Soluble CD26/CD30 levels in visceral leishmaniasis: markers of disease activity

Abstract

Leishmania infantum is the causative agent of zoonotic visceral leishmaniasis (VL). If untreated the disease could be fatal; however, in some cases the infection can run a subclinical course. In subclinical infections a Th1-response predominates, while Th2-responses and/or probably T_reg cells are related to unfavourable outcome of the disease in active VL. In the present study we determined the levels of soluble (s) CD26 and CD30 co-stimulatory molecules in sera from patients with active VL, asymptomatic individuals and healthy volunteers. Results showed a significant difference in both sCD26 and sCD30 between infected cases and normal individuals (P ≤ 0·001). However, there was no significant difference in sCD26 levels between asymptomatic cases and patients, although the difference was not significant. sCD30 levels were significantly higher in VL patients than asymptomatic cases (P ≤ 0·001). These findings suggest a possible association between sCD26 and sCD30 levels and the clinical manifestation of L. infantum infection.

Introduction

The leishmaniases are parasitic diseases with a wide range of clinical symptoms, and the clinical picture depends on determinants related to both the infecting species of leishmania and the host. Zoonotic visceral leishmaniasis (VL) caused by Leishmania infantum is potentially fatal; however, subclinical forms are relatively common [1,2]. In endemic areas L. infantum predominantly affects children and the disease is characterized by fever, hepatosplenomegaly, anaemia, leucopenia and hypergammaglobulinaemia [1–3]. In asymptomatic subjects who have no manifestations of leishmaniasis, however, seroconversion and positive skin test are evidence of previous infection [4–6]. Subclinically infected individuals respond to the infection by a Th1-type response characterized by interferon (IFN)-γ production [5,6]. However, there are conflicting reports on the dominance of interleukin (IL)-4-secreting Th2 cells in active VL [7–9]. High expression of IL-10 in active VL and its association with severity of disease suggest a contributing role for alternative suppressor pathways such as T_reg cells [10–14].

Recent reports have indicated that Th1/Th2 cells exhibit preferential expression of some co-stimulatory molecules such as CD26 and CD30, which are cell surface glycoproteins, and are also present as soluble forms in plasma (sCD26 and sCD30, respectively). These molecules are T cell activation markers expressed preferentially by T lymphocytes producing Th1- and Th2-type cytokines, respectively [15–18]. Several studies of sCD30 levels in plasma have shown correlations with Th2-associated conditions. However, little information was found about the levels of sCD26 in Th1-associated conditions. Serum levels of these markers have also been used to assess disease activity in cases of atopic dermatitis, chronic hepatitis C and B and systemic lupus erythematosus [19–22].

The aim of this study was to evaluate the clinical relevance of VL with serum levels of sCD26 and sCD30 and the informativeness of these soluble proteins as markers of Th1/Th2 response. To our knowledge, serum levels of sCD26 and sCD30 in L. infantum-infected individuals has not been reported previously.

Materials and methods

Subjects

The study was carried out in Meshkin-shahr, a city located in the north-west of Iran in an endemic area of VL. Study groups were as follows:

1. Patients diagnosed as having active VL based on clinical examination and on demonstration of leishmania parasites in bone marrow aspirates. All patients exhibited specific antibody titres of 1/3200 or more by direct agglutination test (DAT).

2. Individuals with asymptomatic VL, defined by positive DAT (specific antibody titres of 1/3200 or more), but no or few mild symptoms.

3. Healthy individuals from non-endemic areas without any history of leishmaniasis were selected as controls. After acquisition of informed consent blood samples were obtained from all these individuals and separated sera were stored at −20°C until use.

Assays

Serum levels of sCD26 and sCD30 were determined by a sandwich enzyme-linked immunosorbent assay (Bendermed System, Vienna, Austria) according to the manufacturer’s instructions. The assay system has a sensitivity of 78 ng/ml for sCD26 and 6·4 U/ml for sCD30.

Statistical analysis

Data were analysed using SigmaStat 2·0 (Jandel Scientific, San Rafael, CA, USA) software. The differences of sample medians were evaluated by the Wilcoxon–Mann–Whitney U-test and P < 0·05 was considered significant.

Results

Thirty-three children (18 female and 15 male) with clinical symptoms of VL and positive DAT were studied. The ages of these patients ranged from 1 to 10 years (mean 3·6 years). Fifteen individuals (eight female and seven male) with positive DAT but no hepatosplenomegaly and constitutional symptoms were included in the asymptomatic group. They were aged 2–11 years (mean 7·3 years). Twenty-seven healthy controls (12 female and 15 male) ranging in age from 3 to 14 (mean 6 years) were also included in the study.

Serum levels of sCD26 and sCD30 are shown in Fig. 1. Patients with active VL and asymptomatic children had significantly higher median levels of sCD26 than controls (P < 0·001) (Table 1). The median level for sCD26 was not significantly different between asymptomatic cases and patients with active disease.

Fig. 1.

Concentration of sCD26 (a) and sCD30 (b) in sera from different study groups.

Table 1.

Serum levels of sCD26 and sCD30 and the median ratio of sCD26 to sCD30.

Groups	Median sCD26 level (range) (ng/ml)	Median sCD30 level (range) (U/ml)	sCD26/sCD30 ratio
Control	630 (375–950)	35·2 (19·6–121·2)	15·5
Active VL	890 (540–1715)	256 (78–800)	3·3
Asymptomatic VL	995 (525–1440)	132 (54·8–244)	5·8

VL: visceral leishmaniasis.

Similarly, patients with active disease and asymptomatic children had significantly higher median levels of sCD30 compared to controls (P < 0·001) (Table 1). Furthermore, a statistically significant difference was observed for median sCD30 levels between VL patients and asymptomatic individuals (P < 0·001). Table 1 also shows that the sCD26/sCD30 ratio was higher in asymptomatic cases than patients and the difference was significant (P < 0·001). Levels of sCD26 and sCD30 did not show any significant correlation with age and sex in any of the groups studied.

Discussion

Visceral leishmaniasis caused by L. infantum predominantly affects children in endemic regions [1–3]. Although the disease is fatal if untreated, most infected children remain asymptomatic or exhibit few mild symptoms [4–6]. Epidemiological studies in Meshkin-shahr, an endemic region of VL in Iran, have revealed that the ratio of disease to infection is 1:13 [2]. In L. infantum infection asymptomatic individuals show a Th1-type response [5,6], whereas in active disease a clear Th2-type response has not been documented conclusively [7–9]. There is evidence that in addition to an aberrant Th2 response, T_reg cells activation can also promote susceptibility in leishmania-infected mice [14,23,24]. In both mice and humans, T_reg cells secrete high levels of IL-10. This cytokine is responsible for suppression of leishmania-specific Th1 immunity in mice [24,25]. It has also been shown that high IL-10 expression is associated with unfavourable clinical outcome in human leishmaniasis. Therefore, a role for T_reg cells in suppression of protective immunity has been postulated [10–14].

Measurement of cytokines in culture supernatants is usually used to classify the immune response into the Th1 and Th2 response. Recently, analysis of some surface molecules and serum soluble markers has been proposed to be a simple, useful and efficient tool in discriminating Th1/Th2 responses. CD26 and CD30 are T cell activation markers which are also present in soluble form in biological fluids. CD26 expression correlates with Th1-associated conditions, whereas elevated levels of sCD30 have been found in Th2-associated conditions such as chronic hepatitis C and B and autoimmune disorders [16,18,20,21,26]. Moreover, a correlation between sCD30 and sCD26 levels and disease activity has been reported [19–21,27,28]. Liao and colleagues [29] reported high levels of sCD30 and normal values of sCD26 in a patient with chronic hepatitis C before therapy; however, sCD30 decreased and sCD26 increased after responding to treatment. A correlation between sCD30 levels and disease activity has also been reported in patients with atopic dermatitis. Plasma level of sCD30 was higher in the exacerbation status of disease than after conventional therapy [19]. On the other hand, decreased levels of sCD26 in active disease and its increase in remission has been shown in anti-neutrophil cytoplasm antibody (ANCA)-associated vasculitides [30]. Similarly, an inverse correlation of sCD26 with disease activity has been reported in patients with systemic lupus erythematosus [22].

In the present study the levels of sCD26 and sCD30 in sera of L. infantum-infected individuals with and without symptoms were measured and compared with the values for normal cases. Our results indicate the higher levels of sCD26 in patients and asyptomatic children compared to those of normal controls, and although the sCD26 level for asymptomatic cases was higher than that obtained for patients with active disease, the difference was not significant. High sCD26 levels in asymptomatic individuals is in agreement with reports on the presence of Th1-dominated response in these cases [5–8]. Moreover, these data extend previous findings proposing the presence of a correlation between the expression of cellular and soluble CD26 and Th1-type responses [16–18,29].

Our data also show that sCD30 level is increased in individuals with L. infantum infection, and it is significantly higher in VL patients compared to asymptomatic cases. These results confirm other studies suggesting that high sCD30 levels are indicative of active disease in Th2-associated conditions.

The sCD26/sCD30 ratio is significantly higher in asymptomatic children than VL patients. It seems that detection of both markers may be more helpful to determine the type of immune response.

From these findings a possible association between sCD26 and sCD30 levels and clinical manifestation of the L. infantum infection could be suggested. Consistent with previous reports, we also found that sCD26 is more representative of Th1 response, whereas sCD30 is more indicative of a Th2 response or of pathways that favour suppression of an effective Th1 immune response. The presence of any possible correlation between CD30 expression and T_reg cells has yet to be defined.

In conclusion, it seems that combined analysis of sCD26 and sCD30 in serum is an easy-to-perform and dependable method and that these factors are reliable markers of disease activity which may be useful in assessing the outcome of the L. infantum infection.