Associations between helminth infection and CD4+ T cell count, viral load and cytokine responses in HIV-1-infected Ugandan adults

TRANSACTIONSOF THE ROYALSOCIETYOF TROPICALMEDICINEAND HYGIENE(2003) 97, 103- 108

Associations between helminth infection and CD4 ÷ T cell count, viral load and cytokine responses in HIV-l-infected Ugandan adults Alison M. Elliott 1'2, Patrice A. M a w a 1, Sarah Joseph 3, Proscovia B. Namujju 1, M o s e s Kizza 1, Jessica S. Nakiyingi ~, Christine Watera 1, D a v i d W. D u n n e 3 and James A. G. Whitworth ~'2 1 Uganda Virus Research Institute, P. 0. Box 49, Entebbe, Uganda; 2London School of Hygiene and Tropical Medicine, Keppel Street, London WC1E 7HT, UK; 3Department of Pathology, Cambridge University, Tennis Court Road, Cambridge CB2 1QP, UK Abstract It has been proposed that helminth infection may exacerbate HIV progression by promoting activation of 'type 2' immune responses. To examine this hypothesis, we investigated helminth infection in a cohort of HIV-l-seropositive adults in Entebbe, Uganda, during November 1999 to January 2000. Individuals with helminths were treated. At enrolment, after 5 weeks and after 4 months, CD4 + and CD8 + T cell counts and viral load were measured. Cytokine responses (interferon [IFN]-y, interleukin [ILl-2, IL-4 and IL-5) to Schiswsoma mansoni adult worm antigen (SWA), Mycobacterium tuberculosis culture filtrate proteins (CFPs) and phytohaemagglutinin (PHA) were measured in a whole blood assay. At baseline, CD4 + T cell counts and CD4+:CD8 + ratios were higher in individuals with helminths than in those without (median CD4 + T cell counts 467/pL and 268/pL, respectively, P = 0.005). Viral load was lower among those with helminths but this was not statistically significant. During follow-up, CD4 ÷ T cell counts and cytokine responses to PHA fell among individuals without helminths. Among those treated for helminths, CD4 ÷ counts remained stable. Viral loads showed a transient increase at 5 weeks, which was more marked among those treated for helminths, but the levels at 4 months were similar to baseline in both groups. Among those with schistosomiasis, IFN-y and IL-2 responses to CFP, and IL-2 and IL-4 responses to PHA declined but there was a sustained increase in cytokine responses to SWA following treatment. These data do not support the hypothesis that helminth infection exacerbates HIV infection. The possibility that chronic helminth infection may suppress HIV replication and that effects on HIV replication may vary during helminth infection and treatment should be considered.

Keywords: HIV, helminths, immunobiology, T cells, cytokine responses, Uganda Introduction It has been speculated that helminth infection may exacerbate HIV infection (Bentwich et al., 2000) and that this interaction may contribute to more rapid HIV progression in co-infected people. This hypothesis is based on immunological models but has not yet been substantiated by clinical data. Indeed, recent epidemiological studies suggest that the rate of HIV progression may be similar in rural Africa, where helminths are common, to the rate in the developed world (Morgan & Whitworth, 2001). The hypothesis proposes that helminth infections are associated with chronic 'immune activation' and hence lead to increased viral transcription and replication (Bentwich et al., 1995). More specifically, helminth infections tend to promote a type 2 bias in the immune response. This is characterized by production of cytokines such as interleukin (IL)-4 and IL-5, which promote production of antibody, including immunoglobulin (Ig)-E, and eosinophilia. This bias can affect the response both to helminth antigens and to unrelated antigens (Del Prete et al., 1991; Kullberg et al., 1992; Pearlman et al., 1993; Sabin et al., 1996). A strong type 2 bias might therefore promote HIV replication in several ways. First, type 1 responses, such as interferon (IFN)- 7 production and cytotoxic T cell activity, might be reduced and there is evidence that this might be associated with delayed clearance of viral infections (Actor et al., 1993). Second, there is evidence that HIV might replicate more rapidly in T helper (Th) 2 or Th0 lymphocytes than in T h l lymphocytes (Maggi et al., 1994). In addition, it has been suggested that replication of HIV in eosinophils might be significant in individuals with helminths and eosinophilia (Fincham et al., 1999). To investigate this hypothesis further, we studied helminth infections and

Address for correspondence: Dr A. M. Elliott, Uganda Virus Research Institute, P.O. Box 49, Entebbe, Uganda; phone +256 41 320272/320042, fax +256 41 321173, e-mail [email protected]

their association with HIV infection in a cohort of HIV1-seropositive adults in Entebbe, Uganda.

Methods The study of the interaction between helminths and HIV infection was based at the Uganda Virus Research Institute and at The AIDS Support Organisation (TASO) in Entebbe, Uganda. During the period November 1999 to January 2000, adults who were part of a continuing cohort of HIV-seropositive individuals (French et al., 2000) attending for enrolment into the cohort or for routine 6-monthly follow-up visits were asked to participate in this study. At enrolment to this study clinical information was collected, a blood sample was obtained and a stool sample was requested. Participants found to have helminth infections were visited by a fieldworker and treated. Schistosomiasis was treated with praziquantel, two doses of 20 mg/kg given 4 h apart. Intestinal nematodes were treated with mebendazole, 100 mg twice daily for 3 d. The first dose of treatment was observed. All participants were given follow-up appointments at 5 weeks and at 4 months. None of the participants was known to be taking antiviral drugs. Ethical approval was given by the Uganda Virus Research Institute and the Uganda National Council for Science and Technology. Blood samples were examined for CD4 + T cell count (FACScount, Becton Dickinson International, Brembodegem, Belgium). HIV serology was conducted using a rapid test (Capillus HIV-1/HIV-2, Cambridge Diagnostics Ireland Ltd, Galway, Ireland), confirmed by indirect enzyme immunoassay (Recombigen (env & gag) HIV-1 EIA; Cambridge Diagnostics Ireland Ltd) and a competitive enzyme-linked immunosorbent assay (ELISA) (Wellcozyme HIV recombinant, Murex Diagnostics Ltd, Dartford, UK). Sera showing discrepant results were examined by HIV-1 western blot (Cambridge Biotech HIV-1 Western Blot, Calypte Biomedical Corporation, Rockville, MA, USA). Plasma was stored at - 8 0 °C for HIV viral load assay (Cobas Amplicor HIV-1 monitor version 1.5, Roche Diagnostics, Branchburg, NJ, USA). Stool samples were examined by the Kato-Katz

104

method (Montresor et al., 1998). For schistosomiasis, serum was also examined for circulating anodic antigen (CAA) of Schistosoma mansoni (Deelder et al., 1989). For this assay, sera were pre-treated with trichloroacetic acid and samples were assayed in duplicate. The ELISA was conducted using anti-CAA monoclonal antibody McAb 147 for capture and biotinylated McAb 147 with alkaline phosphatase conjugate of streptavidin (DAKO, Glostrup, Denmark) and p-nitrophenyl phosphate (Sigma, Irvine, UIO for detection. A standard series of dilutions of trichloroacetic acid-soluble fraction of adult worm antigen from 1000 to 0.25 ng/mLwas included on each plate. The sensitivity of the assay was 4 ng/mL. Individuals with a positive stool sample for S. mansoni, a positive CAA result or both, were defined as infected with S. mansoni. Cytokine responses were examined using a whole blood assay as previously described (Elliott et al., 1999). Briefly, unseparated, heparinized blood was diluted to a final concentration of 1 in 4 using RPMI supplemented with penicillin, streptomycin and glutamine, plated in 96-well plates and stimulated with antigen or mitogen at a final concentration of 10 gg/mL or left unstimulated. Supernatants were harvested on day 1 for IL-2 and IL-4 and day 6 for IFN-y and IL-5 and frozen until analysed. Antigens were crude culture filtrate proteins (CFPs) of Mycobacterium tuberculosis and S. mansoni adult worm antigen (SWA), kindly provided by Dr J. T. Belisle, Colorado State University, USA and Dr D. W. Dunne, respectively. The mitogen was phytohaemagglutinin (PHA; Sigma). Cytokine concentrations were measured by ELISA (PharMingen, San Diego, CA, USA). The sensitivity of the assays was 8 p g / m L to 1 8 0 0 p g / m L Low-level production of cytokine in unstimulated wells was subtracted from the concentration produced in response to stimulation. Data were entered and analysed using FoxPro for Windows, version 2.6 (Microsoft Inc., Seattle, WA, USA) and Stata software, version 6 (Stata Corp., College Station, Texas, USA). Initial analyses of association were made using standard 2 × 2 tables. Crude and adjusted odds ratios (OR and AOR) were estimated using logistic regression. For qualitative analyses, cytokine responses were defined as positive for any result above the limit of sensitivity of the assay. Quantitative parameters showing skewed distributions (cytokine concentrations, cell counts, viral load) were compared using Wilcoxon's tests for paired or unpaired data, as appropriate. Results One hundred and twenty HIV-l-seropositive individuals were enrolled in the study. Twelve of these were excluded from the analysis, four who had no initial viral load result and eight who had no initial CD4 + T cell count. Of the remaining 108 participants, 39 (36%) had helminth infections identified by stool examination or examination of serum for Schistosoma CAA. Twenty-eight (26%) had schistosomiasis, 20 (18%) had intestinal nematodes (15 hookworm, 6 Trichuris, 2 Ascaris) and one had taeniasis. The median egg counts for those with positive findings were 50 eggs per gram (epg) of stool for S. mansoni, 160 epg for hookworm and 125 epg for Trichuris. The two participants with Ascaris had egg counts of 2110 and > 20 000 epg; the participant with Taenia had 10 epg. Some participants had more than one helminth infection, including 10 who had both schistosomiasis and intestinal nematodes. Individuals with helminths were slightly more likely to be male, to be younger and to be of somewhat poorer socio-economic status than those without, but had been in the cohort for a similar length of time prior to this study (Table 1).

A.M. ELLIOTT ETAL.

Table 1. A s s o c i a t i o n s b e t w e e n h e l m i n t h i n f e c t i o n and age, gender, o c c u p a t i o n and t i m e i n the cohort prior to this study i n H I V - l - s e r o p o s i t i v e adults, Entebbe, Uganda No

helminths Helminths (n = 69) (n = 39) Gender Female 53 (77%) 27 (69%) Male 16 (23%) 12 (31%) Age (years) b 35.0 (7.9) 32.6 (7.9) Occupation Regular income 13 (19%) 2 (5%) Irregular or low 15 (22%) 10 (26%) income No regular income 41 (59%) 27 (69%) Time in the cohort 2.3 (1.2) 2.3 (1.4) prior to this study (years) b

pa 0.39 0.14 0.14

0.98

~Z2 test for proportions; t test for means. bMean (SD). CD4 + T cell count, viral load and cytokine responses at enrolment At enrolment, individuals with helminth infections had significantly higher CD4 + T cell counts than individuals without helminths. There were no statistically significant differences in CD8 + T cell counts and thus CD4+:CD8 + T cell ratios were markedly higher in those with helminths (Table 2). Adjusting for age, gender, socio-economic status and time in the cohort prior to this study did not explain these associations (Table 3). Viral loads were lower among individuals with helminths, although this difference was not statistically significant. The proportion of individuals with or without helminths who showed a response to PHA was 80% or more for IFN-y, IL-2 and IL-5. Seventy-two percent of those with helminths and 67% of those without helminths produced IL-4. The concentration of cytokines produced in response to PHA was generally higher for individuals with helminth infection but these differences were not statistically significant. The proportion of individuals with a positive response to CFP was also generally higher in the helminth-infected group. A significantly higher proportion of individuals with helminths produced IL-5 in response to CFP (Table 2) but this association was reduced after adjusting for CD4 + T cell count (OR = 2.4, P = 0.04; AOR for in CD4 + T cell count = 1.6, P = 0.29). Effect of treatment for helminths on CD4 + T cell count, viral load and cytokine responses Individuals who attended and had complete results for all time points were included in a longitudinal analysis. Of the 69 participants who had no helminths at enrolment, 12 were excluded from the follow-up analysis because they were found to have helminths and treated at later time points. In addition, 13 (of whom two died) were excluded because they failed to attend one or more visits and 10 because stool or blood results were incomplete. Of the 39 participants with helminths at enrolment, five were excluded from the follow-up analysis, four (of whom one died) because they failed to attend and one because results were incomplete. Thus 34 participants with no helminths at any time and 34 with helminths at enrolment (24 with schistosomiasis at enrolment, 19 with nematodes and one with Taenia) were analysed for follow-up. Of the 34 participants with helminths who were analysed, 13 still had helminth infections at 5 weeks and 13 at 4 months (eight had helminths at both follow-up visits). Analysis of all 34 individuals without helminths

I N T E R A C T I O N B E T W E E N H E L M I N T H S A N D HIV I N F E C T I O N

105

Table 2. Associations between helminth infection and C D 4 + T c e l l counts, viral load and cytokine responses at study enrolment in H I V - l - s e r o p o s i t i v e adults, Entebbe, Uganda, N o v e m b e r 1999-January 2000 No

helminths (n = 69)

Helminths (n = 39)

Number 69 C D 4 + T cell count b 268 ( 1 0 0 - 4 6 2 ) C D 4 + : C D 8 + T cellratio b 0.223 (0.121-0.498) LOgl0 viral load u 5.00 (4.40-5.33) Response to PHA: median concentration (pg/mL) IFN-y 1641 IL-2 68 IL-4 21 IL-5 832 Response to CFP: % with positive response IFN-y 32% IL-2 30% IL-4 3% IL-5 26%

39 467 ( 2 3 7 - 6 4 3 ) 0.497 ( 0 . 2 4 6 - 0 . 8 7 4 ) c 4.79 (4.31-5.26)

P~ 0.005 0.001 0.33

> 1800 87 20 1059

0.51 0.86 0.36 0.22

49% 38% 3% 46%

0.08 0.40 0.92 0.03

PHA, phytohaemagglutinin; CFP, culture filtrate proteins of Mycobacterium tuberculosis; IFN-y, interferon-y; IL, interleukin. aWilcoxon rank-sum test for medians; X2 test for proportions. bMedian (interquartile range). cOne missing value (CD8 + T cell count not available).

Table 3. Association between helminth infection and CD4 ÷ T cell count, with adjustment for potential explanatory or confounding factors, in H I V - l - s e r o p o s i t i v e adults, Entebbe, Uganda, N o v e m b e r 1999J a n u a r y 2000 Odds ratio for helminth infection (P value, test for trend)

CD4" T cell count 0-199 200-499 >500

Number of subjects

Number (%) with helminth infection

34 44 30

8 (24%) 16 (36%) 15 (50%)

Unadjusted 1 (0.03) 1.86 3.25

showed a progressive decline in C D 4 ÷ T cell counts during the follow-up period (Figure, A). A decline was also seen when the analysis was restricted to the 13 subjects without helminths, who had the highest C D 4 + T cell counts, to give a median C D 4 + T cell count at enrolment which was similar to the helminth-infected group (median C D 4 + T cell count at enrolment, 468/ btL; at 5 weeks, 387/1aL [P = 0.01 for comparison with enrolment]; at 4 months, 3 9 6 / ~ L [ P = 0.03 for comparison with enrolment]). By contrast, a m o n g individuals who had helminths at enrolment and were treated, C D 4 ÷ T cell counts showed no significant change. T h e r e was no statistically significant change in viral load between enrolment and at 4 months for individuals either with or without helminths but individuals in both groups showed an increase in viral load at 5 weeks (Figure, B). This was statistically significant for those without helminths ( P = 0.03) but was more marked for those treated for helminths ( P = 0.008). For those with schistosomiasis this was followed by a statistically significant decline in viral load between 5 weeks and 4 months (P = 0.01). A m o n g individuals with helminths at enrolment, the changes in C D 4 + T cell count and viral load were similar at 4 months, whether or not helminth infection was still present (median change in log10 viral load, +0.11 [helminths not present at 4 months, n = 21], +0.10 [helminths present, n = 13], P - 0 . 9 9 ; median change in C D 4 + T cell counts, +1 cell/gL [helminths not present at 4 months], +2 cells/~L [helminths present], P = 0.80).

Adjusted for age and gender

Adjusted for occupational status

Adjusted for time in the cohort prior to this study

1 (0.05) 1.89 3.00

1 (0.03) 1.62 3.38

1 (0.03) 1.85 3.31

Cytokine responses during the follow-up period are shown in Table 4. Results for participants without helminths are compared to participants with schistosomiasis, for w h o m helminth-specific responses could be assessed using SWA. Changes in responses to C F P are shown only for individuals who had a response with at least one of the cytokines at baseline, since other participants may have been either anergic or not sensitized to M. tuberculosis. Interleukin-4 responses are shown only for P H A because less than 50% showed a response to C F P or S W A in all groups. A m o n g individuals without helminths there was a progressive, statistically significant decline in all four cytoldne responses to P H A . Responses to C F P were low in this group and remained unchanged. A m o n g individuals with schistosomiasis, the median I F N - y response to P H A remained close to or above the m a x i m u m sensitivity of the assay, so quantitative changes may not have been detected. T h e IL-2 response and the IL-4 response declined while the IL-5 response was sustained. T h e I F N - y and IL-2 responses to C F P also declined markedly following treatment a m o n g those who had had an initial response. Interleukin-5 production in response to C F P appeared to show a transient increase at 5 weeks, although this effect was not statistically significant. By contrast, responses to S W A increased and, for IL-2 and IL-5, these increases were highly statistically significant and were sustained at 4 months.

Discussion These data raise doubts regarding the hypothesis that helminth infection exacerbates H I V infection and con-

106

A.M. ELLIOTT ETAL.

A

B

6-

600-

i,-,~

5-

500,,16

ll'"-

400 -

N

-~ 3ooo

'r.

4-

3-

o

#. •.

hi0

-A--A. ''%o

200-

............................... • ,t--g

"i, .....

.....

.....

2-

.. " ' - - . . . . . . . i

1

100-

0

r

i

i

i

1 2 3 Months from start of treatment

4

"

i

i---$--- no helminths [ l helminths

0

0

i

i

i

i

1 2 3 Months from start of treatment

4

i

Figure. Changes in (A) median CD4 + T cell counts and (B) logl0 viral load following treatment of helminth infections in HIV-1seropositive adults, Entebbe, Uganda, November 1999-January 2000. Asterisks indicate P values for comparison of median values at 5 weeks or 4 months with values at enrolment: *P ~< 0.05; ~ P ~< 0.01.

Table 4. Changes in cytokine concentration during follow-up in HIV-1seropositive adults without helxninths or with schistosomiasis, Entebbe, Uganda, November 1999-January 2000 M e d i a n cytokine c o n c e n t r a t i o n ( p g / m L ) Five weeks after enrolment Enrolment No helminths P H A b (n = 34) IFN-y IL-2 IL-4 IL-5 C F P c (n = 14) IFN-y IL-2 IL-5 Schistosomiasis P H A b (n = 24) IFN-y IL-2 IL-4 IL-5 C F P c (n = 16) IFN- 7 IL-2 IL-5 S W A a (n = 24) IFN- 7 IL-2 IL-5

pa

1329 60 11 302 17 0.5 0

>1800 115 60 1132

F o u r m o n t h s after enrolment p~

1324 31.5 1 230

0.30 0.007 0.002 0.47

812 34 5.5 226

0.02 0.02 0.01 0.007

17 0 0

0.57 0.90 0.82

8 0 0

0.70 0.64 0.69

>1800 65.5 28 1142

74 14 20

64 6 60

3 0 53

22 2 715

0.30 0.003 0.30 0.95

1732 69.5 40 1215

0.48 0.001 0.03 0.25

0.72 0.94 0.50

21 0 12

0.04 0.02 0.64

0.10 0.01 0.003

6 21 502

0.50 0.003 0.003

IFN-y, interferon-y; IL, interleukin. aComparison of median cytokine concentrations at enrolment, at 5 weeks follow-up and at 4 months follow-up (Wilcoxon signed-rank test). bResponses to phytohaemagglutinin. CResponses to culture filtrate proteins of Mycobacterium tuberculosis for individuals who had a positive response with one or more of the three cytokine responses at enrolment. dResponses to Schistosoma mansoni adult worm antigen. flict with preliminary data r e p o r t e d f r o m Ethiopia (Borkow & Bentwich, 2000). If the hypothesis were correct, individuals with h e l m i n t h s m i g h t be e x p e c t e d to have higher viral loads, and p e r h a p s lower C D 4 + T cell

counts, t h a n those without. I n d e e d , low C D 4 + T cell counts and C D 4 + : C D 8 + T cell ratios have b e e n s h o w n to be associated with h e l m i n t h infection in HIV-negative subjects (Kalinkovich et al., 1998). H o w -

INTERACTIONBETWEENHELMINTHSAND HIV INFECTION ever, we found at enrolment that individuals with helminth infection had slightly lower viral loads and significantly higher CD4 + T cell counts and CD4+:CD8 + T cell ratios. Given the cross-sectional nature of the analysis, these results need to be interpreted with caution. They could mean that helminth infection was actually associated with protection against HIV progression but a number of other interpretations are possible. For example, individuals with helminths might have presented earlier in the course of HIV disease than those without or might have died or left the cohort with higher C D 4 + T cell counts; or individuals with more advanced, symptomatic HIV infection might be more likely to have taken anthelmintics recently. T h e somewhat younger age of individuals with helminths suggests some such explanation, but adjustment for age, or for time under followup in the cohort prior to this study (and hence, perhaps, likelihood of anthelmintic treatment), did not account for the observed effect. Alternatively, helminth infection might be less readily detected in individuals with advanced immunosuppression. Immune deficiency has been shown to be associated with reduced egg excretion in schistosomiasis, probably because an inflammatory response is required for eggs to pass from the portal circulation though the intestinal mucosa (Karanja et al., 1997). However, the use of a serum antigen detection assay in addition to stool analysis may mean that this would be less likely to explain the findings in this study. Another possibility is that effects of helminths were underestimated in this study because of misclassification. Only one stool sample was obtained at each visit, so it is likely that predominantly high-intensity infections were identified, and the K a t o Katz method gives a poor yield for Stronglyloides. Some participants may have had helminth infections and been treated for them in the recent past, and these infections may have had some persistent immunological effects. Filarial infections were not sought. Most of these potential misclassifications would be expected to reduce the differences observed between the groups, but not to alter the direction of the effect. These possibilities could be addressed in more detailed studies but the cross-sectional component of this study, at least, gave no evidence that helminth infection was detrimental. During the 4 month follow-up period, C D 4 + T cell counts declined and cytoldne responses to P H A decreased in individuals without helminths, as expected in progressive HIV infection. On the other hand, individuals treated for helminths showed sustained CD4 + T cell counts, whether or not the helminth infection was still present. Between enrolment and 4 months there was little change in viral load in either group, but there was a transient increase in viral load at 5 weeks. This was most marked among individuals treated for schistosomiasis, in whom the viral load had fallen again at 4 months. Some of the effects seen in the helminth groups may be related to immunological changes following treatment, indicated by the changes in cytokine responses observed in participants with schistosomiasis. There was a marked increase in IL-2 and IL-5 production in response to Schistosorna worm antigen following treatment. Th e increase in IL-5 response to C F P at 5 weeks (although not statistically significant) suggested an increased tendency to type 2 responses to unrelated antigens. The associated, transient increase in viral load in schistosomiasis patients at 5 weeks is in keeping with the hypothesis that a burst of immune activation following treatment, possibly with a type 2 bias, was associated with increased viral replication. While the cyto!dne responses at baseline suggested that individuals with helminths had relatively intact immune function, the changes during follow-up suggested progressive loss of selected aspects of immune

107 function in participants both with and without helminths, despite sustained CD4 + T cell counts in the helminth-infected participants. There was a marked decline in IFN-~/and IL-2 responses to CFP and in IL2 and IL-4 responses to P H A in the schistosomiasis group. Only IL-5 responses (which may, to some extent, be contributed by eosinophils) were clearly sustained. Th e contrasting increase in responses to SWA suggests that the sustained CD4 + T cell counts in these subjects may be due in part to a relative expansion of helminth-specific cells. In the light of these findings, the effects of chronic helminth infection and of the treatment of helminths on HIV replication need to be examined further, so that recommendations regarding treatment can be made. In this study we felt that it was incumbent upon us to treat individuals identified as having helminths, so we were unable to distinguish the effects of treated and untreated helminth infections on HIV disease progression. Furthermore, the number of participants in this study was too small to analyse differences between those whose helminth infections cleared and those with persistent, or new, infections following treatment. Future studies should be designed to examine these differences, if possible, given our evidence that helminth infection may not be as detrimental as previously thought. Th e pathways by which 'immune activation' may stimulate viral replication and the mechanisms by which this effect may be blocked are not certain (McComsey et al., 2001). In the development of the hypothesis that helminth infection may exacerbate HIV infection, the concept of a detrimental effect of 'activation' of the immune system by helminths, and of a type 2 bias, may have been oversimplified. Helminth infections also induce production of anti-inflammatory cytokines, notably IL-10 and tissue growth factor (TGF)-I3, which suppress antigen presentation and both T h l and Th2 lymphocyte activity (Yazdanbakhsh et al., 2001). Interleukin-10 inhibits inflammatory responses mediated by the transcription factor NF-~:B (Schottelius e~ al., 1999). Hence, IL-10 might also inhibit activation of viral replication by NF-~cB, which triggers replication by binding the long terminal repeat sequence of proviral HIV (Duh et al., 1989; Osborn et al., 1989) and chronic helminth infection might suppress HIV replication. On the other hand our data, especially for patients with schistosomiasis, suggest that the boost in helminth specific responses following treatment may influence responses to other antigens and trigger an increase in HIV replication. Thus further studies which distinguish between the effects of treated and untreated helminth infections are needed, and measurement of viral load may give a better indication of the effects of helminths than CD 4 ÷ T cell counts since the latter may be influenced by changes in helminth-specific responses.

Acknowledgements We thank the clients of TASO and UVRI clinics for participating in the study. We thank the medical and nursing staff of TASO, and the medical staff, field workers, and laboratory and data entry staff of the MRC Programme on AIDS in Uganda. We thank the Vector Control Division, Ministry of Health, Uganda, for training and provision of equipment for stool analyses. We thank Wendy Hodsdon for assistance with laboratory work, Maria Quigley for statistical advice and Michael Brown for comments on the manuscript. The study was funded by a Wellcome Trust Career Development Fellowship to Dr Alison Elliott (grant reference 049523). Additional funding for viral load assays and CD4 + T cell counts was arranged by Dr N. J. Robinson (Glaxo Smith Kline, Greenford, UK). References Actor, J., Shirai, M., Kullberg, M., Bullet, R., Sher, A. & Berzofsky, J. (1993). Helminth infection results in decreased virus-specific CD8+ cytotoxic T-cell and Thl cytokine

A.M. ELLIOTT ETAL.

108 responses as well as delayed virus clearance. Proceedings of the NationalAcademy of Sciences of the USA, 90, 948-952. Bentwich, Z., Kalinkovich, A. & Weisman, Z. (1995). Immune activation is a dominant factor in the pathogenesis of African AIDS. Immunology Today, 16~ 187-191. Bentwich, Z., Maartens, G., Torten, D., Lal. A. A. & Lal, R. B. (2000). Concurrent infections and HIV pathogenesis. AIDS, 14, 2071-2081. Borkow, G. & Bentwich, Z. (2000). Eradication of helminth infections may be essential for successful vaccination against HIV and tuberculosis. Bulletin of the WorldHealth Organization, 78, 1368-1369. Deelder, A. M., De Jonge, N., Boerman, O., Fillie, Y. E., Hilberath, G. W., Rotmans, J. P., Gerritse, M. J. & Schut, D. W. O. L. (1989). Sensitive determination of circulating anodic antigen in Schistosoma mansoni infected individuals by an enzyme-linked immunosorbent assay using monoclonal antibodies. American Journal of Tropical Medicine and Hygiene, 40, 268-272. Del Prete, G. F., De Carli, M., Mastomauro, C., Biagotti, R., Macchia, D., Falagiani, P., Ricci, M. & Romagnani, S. (1991). Purified protein derivative of Mycobacterium tuberculosis and excretory antigen(s) of Toxocara canis expand in vitro human T cells with stable and opposite (type 1 T helper or type 2 T helper) profile of cytokine production. Journal of ClinicalInvestigation, 88, 346-350. Dub, E. J., Maury, W. J., Folks, T. M., Fauci, A. S. & Rabson, A. B. (1989). Tumor necrosis factor ct activates human immunodeficiency virus type 1 through induction of nuclear factor binding to the NF-KB sites in the long terminal repeat. Proceedings of the National Academy of Sciences of the USA, 86, 5974-5978. Elliott, A. M., Hurst, T. J., Baleyku, M., Quigley, M., Kaleebu, P., French, N., Biryawaho, B., Whitworth, J. A. G., Dockrell, H. M. & Hayes, R. J. (1999). The immune response to Mycobacterium tuberculosis in HIV infected and uninfected adults: application of the whole blood cytokine assay in an epidemiological study in Uganda. International Journal of Tuberculosis and Lung Disease, 3, 239-247. Fincham, J. E., Markus, M. B. & Mansvelt, E. P. G. (1999). Could non-selective anthelminthic treatment programmes contribute to control of the spread of HIV infection and AIDS? Transactions of the Royal Society of Tropical Medicine and Hygiene, 93, 536. French, N., Naldyingi, J., Carpenter, L. M., Lugada, E., Watera, C., Moi, K., Moore, M., Antvelink, D., Mulder, D., Janoff, E. N., Whitworth, J. & Gilks, C. F. (2000). 23valent pneumococcal polysaccharide vaccine in HIV-l-infected Ugandan adults: double-blind, randomised and placebo controlled trial. Lancet, 355, 2106-2111. Kalinkovich, A., Weisman, Z., Greenberg, Z., Nahmias, J., Eitan, S., Stein, M. & Bentwich, Z. (1998). Decreased CD4 and increased CD8 counts with T cell activation is asso-

ciated with chronic helminth infection. ClinicalExperimental

Immunology, 114, 414-421. Karanja, D. M. S., Colley, D. G., Nahlen, B. L., Ouma, J. H. & Secor, W. E. (1997). Studies on schistosomiasis in Western Kenya: I. Evidence for immune-facilitated excretion of schistosome eggs from patients with Schistosoma mansoni and human immunodeficiency virus coinfections. American Journal of TropicalMedicine and Hygiene, 56, 515- 521. Kullberg, M. C., Pearce, E. J., Hieny, S. E., Sher, A. & Berzofsky, J. A. (1992). Infection with Schistosoma mansoni alters T h l / T h 2 cytokine responses to a non parasite antigen. Journal of Immunology, 148, 3264-3270. Maggi, E., Mazzetti, M., Ravina, A., Annunziato, F., De Carli, M., Piccinni, M. P., Manetti, R., Carbonari, M., Pesce, A. M., Del Prete, G. & Romagnani, S. (1994). Ability of HIV to promote a T h l to Th0 shift and to replicate preferentially in Th2 and Th0 cells. Science, 265, 244-248. McComsey, G. A., Whalen, C. C., Mawhorter, S. D., Asaad, R., Valdez, H., Patki, A. H., Klaumunzner, J., Gopalakrishna, K. V., Calabrese, L. H. & Lederman, M. M. (2001). Placebo-controlled trial of prednisone in advance HIV-1 infection, AIDS, 15, 321-327. Montresor, A., Crompton, D. W. T., Hall, A., Bundy, D. A. P. & Savioli, L. (1998). Guidelinesfor the evaluation of soil-

transmitted helminthiasisand schistosomiasis at community level. Geneva: World Health Organization, WHO/CTD/SIP/98.1. Morgan, D. & Whitworth, J. A. G. (2001). The natural history of HIV- 1 infection in Africa. Nature Medicine, 7, 143-145. Osborn, L., Kunkel, S. & Nabel, G. J. (1989). Tumor necrosis factor c~ and interleukin 1 stimulate the human immunodeficiency virus enhancer by activation of the nuclear factor

~B. Proceedings of the National Academy of Sciences of the USA, 86, 2336-2340. Pearlman, E., Kazura, J. W., Hazlett, F. E. & Boom, W. H. (1993). Modulation of murine cytokine responses to mycobacterial antigens by helminth-induced T helper 2 cell responses. Journal of Immunology, 151, 4857-4864. Sabin, E. A., Ilma Araujo, M., Carvalho, E. M. & Pearce, E. J. (1996). Impairment of tetanus toxoid-specific Thl-like immune responses in humans infected with Schiswsoma mansohi. Journal of Infectious Diseases, 173, 269-272. Schottelius, A. J. G., Mayo, M. W., Sartor, R. B. & Baldwin, A. S. (1999). Interleukin- 10 signaling blocks inhibitor of KB kinase activity and nuclear KB DNA binding. Journal of Biological Chemistry, 274, 31868-31874. Yazdanbakhsh, M., van den Biggelaar, A. & Maizels, R. (2001). Th2 responses without atopy: immunoregulation in chronic helminth infections and reduced allergic disease. Trends in Immunology, 22, 372-377.

Received 28 March 2002; revised 27 June 2002; accepted for publication 27June 2002

Announcement R O Y A L S O C I E T Y OF T R O P I C A L M E D I C I N E A N D H Y G I E N E D e n i s Burkitt F e l l o w s h i p s T h e D e n i s Burkitt F u n d was set u p b y his family in m e m o r y of D e n i s Burkitt, FRS, w h o died in 1993; it is a d m i n i s t e r e d by the Royal Society of T r o p i c a l M e d i c i n e a n d Hygiene. O n e Fellowship ( m a x i m u m value £ 7 0 0 0 ) or two separate Fellowships (of £ 3 5 0 0 each) are a w a r d e d a n n u a l l y for practical training, travel, or direct assistance with a specific project (preferably clinico-pathological, geographical or epidemiological studies of n o n - c o m m u n i c a b l e diseases in Africa). Applications m u s t be m a d e at least six m o n t h s before the c o m m e n c e m e n t of the p r o p o s e d s t u d y (by 15 M a r c h in each year). A short r e p o r t o n the s t u d y s h o u l d b e s u b m i t t e d , w i t h i n 3 m o n t h s of the recipient's return. A p p l i c a t i o n forms are available from the A d m i n i s t r a t o r , Royal Society o f T r o p i c a l M e d i c i n e a n d Hygiene, M a n s o n H o u s e , 26 P o r t l a n d Place, L o n d o n , W 1 B 1EY, U K ; fax + 4 4 (0)20 7436 1389, e-mail m a i l @ r s t m h . o r g