The Clinical Features and Immunological Signature of Cyclospora cayetanensis Co-Infection among People Living with HIV in Ghana

Background: There is a paucity of information on the contemporary burden, disease patterns, and immunological profile of people living with HIV who are co-infected with C. cayetanensis in the post-antiretroviral therapy era. Methods: For this cross-sectional study, stool samples of 640 HIV-positive and 83 HIV-negative individuals in Ghana were tested for C. cayetanensis. Additionally, sociodemographic parameters, clinical symptoms, medical drug intake, and immunological parameters were assessed. Results: The prevalence of C. cayetanensis was 8.75% (n = 56) in HIV-positive and 1.20% (n = 1) in HIV-negative participants (p = 0.015). Within the group of HIV-positive participants, the prevalence reached 13.6% in patients with CD4+ T cell counts below 200 cells/µl. Frequencies of the clinical manifestations of weight loss and diarrheal disease were significantly higher in patients with C. cayetanensis compared to those without co-infection (36.36% vs. 22.59%, p = 0.034 and 20.00% vs. 4.90%, p < 0.001, respectively). The expression of markers of immune activation and exhaustion of T lymphocyte sub-populations was significantly elevated in patients colonized with C. cayetanensis. Conclusions: In the modern post-combined antiretroviral therapy (cART) era, the acquisition of C. cayetanensis among PLWH in Ghana is driven largely by the immunosuppression profile characterized by high expression of markers of immune activation and immune exhaustion.


Introduction
Cyclospora cayetanensis, a coccidian parasite, causes enteric disease among humans via fecal-oral transmission [1,2]. The life cycle of this parasite involves an obligatory sporulation step at a temperature of 25-30 • C for at least 1-2 weeks in the environment to become infective [3]. Therefore, direct person-to-person transmission is rather unlikely [2]. Infection with C. cayetanensis is mainly transmitted through the ingestion of contaminated food or water with oocysts and heightened transmission during the rainy season [4][5][6]. C. cayetanensis establishes as an infection in the upper small intestinal tract, which is evanescent in immunocompetent hosts, but among immunocompromised individuals, complications that may ensue include chronic diarrhea and malabsorptive syndromes due to villous atrophy and crypt hyperplasia [7]. These characteristics of C. cayetanensis make it an important pathogen among people living with HIV (PLWH) as a differential diagnosis of chronic diarrheal disease in this population.
Infection with Cyclospora cayetanensis is treated with co-trimoxazole, a commonly prescribed antibiotic used prophylactically and therapeutically among PLWH in resourcelimited settings [8,9]. Ciprofloxacin is less effective than co-trimoxazole but is suitable for patients who are intolerant to sulfonamide drugs [10].
There is, however, a paucity of information on the contemporary burden, disease patterns, and immunological profile of PLWH who are co-infected with C. cayetanensis in the post-antiretroviral therapy era. Recent reviews confirmed diarrhea and a low CD4+ T cell count to be significantly associated with C. cayetanensis infection in PLWH [3,11]. We, therefore, sought to conduct a study to evaluate the clinical and immunological profile of a large cohort of PLWH according to infection status with C. cayetanensis in a Ghanaian health system. Furthermore, we aimed to provide new information on the association of this parasite with markers of immune activation, cell proliferation, terminal differentiation, and exhaustion of CD4+ and CD8+ T lymphocyte sub-populations stratified for the combined antiretroviral therapy (cART) status.

Study Design and Population
This cross-sectional observational study is part of a work investigating the prevalence of Helicobacter pylori and other gastrointestinal co-infections in HIV-positive and -negative adults and was conducted at the Komfo Anokye Teaching Hospital, a tertiary referral hospital in the Ashanti Region of Ghana [12][13][14][15]. Between November 2011 and November 2012, consecutive HIV-positive patients presenting to the HIV outpatient department, and HIV-negative blood donors presenting to the blood bank of the hospital, were offered participation in the study. All participants gave written informed consent prior to enrolment. The study was approved by the appropriate ethics committees in Ghana (CHRPE/AP/12/11) and Germany (PV3771).

Data Collection and Laboratory Methods
Demographic, socioeconomic, and clinical data, as well as a detailed medical history, were recorded using standardized questionnaires, which were completed by trained study personnel. In particular, time since diagnosis of HIV infection, duration and kind of cART, co-medications, and clinical parameters were documented. Blood samples were collected, and the analysis of CD4+ T cell count was performed locally using a FACSCalibur flow cytometer (Becton Dickinson, Mountain View, CA, USA). HIV-1 viral load was measured using the Real-Time HIV-1 PCR system (Abbott Diagnostics, Wiesbaden, Germany).
Peripheral blood mononuclear cells (PBMCs) were isolated by centrifugation of heparinized venous blood on a Ficoll/Hypaque (Biocoll Separating Solution, Biochrom AG, Berlin, Germany) density gradient. Cells were washed in phosphate-buffered saline and resuspended in Roswell Park Memorial Institute 1640 medium (both Gibco Invitrogen, Carlsbad, CA, USA) supplemented with heat-inactivated fetal calf serum (Biochrom AG, Berlin, Germany). PBMCs were cryopreserved and shipped to Germany on liquid nitrogen. Cell surface markers for immune activation were stained as described elsewhere [16]. Flow cytometric data were acquired using the LSRII flow cytometer (BD Biosciences, Heidelberg, Germany) and analyzed using FlowJo version 9.6.2 (Tree Star, San Carlos, CA, USA).
Aliquots of native stool samples were freshly frozen and stored at −80 • C before being transported to Germany on dry ice. All stool samples were subjected to nucleic acid extraction applying the QiaAMP DNA Stool Mini kit (Qiagen, Hilden, Germany) as suggested by the manufacturer. The real-time PCR targeting the small subunit ribosomal RNA (SSU rRNA) gene of C. cayetanensis was run as described previously on a Rotor-Gene Q cycler (Qiagen, Hilden, Germany) with some minor modifications [17]. In short, the forward primer sequence was 5 -TAGTAACCGAACGGATCGCATT-3 , the reverse primer sequence was 5 -AATGCCACGGTAGGCCAATA-3 , and the probe sequence was 5 -CCGGCGATAGATCATTCAAGTTTCTGACC-3 . The reaction mix comprised a 10 µL HotStar master mix (Qiagen, Hilden, Germany), 5.0 mM total MgCl 2 , and 2.0 µL DNA eluate in 20 µL volumes. The run conditions were activation at 95 • C for 15 min, 45 cycles of 15 s denaturation at 95 • C, followed by a touchdown for the 30-second-long combined annealing and elongation step from 72 • C to 67 • C over 13 cycles in 0.5 • C steps, with final cooling down to 40 • C for an additional 30 s at the end of the run. As estimated in previous studies, calculated sensitivity for the C. cayetanensis-specific real-time PCR ranges from 32.0% to 81.8%, specificity from 98.7% to 99.7% with a limit of detection of less than 10 copies per µL eluate [18,19]. All runs included a PCR-grade-water-based negative control and positive control with a plasmid containing the target sequence 5 -GATTCATAGTAACCGAACGGATCGCATTTGGCTTTAGCCGGCGATAGATCATTCAAG-TTTCTGACCTATCAGCTTTCGACGGTAGGGTATTGGCCTACCGTGGCATTGACGGG-3 . An inhibition control PCR targeting a phocid herpes virus (PhHV) sequence fragment as described elsewhere was run with all samples to exclude relevant sample inhibition [20].

Statistical Analysis
Categorical variables were compared using either the χ 2 test or the Fisher exact test, as appropriate. Continuous variables were expressed as median (interquartile range, IQR) or mean ± standard deviation (SD) and compared using the Wilcoxon rank-sum test or the unpaired Student's t-test. Multiple regression models were accomplished using the "forestmodel" package in R (version 4.0.5, R Foundation for Statistical Computing, Vienna, Austria). To account for moderation, models were additionally run with included interaction terms. The Spearman rank correlation coefficient ρ was calculated as a measure of the strength of the relationship between continuous variables. Two-sided p-values were presented, and an α of 0.05 was determined as the cutoff for significance.

Composition of the Study Population
A total of 1095 HIV-positive individuals and 107 HIV-negative blood donors were recruited for this study. Residual stool samples for C. cayetanensis testing were available for 723 individuals (640 HIV-positive and 83 HIV-negative). The prevalence of C. cayetanensis was 8.75% (n = 56) in HIV-positive participants and 1.20% (n = 1, p = 0.015) in HIV-negative participants. Among HIV-positive patients, the detection rate was higher in patients with detectable HIV viral load, with CD4+ T cell counts below 200/µL, and in cART naïve patients ( Figure 1).

Comparison of Demographic and Clinical Characteristics of the HIV Cohort by C. cayetanesis Status
As shown in Table 1, there were no differences in the mean age, sex, and socioeconomic parameters of those infected with C. cayetanensis compared to those who were uninfected. There were also no differences in the proportion of cART and co-trimoxazole chemoprophylaxis between the two groups, but those with C. cayetanensis infection had been on cART for a significantly shorter median (IQR) duration of 17.3 (6.9-36.6) months compared to 55.2 (28.6-79.5) months for those not co-infected (p = 0.001).

Comparison of Demographic and Clinical Characteristics of the HIV Cohort by C. cayetanesis Status
As shown in Table 1, there were no differences in the mean age, sex, and socioeconomic parameters of those infected with C. cayetanensis compared to those who were uninfected. There were also no differences in the proportion of cART and co-trimoxazole chemoprophylaxis between the two groups, but those with C. cayetanensis infection had been on cART for a significantly shorter median (IQR) duration of 17.3 (6.9-36.6) months compared to 55.2 (28.6-79.5) months for those not co-infected (p = 0.001). Clinically, those with C. cayetanensis co-infection compared to those without co-infection were more likely to report weight loss over the previous 6 months (36.36% vs. 22.59%, p = 0.034) and acute or chronic diarrhea (20.00% vs. 4.90%, p < 0.001), as shown in Figure  2. There were no significant differences in other symptoms such as fever or chronic cough.  Clinically, those with C. cayetanensis co-infection compared to those without co-infection were more likely to report weight loss over the previous 6 months (36.36% vs. 22.59%, p = 0.034) and acute or chronic diarrhea (20.00% vs. 4.90%, p < 0.001), as shown in Figure 2. There were no significant differences in other symptoms such as fever or chronic cough.

Comparison of Virological and Immunological Characteristics of the HIV Cohort by C. cayetanesis Status
Participants with C. cayetanensis co-infection had a significantly higher median HIV-1 viral load in log10 copies/mL ( [21.4-44.0], p = 0.042). There were no significant differences in the expression of surface markers for terminal differentiation or cell proliferation in C. cayetanensis positive and negative patients.

Comparison of Virological and Immunological Characteristics of the HIV Cohort by C. cayetanesis Status
Participants with C. cayetanensis co-infection had a significantly higher median HIV-1 viral load in log10 copies/mL (

Immunological Significance of C. cayetanensis Infection in PLWH
In a multiple logistic regression model, we identified a CD4+ T lymphocyte value < 200 cells/µL to be significantly associated with the odds of C. cayetanensis coinfection among PLWH, the adjusted odds ratio was 2.11 (95% CI: 1.17, 3.79), p = 0.01 (Supplementary Figure S1a). However, age, sex, and co-trimoxazole use were not associated with C. cayetanensis infection. Furthermore, C. cayetanensis infection among PLWH was significantly and independently associated with markers of increased immune activation with a parameter estimate from a multiple linear regression model of 7.46 (95% CI: 1.65, 13.27, p = 0.01). As expected, a low CD4+ T lymphocyte count of <200/µL also was associated with elevated immune activation, as evidenced by increased expression of HLA-DR+ CD38+ CD4+ (Supplementary Figure S1b).

Correlations of Cycle Threshold (Ct) Values with HIV Viral Load and CD4+ Cell Count
The correlation analysis of cycle threshold (Ct) values and HIV viral load as well as CD4+ T cell count in C. cayetanensis positive participants revealed a significant inverse correlation for the HIV viral load and the Ct values of C. cayetanensis but not for the CD4 lymphocyte count (rho = −0.36, p = 0.008 and rho = 0.10, p = 0.469, respectively).

Discussion
In this large cohort of Ghanaians living with HIV in the post-cART era, the prevalence of Cyclospora cayetanensis was 8.75% compared to 1.20% among HIV-negative control group. Compared with data from other regions of the globe, the prevalence of C. cayetanensis among the PLWH in Ghana is higher than 0.3% in Iran, 1% in Tanzania, 1.1% in Thailand, 0.7-3.3% in India, 3.0% in Cuba, 3.4% in Nigeria, and 3.9% in Venezuela but lower than 11% in a Haitian cohort [21][22][23][24][25][26][27]. The differences in prevalence observed in the various studies may reflect differences in relative proportions of cART exposure in the HIV cohort, diagnostic techniques utilized (microscopy or molecular approaches such as PCR), prophylactic use of co-trimoxazole and perhaps also the level of sanitation in the various countries where these studies were conducted. In the present study, the presence of C. cayetanensis was assessed using an in-house real-time PCR targeting the small subunit ribosomal RNA (SSU rRNA) gene of C. cayetanensis [19]. In a previous head-to-head-comparison of described C. cayetanensis-specific real-time PCR assays, the applied approach was confirmed to be particularly sensitive for the identification of C. cayetanensis in human stool samples [19].
In our cohort, the prevalence of C. cayetanensis co-infection was higher among PLWH with CD4+ T cell counts below 200 cells/µL. Furthermore, individuals on cART with C. cayetanensis infection had been on cART for a significantly shorter duration compared to the group without this coccidian infection. These observations strongly indicate that co-infection is correlated with the level of immune suppression of the host. Importantly, co-infection with C. cayetanensis, a coccidian intestinal parasite, was associated with clinical symptoms, namely diarrhea over the previous 6 months and marked weight loss among those infected, which is in line with previous observations [7]. These findings are also broadly consistent with the body of literature demonstrating an association between immunosuppression and the risk of diarrhea-causing co-infection acquisition [28][29][30]. Intriguingly, however, the use of co-trimoxazole was not associated with significantly lower odds of co-infection; the adjusted odds ratio was 0.69 (95% CI: 0.35-1.29). In Ghana, as in many low-income countries, prophylactic use of co-trimoxazole, which shows antimicrobial activity on C. cayetanensis, is withdrawn among PLWH whose CD4+ T cell counts are above 200 cells/µL after initiation of cART in accordance with WHO guidelines [31]. A study of HIV-positive patients in Haiti found that symptomatic infection recurred in 43% who were followed up for more than one month after an initial 10-day course of co-trimoxazole treatment [26].
Apart from the known associations between low CD4+ T lymphocyte cell count and risk of acquisition of C. cayetanensis among PLWH, we provide new information on the association of this parasite with markers of immune activation, cell proliferation, terminal differentiation, and exhaustion of CD4+ and CD8+ T lymphocyte sub-populations stratified for cART status. There were no significant differences observed in the markers of cellular proliferation (Ki67) and terminal differentiation (CD57+) between those with and without C. cayetanensis co-infection. However, PLWH with C. cayetanensis co-infection exhibited a distinct immune profile, and thereby, CD4+ and CD8+ T-lymphocytes significantly expressed elevated markers of unfavorable immune activation (HLA-DR+CD38+) and CD8+ T-lymphocytes sub-populations demonstrated significant immune exhaustion (PD-1). The expression of the Programmed Death-1 (PD-1) marker, an inducible molecule on lymphocytes that portends high susceptibility to apoptosis and negatively regulates T cell activity, may be required to prevent the acquisition of diarrhea-causing coccidian parasites such C. cayetanensis.

Implications of the Study Findings
It has been advocated that regular 6-monthly screening of PLWH with low CD4+ T cell counts for the presence of stool parasites may be a cost-effective approach for targeted therapy against diarrhea. Patients may be non-adherent to co-trimoxazole, especially when cART is initiated, and are prone to co-infections with parasites such as C. cayetanensis, especially in the setting of sub-optimal immune reconstitution. This is important, especially among those on cART who are virologically suppressed but have low CD4+ T cell counts. As this study has a cross-sectional design, any causal relationship between the presence of C. cayetanensis and elevated immune activation parameters cannot be definitely confirmed. However, as shown in our study, patients colonized with C. cayetanensis suffer more often from diarrheal disease and weight loss and present with a distinctly unfavorable immune profile, so a prolongated chemoprophylaxis with co-trimoxazole in cases with clinical symptoms may be considered.

Limitations of the Study
A recent study comparing PCR assays for the detection of Cyclospora cayetanensis in stool samples demonstrated differences in the sensitivity of the tests evaluated [19]. Consequently, comparisons of detection rates across studies using different diagnostic methods should be made cautiously. Nevertheless, all methods evaluated have shown good specificity making a significant proportion of false-positive results rather unlikely. Another point that should be kept in mind is the seasonality of C. cayetanensis [32]. However, as all subgroups in our study population were included over the period of one year in a parallel design, seasonal fluctuations should have impacted all subgroups, and therefore, the risk for bias for the group comparisons is graded as low. Another limitation of the present study is that we did not record the duration of individual co-trimoxazole prophylaxis. Recurrence is common after short treatment periods. This missing information might have brought additional insights.

Conclusions
In the modern post-cART era, acquisition of C. cayetanensis among PLWH in Ghana is largely associated with an immunosuppression profile characterized by high expression of markers of immune activation and immune exhaustion. Data Availability Statement: All relevant data are provided in the manuscript. Raw data can be made available upon reasonable request.