Reduced efficacy of treatment of strongyloidiasis in HTLV-I carriers related to enhanced expression of IFN-γ and TGF-β1

Abstract

Strongyloidiasis, a human intestinal infection caused by Strongyloides stercoralis (S. stercoralis), is difficult to cure with drugs. In particular, a decrease of the efficacy of treatment has been reported in patients dually infected with S. stercoralis and human T-cell leukaemia virus type I (HTLV-I), both of which are endemic in Okinawa, Japan. However, the factors influencing this resistance remain unclear. In the present study, patients infected with S. stercoralis, with or without HTLV-I infection, were treated with albendazole, followed up for one year and separated into two groups, cured and non-cured. The cure rate of S. stercoralis was lower in HTLV-I carriers (P < 0·05). Serum levels of S. stercoralis-specific IgA, IgE, IgG, IgG1 and IgG4 antibodies were estimated, and a decrease of IgE (P < 0·05) and an increase of IgG4 (P < 0·05) were observed in the non-cured group, especially in HTLV-I carriers. RT-PCR of cytokines using peripheral blood mononuclear cells revealed that S. stercoralis patients with HTLV-I showed a high frequency of expression of IFN-γ and TGF-β1, whereas those without HTLV-I showed no expression of these cytokines. IFN-γ- and TGF-β1-positive HTLV-I carriers showed a decrease of IgE (P < 0·05), an increase of IgG4 (P < 0·01) and a lower cure rate (P < 0·01) compared with those who were negative for both cytokines. These results suggest that persistent infection with HTLV-I affected S. stercoralis-specific immunity and reduced therapeutic efficacy.

INTRODUCTION

Strongyloidiasis is due to a human intestinal nematode infection of Strongyloides stercoralis (S. stercoralis) which causes chronic bowel problems. Strongyloidiasis is relatively common in tropical and subtropical areas, such as the south-western islands and Okinawa in Japan, where human T-cell leukaemia virus type I (HTLV-I) is also endemic [1,2]. HTLV-I is a persistent virus, infecting 10–20 million people worldwide. Most infected people remain healthy, but 1–2% develop tropical spastic paraparesis/HTLV-I-associated myelopathy [3,4], and a further 2–3% develop adult T-cell leukaemia (ATL) [5,6].

An elevated proportion of HTLV-I carriers among patients with S. stercoralis has been reported [7], and ATL is frequently observed in patients dually infected with S. stercoralis and HTLV-I [8,9]. A decrease in the efficacy of treatment of S. stercoralis has also been reported in dually-infected patients [10,11]. However, the factors involved in resistance to treatment remain unknown.

In general, there are two factors determining the effectiveness of an antimicrobial drug. The first is its pharmacological effect, including specific cytotoxicity and pharmacokinetics. The other factor is host immunity [12]. In HTLV-I carriers, modulation of host immunity, such as a decrease of IgE, has been reported [13,14], and this may be one of the reasons for the reduced efficacy of treatment of strongyloidiasis. HTLV-I has a unique genomic region which can transactivate the transcription of various cellular genes as well as that of the virus genome. Among the genes that are transactivated are those of cytokines such as IFN-γ and TGF-β1 [15,16]. These cytokines have been reported to influence the level of production of IgE antibody [17–19]. Therefore, examining the relation between cytokines and S. stercoralis-specific antibodies in HTLV-I carriers should help to clarify the factors which are involved in the resistance of strongyloidiasis to treatment. Shimamoto et al. [20] reported an increase of IFN-γ production in cultured peripheral blood mononuclear cells (PBMC) from HTLV-I carriers. Moreover, Neva et al. [21] reported the relationship between an increase of IFN-γ production in the cultured PBMC and a decrease of IgE in patients who had S. stercoralis infections and were HTLV-I carriers. However, up-regulation of the expression of HTLV-I in cultured PBMC has also been reported [22] and therefore, examination of the in vivo expression of cytokines is needed in patients who have S. stercoralis infections and are HTLV-I carriers.

The purpose of this study was to determine the factors related in vivo to host immunity that influence the resistance to treatment of S. stercoralis infection in HTLV-I carriers. In the current study, we demonstrated that reduced efficacy of treatment for strongyloidiasis was closely related to overexpression of IFN-γ and TGF-β1 in HTLV-I carriers.

MATERIALS AND METHODS

Study population

The efficacy of treatment was evaluated in 79 patients with S. stercoralis, including 32 HTLV-I-positive (17 males and 15 females) and 47 HTLV-I-negative (30 males and 17 females) patients. The ages (mean ± s.d.) were 65·1 ± 11·6 and 66·7 ± 8·73, respectively. Twenty-one S. stercoralis-negative individuals were also enrolled in this study, including nine HTLV-I positive (seven males and two females) and 12 HTLV-I-negative (nine males and three females) individuals. Patient ages (mean ± s.d.) were 62·9 ± 8·04 and 63·3 ± 8·16, respectively. All patients in this study were diagnosed with S. stercoralis by agar plate faecal culture [23] at the 1994 annual regional health examination performed in Okinawa prefecture, Japan. Informed consent was obtained from all patients. Protocols involving human subjects were approved by the regional review boards of the University of the Ryukyus.

HTLV-I antibody

Individuals seropositive for HTLV-I were identified by the particle agglutination test (Serodia HTLV-I; Fujirebio, Tokyo, Japan) and also by an indirect immunofluorescence assay [24].

Treatments

Because of lower side effects and availability, albendazole was used in this study. The agent was administered after the diagnosis at a dosage of 400 mg/day for three consecutive days. The same therapeutic course was repeated after 2 weeks. The efficacy of treatment was assessed by stool examination three times: at 2 weeks, 6 months and 1 year after treatment. Stool examinations were performed three times for each patient at different time points. Cured cases were free of parasites in all three stool examinations (at 2 weeks, 6 months and 1 year). All other cases were assessed as non-cured.

Antigen

Somatic S. stercoralis filariform larval antigen (S.s. Ag) was prepared as previously described with minor modifications [25]. Briefly, third-stage filariform larvae were collected from faecal cultures obtained from parasite-free laboratory-reared beagles experimentally infected with a human strain of S. stercoralis. The larvae were washed five times in phosphate-buffered saline (PBS) containing antibiotic-antimycotic (1/100: Life Technologies, Inc., Rockville, MD, USA) and gentamicin reagent solution (1/200: Life Technologies, Inc.), washed again three times in sterile PBS, and frozen for storage at – 70°C. After sufficient numbers of larvae were collected, they were thawed and resuspended in sterile PBS containing 0·2 mm aminoethyl benzenesulphonylfluoride (Calbiochem, San Diego, CA, USA), 1·0 mm ethylene-diamine-tetraacetic acid (Wako Pure Chemical, Osaka, Japan), 1·0 mm leupeptin (Wako Pure Chemical) and 1·0 mm pepstatin A (Wako Pure Chemical). The suspended larvae were then homogenized with a Teflon homogenizer and fragmented by a 2 min sonication on ice. The suspension of fragmented larvae was stirred in PBS for 18 h at 4°C to extract antigenic components. The supernatant fluid was collected by centrifugation at 8000 g for 1 h, filtered through a 0·45 μm pore size membrane filter (Acrodisc; Gelman Sciences, Ann Arbor, MI, USA) and stored at – 70°C until use. The protein concentration was determined using a Micro BCA kit (Pierce, Rockford, IL, USA).

Determination of specific antibody titre to S. stercoralis

Specific antibodies to S.s. Ag were measured before treatment as previously described with minor modifications [26]. Briefly, ELISA plates (Luminoplate; Labsystems, Helsinki, Finland) were coated overnight at 4°C with S.s. Ag (5 μg/ml) and blocked with 0·2% blocking reagent (Boehringer Mannheim, Mannheim, Germany) in 0·1% Tween 20 (Wako Pure Chemical) in PBS for 2 h at 37°C. Plates were incubated with serum at an optimal dilution for each antibody class or subclass (IgA: 1/15 000, IgE: 1/100, IgG: 1/60 000, IgG1: 1/10 000, IgG4: 1/15 000) overnight at 4°C. Horseradish peroxidase (HRP)-conjugated mouse anti-human IgG1, IgG4 (Southern Biotechnology Associates, Birmingham, AL, USA) or HRP-conjugated goat anti-human IgA, IgE or IgG (Biosource International, Camarillo, CA, USA) was added to each well and the plates were incubated for 1 h at room temperature. After washing with 0·1% Tween 20 in PBS, substrate (Super Signal Substrate; Pierce) was added and the luminescence intensity was read with a microplate reader (Luminoskan; Labsystems). For each isotype–antigen combination, standard sera were assayed and assigned units. The antibody levels of the samples were expressed as units relative to the standard serum calculated by the following formula:

serum antibody units = counts per minute (CPM) of the sample/CPM of the standard serum (appropriately diluted) × 100.

Extraction of RNA from the PBMC

Heparinized blood was collected before treatment and the PBMC were separated by density gradient centrifugation using Lymphoprep (Nicomed, Oslo, Norway) and stored at – 70°C until use. Total cytoplasmic RNA was extracted from the cytoplasmic fraction using vanadyl ribonucleoside complex [27].

RT-PCR

Reverse transcription-PCR (RT-PCR) was employed to study cytokine gene expression in the PBMC, as described before [28]. Complementary DNA (cDNA) was synthesized from 1μg total RNA using random hexamer (Perkin Elmer, Foster City, CA, USA) in a total volume of 25 μl, and this mixture was diluted 1:5 with sterile distilled water after the reaction was finished. A 2μl aliquot of the cDNA was amplified by PCR. The following sense and antisense primer sequences were used: IFN-γ, 5′-AAGAGTGTGGAGACCAT CAA-3′, 5′-CTGACTC CTTTTTCGCTTCC-3′ [28]; TGF-β1, 5′-AAGTG GATCCACG AGCCCAA-3′, 5′-GCTGCACTTGCAGGAG CGCA-3′ [29]; IL-4, 5′-GTGCGATATC ACCTTACAGG-3′, 5′-TTCAGGAATCGGATCAGCTG-3′ [28]; IL-13, 5′-ACGGTCA TTGCTCTCACTTGCC-3′, 5′-CTTCCCGCCTACCCAAGA CATT-3′ [30]; β-actin, 5′-ATGGATGATGATATCGCCGCG-3′, 5′-TTCTCCATGTCGTCCCAGTTG-3′ [28]. All the primer pairs used in the gene expression study were designed to be located in different exons of the target genes so that the PCR products from cDNA templates could be distinguished easily from those of genomic DNA. The reaction volume of 20 μl contained each primer at 0·5 μm, each of the four nucleotide triphosphates at 200 μm (Perkin Elmer), 1 mm magnesium chloride and 0·5 U of AmpliTaq polymerase (Perkin Elmer). The amplification was performed in a thermal cycler (GeneAmp PCR system 9600; Perkin Elmer) with a programme of 40 cycles consisting of denaturation at 94°C for 1 min, annealing at 52°C (for IFN-γ), 55°C (for IL-4, IL-13 and β-actin) or 65°C (for TGF-β1) for 1 min, and extension at 72°C for 1 min. The PCR products were analysed by electrophoresis on 1·5% agarose gels and visualized by ethidium bromide staining.

Statistical analysis

The Mann–Whitney U-test and the χ² test were used to analyse the statistical significance of differences.

RESULTS

The efficacy of treatment in HTLV-I-positive and -negative groups of patients with S. stercoralis

Thirty-two HTLV-I-positive and 47 HTLV-I-negative patients with S. stercoralis were treated with albendazole. Of the 32 HTLV-I-positive patients with S. stercoralis, 13 were cured. The cure rate in the HTLV-I-positive group (40·6%) was significantly lower than that in the HTLV-I-negative group (66·0%) (P < 0·05) (Table 1).

Table 1.

Effect of concurrent HTLV-I infection on treatment ofstrongyloidiasis N

	Number of patients with S. stercoralis
HTLV-I	Cured	Non-cured
Positive (n = 32)	13	19
Negative (n = 47)	31	16

P < 0·05.

Comparison of S. stercoralis-specific antibody titres between HTLV-I-positive and HTLV-I-negative patients, and between cured and non-cured patients

To examine the relationship of HTLV-I infection and the efficacy of treatment, S. stercoralis-specific antibody titres, S. stercoralis-specific IgA, IgE, IgG, IgG1 and IgG4, were compared between 31 HTLV-I-positive (13 cured and 18 non-cured) and 46 HTLV-I-negative (30 cured and 16 non-cured) patients. The S. stercoralis-specific IgE antibody titre was significantly lower and the S. stercoralis-specific IgG4 antibody titre was significantly higher in the non-cured than in the cured patients with S. stercoralis in HTLV-I carriers (P < 0·05). No significant differences were observed in the other antibody subtypes ⁽Table 2). S. stercoralis-specific IgE and IgG4 were therefore suspected of influencing the efficacy of strongyloidiasis treatment in HTLV-I carriers.

Table 2.

Comparison of S. stercoralis-specific antibody titres between cured and non-cured patients in HTLV-I-pisitive and -negative groups

	S. stercoralis-specific antibody titre (log10):median (range)
	IgA		IgE		IgG		IgG1		IgG4
Patients	Cured	Non-cured	Cured	Non-cured	Cured	Non-cured	Cured	Non-cured	Cured	Non-cured
HTLV-I (+)(n = 31)	1·16 (1·45)	1·29 (1·76)	0·42 (1·64)*	0·00 (1·17)*	1·43 (1·70)	1·54 (2·63)	1·05 (1·57)	1·25 (2·60)	1·50 (2·71)*	2·07 (2·90)*
HTLV-I (−)(n = 46)	0·85 (2·09)	1·22 (1·79)	0·71 (2·08)	0·19 (1·70)	1·25 (2·12)	1·52 (0·90)	0·82 (2·05)	1·15 (1·84)	1·01 (2·48)†	1·79 (2·57)†

^*p < 0·05.

^†P < 0·05.

Cytokine expression of PBMC

To investigate the factors which influence S. stercoralis-specific antibody titres, the expression of cytokines in PBMC was analysed by RT-PCR. Both IFN-γ and TGF-β1 were expressed in all HTLV-I carriers not infected with S. stercoralis (nine out of nine) and they were expressed in some of the patients dually infected with S. stercoralis and HTLV-I (IFN-γ: 16 out of 27; TGF-β1: 17 out of 27). However, their expression was observed in very few of the patients infected only with S. stercoralis (IFN-γ: 0 out of 24; TGF-β1: one out of 24) and normal controls (IFN-γ: 0 out of 12; TGF-β1: one out of 12). IL-4 and IL-13 were not observed in most patients, either HTLV-I-positive or HTLV-I-negative patients. Representative results are displayed in Fig. 1.

Fig. 1.

Comparison of the expression of IFN-γ, TGF-β1, IL-4 and IL-13 among the patients with or without S. stercoralis or HTLV-I. Sixteen IFN-γ-positive and 17 TGF-β1-positive patients were observed among 27 patients co-infected with S. stercoralis and HTLV-I. There were no IFN-γ-positive and one TGF-β1-positive patient observed among 24 patients with S. stercoralis but not HTLV-I. Both IFN-γ and TGF-β1 were positive in all HTLV-I carriers without S. stercoralis (nine of nine), while very few expressions were found in normal controls (0 of 12 for IFN-γ and one of 12 for TGF-β1). Very few expressions and no significant differences were observed for IL-4 and IL-13 in all groups. Representative results are displayed. M, marker; –, negative control; +, positive control.

The influence of IFN-γ and TGF-β1 on S. stercoralis-specific IgE and IgG4 antibody titres

Because IFN-γ and TGF-β1 were mostly observed in HTLV-I-positive carriers, and their expression has been reported to be involved in the production of IgE or IgG4, the relationships between the expression of IFN-γ or TGF-β1 and S. stercoralis-specific IgE or IgG4 antibody titres were examined. The S. stercoralis-specific IgE antibody titre was significantly lower (P < 0·01) and the S. stercoralis-specific IgG4 antibody titre was higher (P < 0·1) in IFN-γ-positive or in TGF-β1-positive patients than in patients negative for both cytokines. This effect was even more significant in patients who expressed both IFN-γ and TGF-β1 (P < 0·05). However, the S. stercoralis-specific IgG4 antibody titre was not higher in patients who expressed only TGF-β1. Therefore, most of the effects on the S. stercoralis-specific IgG4 levels were suspected to have been influenced by IFN-γ (Table 3).

Table 3. Influence of the expression of IFN-γ and TGF-β1 on Strongyloides stercoralis-specific IgE and IgG4.

The influence of IFN-γ and TGF-β1 on the efficacy of treatment in the patients with S. stercoralis

The cure rate was analysed to determine the influence of IFN-γ and TGF-β1 on the efficacy of treatment. In the patients with S. stercoralis, the cure rate was lower in the patients who were positive for either IFN-γ or TGF-β1 than in the patients who were negative for both cytokines (P < 0·01). The cure rate was also lower in patients who expressed both IFN-γ and TGF-β1 (P < 0·01) (Table 4). The expression of IFN-γ and TGF-β1 was therefore suspected of influencing the efficacy of treatment in the patients with S. stercoralis in HTLV-I carriers.

Table 4. Influence of IFN-γ and TGF-β1 positivity on treatment of strongyloidiasis.

DISCUSSION

Because S. stercoralis is more difficult to cure in cases with immunosuppression, such as acquired immunodeficiency syndrome patients [31], host immunity in strongyloidiasis is undoubtedly important in the efficacy of treatment, as is true in schistosomiasis [12]. In the present study, we showed that the cytokine expression of HTLV-I carriers infected with S. stercoralis allowed their classification into IFN-γ- and TGF-β1-positive and -negative groups. Moreover, we showed that these cytokines may have affected the efficacy of treatment for strongyloidiasis through effects on S. stercoralis-specific IgE in the IFN-γ- or TGF-β1-positive group, and effects on IgG4 antibodies in the IFN-γ-positive group.

Modulation of host immunity, including impaired response of IgE [13,14], has already been reported in HTLV-I carriers [32,33]. Because IgE and mast cells are suspected of being involved in the protective immunity against strongyloidiasis [14,34], the low level of IgE may reduce the efficacy of treatment through the impairment of IgE–mast cell interaction in HTLV-I carriers infected with S. stercoralis.

IgG4 has been reported to be an interesting antibody subclass with respect to parasitic infections [25,35,36], and an elevation of the antigen-specific IgG4 antibody titre has been reported to enhance the efficacy of immunotherapy for rhinitis or asthma [37,38]. It has been suggested that IgG4 may block IgE-mediated protective effector functions in Schistosoma mansoni [39]. Genta et al. [25] reported that serum factors capable of competing with IgE for S. stercoralis antigens are present on the basophil surface, and suggested that such a mechanism could block the IgE–basophil system in strongyloidiasis. These findings suggest that the specific IgG4 antibodies act as blocking antibodies against a series of immunological reactions mediated through specific IgE antibodies. Our findings that serum IgG4 levels are elevated in HTLV-I carriers infected with S. stercoralis suggested that similar mechanisms may operate in S. stercoralis infection, especially in HTLV-I carriers.

Estimation of the mRNA expression of cytokines IFN-γ, TGF-β1, IL-4 and IL-13 in the PBMC of HTLV-I carriers revealed reduced expression of IFN-γ and TGF-β1 in S. stercoralis-positive compared with S. stercoralis-negative individuals. This suggested that there was inhibition of the expression of IFN-γ and TGF-β1 by S. stercoralis infection in HTLV-I carriers, although the mechanism of inhibition by S. stercoralis infection was not determined in this study. On the other hand, in S. stercoralis patients, HTLV-I infection stimulated the levels of expression of IFN-γ and TGF-β1 in PBMC (Fig. 1), and resulted in lower efficacy of treatment. IFN-γ has been reported to block IL-4-induced IgE production [40], and it has also been reported that high IFN-γ to IL-4 ratios enhance IgG4 production and simultaneously decrease IgE production [41,42]. IFN-γ has also been reported to exert an inhibitory effect of the functions of mast cells [43,44]. TGF-β1 is a multifunctional cytokine capable of a variety of immunological effects. It has been reported to inhibit the production of IgE antibody [45], and to inhibit the viability of and the release of histamine from mast cells [46,47]. These studies indicate that IFN-γ and TGF-β1 may affect the IgE–mast cell function. Antibody-dependent cell-mediated cytotoxicity (ADCC) has been reported as another form of protective immunity against strongyloidiasis, and eosinophils are suspected of being concerned with ADCC in strongyloides infection [48–50]. IFN-γ has been suggested to decrease the peripheral eosinophil level [51,52], and TGF-β1 has also been reported to inhibit eosinophil survival and function [53,54]. A decrease of eosinophils has also been reported in HTLV-I carriers [55]. Therefore, IFN-γ and TGF-β1 may also influence ADCC through the modulation of eosinophil activity.

In conclusion, persistent infection of HTLV-I elevated the expression of IFN-γ and TGF-β1 in patients with S. stercoralis, and resulted in reduced efficacy of treatment for S. stercoralis through a decrease and increase of the serum levels of specific IgE and IgG4, respectively.