/12, November/December 2002 Immune response to intestinal helminthic co-infections

Table 3a A. lumbricoides-specific IgG, IgG1, IgG4 and IgE reactivity in sera from patients with intestinal helminth infections and from
negative controls as determined by ELISA

Ascaris-Antigen

ASC + TRIC ASC TRIC EC NEG


Ig Mean OD Mean OD Mean OD Mean OD Mean OD
isotype (± SEM) (± SEM) (± SEM) (± SEM) (± SEM)

IgG 0·71a* 0·77a* 0·72a* 0·56 0·29a


(± 0·09) (± 0·10) (± 0·16) (± 0·07) (± 0·04)
IgG1 0·33 0·26 0·37 0·27 0·2
(± 0·04) (± 0·03) (± 0·11) (± 0·05) (± 0·06)
IgG4 0·81b*,c* 0·4b* 0·4 0·13b 0·13c
(± 0·16) (± 0·11) (± 0·14) (± 0·04) (± 0·06)
IgE 0·34d* 0·13 0·34d* 0·3 0·05d
(± 0·08) (± 0·04) (± 0·09) (± 0·13) (± 0·02)

Table 3b T. trichiura-specific IgG, IgG1, IgG4, and IgE reactivity in sera from patients with intestinal helminth infections and from negative
controls as determined by ELISA

Trichuris-Antigen

ASC + TRIC ASC TRIC EC NEG


Ig Mean OD Mean OD Mean OD Mean OD Mean OD
isotype (± SEM) (± SEM) (± SEM) (± SEM) (± SEM)

IgG 1·25a 0·49a*,b* 1·21b 0·55a*,b* 0·36a*,b*


(± 0·16) (± 0·06) (± 0·18) (± 0·11) (± 0·06)
IgG1 0·57c 0·21c*,d* 0·54d 0·25c*,d* 0·25c*
(± 0·08) (± 0·02) (± 0·08) (± 0·06) (± 0·08)
IgG4 1·27e 0·18e*,f* 1·05f 0·18e*,f* 0·17e*,f*
(± 0·19) (± 0·05) (± 0·20) (± 0·07) (± 0·10)
IgE 0·47 0·23 0·49 0·54 0·20
(± 0·11) (± 0·05) (± 0·11) (± 0·18) (± 0·05)

Shown are the mean optical density (OD) values, with SEM values in parentheses. Significant differences (*P < 0·05) between the patient
groups are indicated by letters a–f. P-values are corrected for multiple testing. Patient groups: ASC + TRIC (n = 23), Ascaris and Trichuris
co-infected; ASC (n = 27), Ascaris mono-infected; TRIC (n = 19), Trichuris mono-infected; EC (n = 17), negative endemic controls; NEG
(n = 9), negative non-endemic controls.

trichiura patients or co-infected patients, significantly ele- co-infected Trichuris patients as well as in endemic controls,
vated (P < 0·05) total IgG levels were found when compared however, these differences were not statistically significant
with mono-infected Ascaris patients or with endemic or when compared to the other groups.
non-endemic controls (Table 3b). In the mono-infected Pearson product correlation between egg counts and
Trichuris patients, parasite-specific IgG1 reactivity was antibody isotype levels revealed a significant positive correla-
higher when compared to mono-infected Ascaris patients tion between T. trichiura egg counts and IgG4 antibody
or endemic controls (both P-values < 0·05). Co-infected levels against T. trichiura antigen in co-infected patients
patients had significantly higher IgG1 levels than mono- (rho = 0·419; P < 0·01).
infected Ascaris patients, endemic controls, or non-endemic
controls (P < 0·05) (Table 3b). Similarly, Trichuris patients
DISCUSSION
and Ascaris/Trichuris co-infected individuals showed stronger
IgG4 reactivity than mono-infected Ascaris patients, endemic Intestinal helminth parasites are among the most prevalent
controls or non-endemic controls (P < 0·05) (Table 3b). For causes of tropical diseases in developing countries, and con-
Trichuris-specific IgE (Table 3b) as well as IgA (data not current infections with multiple helminth species are fre-
shown), serological reactivity was elevated in mono- and quently found in endemic populations. Up to date, however,

© 2002 Blackwell Science Ltd, Parasite Immunology, 24, 499 –509 505
S. M. Geiger et al. Parasite Immunology

only few studies have investigated the effects of multiple pheral cytokine responses corresponded to those occurring
and chronic intestinal helminth infections on the cellular at the site of the infection. Therefore, peripheral cytokine
immune response in infected humans. Normally, these responses may provide valuable information on the parasite-
parasites cause chronic and non-fatal infections (3), even induced cytokine environment. After stimulation with
though it was recently shown that these helminth infections mitogen, our results disclosed a dominant type 1 cytokine
cause T cell alterations and changes in the T cell activation production in endemic controls, represented by elevated
profiles and, as such, exert a profound impact on the host’s levels of TNF-α, IL-12, and IFN-γ, whereas in patently
immune response (18). Furthermore, intestinal helminth infected patients those type 1 and pro-inflammatory
infections will not only induce parasite-specific immune cytokines were detected at lower amounts in their PBMC
responses, but also broadly activate immune responses, cultures. In addition, after stimulation with Ascaris antigen,
which may then interfere with other diseases and, as previ- less TNF-α, IL-12, and IFN-γ were detected in culture
ously shown, may impede (13,19) or improve (20) the capa- supernatants from singly Ascaris or Ascaris and Trichuris
bility of the host to defend itself against pathogens, or may co-infected patients when compared with endemic controls,
even impair the success of vaccination (21,22). Hence, it although this did not reach statistical significance. Previ-
remains important to understand the impact that common ously, no differences in the IL-2, IL-10, or IFN-γ secretion
intestinal helminth infections exert on the host’s immune after mitogen or antigen stimulation were reported between
response and to reveal to which extent parasite co-infections Ascaris-infected individuals and negative individuals from a
may activate, bias or exhaust the immune response of low-transmission area (6). Furthermore, Ascaris-infected
infected individuals, especially children. patients showed a markedly elevated IL-5 secretion after
In this study, we investigated the cellular immune response stimulation with mitogen or specific Ascaris antigens (6). In
and cytokine production of patients singly and co-infected with the present study, we examined parasite-free controls from
A. lumbricoides and T. trichiura. The proliferation of PBMC the same highly endemic area. These negative endemic con-
in response to Ascaris and Trichuris antigens was distinctly trols, in all likelihood, had previous contact with the two
different in the various patient groups. Those mono-infected intestinal helminthic species and, apart from possible pro-
with Ascaris or co-infected with Ascaris and Trichuris tective immune responses, chemotherapy in the past, use of
showed the lowest parasite-specific cellular reactivity, which plant extracts as therapeutic agents or very low-levels of
was in the range of that observed for non-endemic controls. infections cannot be excluded. In order to investigate the
In contrast, responses to Ascaris antigen were highest in immune response in an endemic area, we consider it appro-
endemic controls. Previously, Cooper et al. (6) reported that priate to include such endemic controls, because those
cellular reactivity to Ascaris adult worm and larval stage individuals represent an indigenous group who are exposed
antigens was higher in Ascaris-infected patients when com- but potentially not yet patently infected or that may have
pared to non-endemic controls. However, this may be attrib- acquired protective immunity to the infection. As such,
uted to the lower infection intensities, i.e. lower mean A. negative endemic controls were also used in studies where
lumbricoides egg counts (6728 epg), than those found in the the effect of intestinal helminth infections on vaccination
present study population. Such specifically suppressed cellu- efficiency was investigated (22).
lar reactivity in Ascaris and Trichuris co-infected as well as The lower TNF-α production in females, both after
mono-infected Ascaris patients may be caused by the high mitogen and antigen stimulation, when compared with male
intensity of infection found in the patients of this study. individuals from the endemic area is interesting. Since we
Furthermore, a pronounced cellular responsiveness was were not able to detect a sex-related difference in intensity of
observed in non-infected controls to S. mansoni SWAP, infection, which might explain the lower TNF-α production
whereas this reactivity remained low in helminth-infected in females, it may be explained by the different hormonal
patients. This may reflect not only possible cross-reactivities regulation in pre-adulthood influencing the patients’ immune
between antigenic epitopes, but also, that higher parasite response.
loads will influence negatively cellular responsiveness in In several animal models it has been shown that Th2-type
patients exposed to related parasite antigens, which is the cytokines such as IL-4, IL-5, and IL-13 play an important
case with concurrent helminth infections. role in the elimination of gastrointestinal nematodes (4,5).
In the present study, the secretion of cytokines by antigen- In our study groups, we investigated type 2 cytokine
and mitogen-stimulated PBMC was analysed in vitro. Al- responses, i.e. IL-13 and IL-10, in in vitro stimulated PBMC
though the peripheral blood immune response may not cultures, but we did not detect significant differences
reflect the local responses in the intestinal wall or the drain- between patients and controls, both after mitogen and
ing lymph nodes, in the mouse experimental model it was parasite antigen-specific stimulation. It is noteworthy that
recently shown, that in a T. muris infection (23) the peri- PBMC from non-endemic controls showed the lowest

506 © 2002 Blackwell Science Ltd, Parasite Immunology, 24, 499–509


Volume 24, Number 11/12, November/December 2002 Immune response to intestinal helminthic co-infections

production of IL-10 and IL-13 after stimulation with Ascaris controls, which again is indicative for previous and repeated
antigen. This supports the idea that our non-endemic parasite exposure. Therefore, specific IgE may be important
controls may truly represent negative controls. On the for the development of acquired immunity to A. lumbri-
other hand, recent history of exposure as well as of infection coides infection, as proposed by others (7,31), and should
has imprinted immunological memory and upon in vitro be further investigated, especially in those individuals con-
stimulation with parasite-specific antigens activated a stantly exposed, although without signs of infection. Such
mixed Th1- and Th2-type cytokine production, as observed IgE-mediated protective immune responses were also reported
in our endemic controls. The predominance of either type 1 in reinfection studies with S. haematobium or S. mansoni
or 2 T cell responses occurs only in a chronic helminth (32,33). In addition, specific IgG1 or IgG4 antibodies
infection in immunopathological reactions or in acquired against Trichuris crude adult worm antigens were con-
immunity (24). firmed to be useful tools to detect a patent T. trichiura
Interestingly, IL-13 secretion, which shares some of the infection with low cross-reactivity to A. lumbricoides antigens,
important properties with IL-4, was induced in PBMC cul- as already shown by Lillywhite et al. (34). Interestingly,
tures of patients and endemic controls, and this indicated in patients infected with T. trichiura and A. lumbricoides,
the generation of potentially protective Th2-type cytokine elevated parasite-specific reactivities of both IgG4 and IgE
responses as required for the expulsion of experimental were detected. In S. haematobium infections, elevated IgG4
gastrointestinal nematodes (5). In Ascaris-infected and co- antibody levels were postulated to be involved in the blockage
infected patients with a moderate to high intensity of infec- of the IgE protective response (32); however, the question
tion, cellular unresponsiveness was accompanied by a reduced to what extent IgG4 may block acquisition of protective
secretion of pro-inflammatory TNF-α and other type 1 immunity in ascariasis and trichiuriasis has to be determined
cytokines. Such a down-modulation of type 1 cytokines in further studies.
and inflammatory responses may favour parasite persistence In summary, we observed elevated levels of TNF-α, IL-
but at the same time may serve to reduce intestinal path- 12, and IFN-γ production by PBMC from parasite-free
ology, as proposed for other intestinal helminth infections endemic controls, whereas these cytokines were suppressed
(25,26). As such, elevated local and systemic TNF-α levels in the patient groups with high parasite loads, e.g. mono-
were detected in children with a high intensity of T. trichiura infected Ascaris and co-infected Ascaris and Trichuris patients.
infection and Trichuris dysentery syndrome (TDS), and Whether these elevated levels of Th1-type cytokines
were associated with intestinal pathology (27), whereas play an essential role in parasite expulsion and protective
elevated TNF-α levels were not detected in low-intensity immunity, or whether they may just reflect the parasite-
trichiuriasis (28). free state of infection, has to be further investigated. In
The lack of substantial cytokine secretion after stimula- order to better determine those immune mechanisms that
tion with T. trichiura crude adult worm antigen was recently protect against infection, longitudinal studies remain
reported by different research groups (6,29), whereas ele- necessary to identify and characterize those individuals
vated secretion of IL-10 and TNF-α were detected in super- who remain resistant to infection over extended periods of
natants with whole blood cultures from Trichuris patients time. In filariasis and schistosomiasis, it has recently been
after stimulation with excretory–secretory antigens from shown that resistance in endemic normal individuals is asso-
T. trichiura or T. muris (29). ciated with a mixed rather than with a clearly polarized type
For the parasite-specific humoral immune responses, high 1 or type 2 immune response (35,36). In A. lumbricoides
levels of Ascaris- and Trichuris-specific IgG4 antibodies infection, natural immunity has been associated with a high
were detected in the sera of infected individuals, whereas specific IgE response, favouring a protective type 2 immune
they remained at low levels in both control groups. In co- response (7,31). However, further detailed studies on cellu-
infected Ascaris and Trichuris patients the levels of Trichuris- lar responses and cytokine production profiles in a well-
specific IgG4 antibodies correlated significantly with the defined group of resistant individuals are required to better
T. trichiura egg output. Although such a positive corre- define those parameters necessary for acquisition of protec-
lation was already found in an epidemiological study (30), tive immunity.
the same study showed a negative correlation between
Trichuris-specific IgA antibodies and intensity of infection
ACKNOWLEDGEMENTS
in adults. We were not able to detect such correlations for
parasite-specific IgA antibodies, which might be due to the This work was supported by grants from FAPEMIG and
lower mean age in our patients. In contrast to Cooper et al. FIOCRUZ. S. M. Geiger received postdoctoral research
(6), we were able to detect considerable levels of Trichuris- fellowships at institutions of higher education in the
and Ascaris-specific IgE antibodies in sera of our endemic Federal Republic of Germany from the German Academic

© 2002 Blackwell Science Ltd, Parasite Immunology, 24, 499 –509 507
S. M. Geiger et al. Parasite Immunology

Exchange Service (DAAD) and from Conselho Nacional activity by peripheral blood mononuclear cells from treated but
de Desenvolvimento Científico e Tecnológico (CNPq). J. not active cases of schistosomiasis. J Immunol 1983; 130: 2891–
2895.
Bethony is supported by an International Scientists Research
17 Lowry OH, Rosenbrough NJ, Farr AL, Randall RJ. Protein
Development Award, from the John C. Fogarty Center, measurement with the folin phenol reagent. J Biol Chem 1951;
NIH. 193: 265–275.
18 Kalinkovich A, Weisman Z, Greenberg Z, et al. Decreased CD4
and increased CD8 counts with T cell activation is associated with
REFERENCES chronic helminth infection. Clin Exp Immunol 1998; 114: 414 – 421.
1 Albonico M, Crompton DWT, Savioli L. Control strategies for 19 Gopinath R, Ostrowski M, Justement SJ, Fauci AS, Nutman TB.
human intestinal nematode infections. Adv Parasitol 1999; 42: Filarial infections increase susceptibility to human immun-
277–341. odeficiency virus infection in peripheral blood mononuclear
2 World Health Organization. The World Health Report 1996: cells in vitro. J Infect Dis 2000; 182: 1804 –1808.
Fighting Disease, Fostering Development. Geneva: WHO; 1996. 20 Nacher M, Gay F, Singhasivanon P, et al. Ascaris lumbricoides
3 Chan M-S. The global burden of intestinal nematode infections infection is associated with protection from cerebral malaria.
– fifty years on. Parasitol Today 1997; 13: 438– 443. Parasite Immunol 2000; 22: 107–113.
4 Grencis RK. Enteric helminth infection: Immunopathology and 21 Sabin EA, Araujo MI, Carvalho EM, Pearce EJ. Impairment of
resistance during intestinal nematode infection. Chem Immunol; tetanus toxoid-specific Th1-like immune responses in humans
1997; 66: 41–61. infected with Schistosoma mansoni. J Infect Dis 1996; 173: 269–272.
5 Finkelman FD, Wynn TA, Donaldson DD, Urban JF Jr. The 22 Cooper PJ, Chico ME, Losonsky G, et al. Albendazole treat-
role of IL-13 in helminth-induced inflammation and protective ment of children with ascariasis enhances the vibriocidal anti-
immunity against nematode infections. Curr Opin Immunol body response to the live attenuated oral cholera vaccine CVD
1999; 11: 420–426. 103-HgR. J Infect Dis 2000; 182: 1199–1206.
6 Cooper PJ, Chico ME, Sandoval C, et al. Human infection with 23 Taylor MD, Betts CJ, Else KJ. Peripheral cytokine responses to
Ascaris lumbricoides is associated with a polarized cytokine Trichuris muris reflect those occurring locally at the site of infec-
response. J Infect Dis 2000; 182: 1207–1213. tion. Infect Immun 2000; 68: 1815–1819.
7 McSharry C, Xia Y, Holland CV, Kennedy MW. Natural 24 Spellberg B, Edwards JE Jr. Type 1/type 2 immunity in infec-
immunity to Ascaris lumbricoides associated with immunoglob- tious diseases. Clin Infect Dis 2001; 32: 76–102.
ulin E antibody to ABA-1 allergen and inflammation indicators 25 Coutinho HB, Robalinho TI, Coutinho VB, et al. Immunocy-
in children. Infect Immun 1999; 67: 484 – 489. tochemistry of mucosal changes in patients infected with the
8 Needham CS, Lillywhite JE, Didier JM, Bianco TA, Bundy intestinal nematode Strongyloides stercoralis. J Clin Pathol
DAP. Comparison of age-dependent antigen recognition in two 1996; 49: 717–720.
communities with high and low Trichuris trichiura transmission. 26 Prociv P. Pathogenesis of human hookworm infection: insights
Acta Tropica 1994; 58: 87–98. from a ‘new’ zoonosis. Chem Immunol 1997; 66: 62–98.
9 Weidong P, Xianmin Z, Crompton DWT. Ascariasis in China. 27 MacDonald TT, Spencer J, Murch SH, et al. Immunoepidemi-
Adv Parasitol 1998; 41: 109–148. ology of intestinal helminth infections. 3. Mucosal macrophages
10 Pit DSS, Polderman AM, Baeta S, Schulz-Key H, Soboslay PT. and cytokine production in the colon of children with Trichuris
Parasite-specific antibody and cellular immune responses in trichiura dysentery. Trans Roy Soc Trop Med Hyg 1994; 88:
humans infected with Necator americanus and Oesophagosto- 265–268.
mum bifurcum. Parasitol Res 2001; 87: 722–729. 28 Karyadi E, Gross R, Sastroamidjojo S, Dillon D, Richards AL,
11 Hermanek J, Goyal PK, Wakelin D. Lymphocyte, antibody and Sutanto I. Antihelminthic treatment raises plasma iron levels
cytokine responses during concurrent infections between but does not decrease the acute-phase response in Jakarta
helminths that selectively promote T-helper-1 or T-helper-2 school children. Southeast Asian J Trop Med Public Health
activity. Parasite Immunol 1994; 16: 111–117. 1996; 27: 742–753.
12 Curry AJ, Else KJ, Jones F, Bancroft A, Grencis RK, Dunne 29 Turner J, Faulkner H, Kamgno J, Else K, Boussinesq M,
DW. Evidence that cytokine–mediated immune interactions Bradley JE. A comparison of cellular and humoral immune
induced by Schistosoma mansoni alter disease outcome in mice responses to trichuroid derived antigens in human trichuriasis.
concurrently infected with Trichuris muris. J Exp Med 1995; Parasite Immunol 2002; 24: 83–93.
181: 769–774. 30 Bundy DAP, Lillywhite JE, Didier JM, Simmons I, Bianco AE.
13 Chamone M, Marques CA, Atuncar GS, Pereira ALA, Pereira Age-dependency of infection status and serum antibody levels
LH. Are there interactions between schistosomes and intestinal in human whipworm (Trichuris trichiura) infection. Parasite
nematodes? Trans R Soc Trop Med Hyg 1990; 84: 557–558. Immunol 1991; 13: 629–638.
14 Booth M, Mayombana C, Kilima P. The population biology 31 Hagel I, Lynch NR, DiPrisco MC, Rojas E, Perez M, Alvarez N.
and epidemiology of schistosome and geohelminth infections Ascaris reinfection of slum children: relation with the IgE
among schoolchildren in Tanzania. Trans R Soc Trop Med Hyg response. Clin Exp Immunol 1993; 94: 80–83.
1998; 92: 491–495. 32 Hagan P, Blumenthal UJ, Dunn D, Simpson AJG, Wilkins HA.
15 Katz N, Chaves A, Pellegrino J. A simple device for quantitative Human IgE, IgG4 and resistance to reinfection with Schistosoma
stool thick-smear technique in schistosomiasis mansoni. Rev haematobium. Nature 1991; 349: 243–245.
Inst Med Trop Sao Paulo 1972; 14: 397–400. 33 Dunn DW, Butterworth AE, Fulford AJ, et al. Immunity after
16 Gazzinelli G, Katz N, Rocha RS, Colley DG. Immune treatment of schistosomiasis: association between IgE anti-
responses during human schistosomiasis mansoni X. Pro- bodies to adult worm antigens and resistance to reinfection. Eur
duction and standardization of an antigen-induced mitogenic J Immunol 1992; 22: 1483–1494.

508 © 2002 Blackwell Science Ltd, Parasite Immunology, 24, 499–509


Volume 24, Number 11/12, November/December 2002 Immune response to intestinal helminthic co-infections

34 Lillywhite JE, Bundy DAP, Didier JM, Cooper ES, Bianco AE. 36 Turaga PSD, Tierney TJ, Bennett KE, et al. Immunity to
Humoral immune responses in human infection with the whip- onchocerciasis. Cells from putatively immune individuals pro-
worm Trichuris trichiura. Parasite Immunol 1991; 13: 491–507. duce enhanced levels of interleukin-5, gamma interferon, and
35 Corrêa-Oliveira R, Caldas IR, Gazzinelli G. Natural versus granulocyte-macrophage colony-stimulating factor in response
drug-induced resistance in Schistosoma mansoni infection. Par- to Onchocerca volvulus larval and male worm antigens. Infect
asitol Today 2000; 16: 397–399. Immun 2000; 68: 1905–1911.

© 2002 Blackwell Science Ltd, Parasite Immunology, 24, 499 –509 509