Serological survey of Leishmaniainfection in blood donors in Salvador, Northeastern Brazil

Background Visceral Leishmaniasis is endemic to Brazil, where it is caused by Leishmania infantum (syn. Leishmania chagasi). Following parasite inoculation, individuals may experience asymptomatic infection, raising the possibility of parasite transmission through the transfusion of contaminated blood products. In the present work, we evaluated the prevalence of asymptomatic Leishmania infection among blood donors in Salvador, northeastern Brazil. Methods Peripheral blood was collected from 700 blood donors attending the Blood Bank of Bahia (HEMOBA/SESAB), from January to September 2010. We evaluated anti-Leishmania serology by ELISA, employing Soluble Leishmania Antigen (sensitivity 90% and specificity 95%). The presence of parasite DNA was determined by qPCR, targeting a single copy gene (G6PD), and by end-point PCR, targeting multiple targets, namely a segment located in the Leishmania rRNA locus (ITS) and the conserved region of kinetoplastid DNA (kDNA) minicircles. Results The blood-donor population was comprised of 74.5% of males with a mean age of 34 years. Anti-Leishmania serology by ELISA was positive in 5.4% (38/700) individuals. One individual was also positive for Chagas’ disease and another tested positive for Syphilis. Employing qPCR, parasite DNA was not found in any of 38 seropositive samples. However, by ITS PCR, 8/38 (21%) samples were positive and this positivity increased to 26/38 (68%) when we targeted kDNA amplification. Agreement between both techniques (ITS and kDNA PCR) was fair (kappa = 0.219). Conclusions These results indicate that asymptomatic infection is present among the blood donor population of Salvador, a finding that warrants a broader discussion regarding the need to implement specific screening strategies. Electronic supplementary material The online version of this article (doi:10.1186/1471-2334-14-422) contains supplementary material, which is available to authorized users.


Background
Visceral leishmaniasis (VL) or kalazar is a serious public health problem worldwide, and approximately 500,000 new cases are reported each year { [1] #67}. Leishmania infantum is the causative agent of VL in Brazil, which accounts for 90% of cases in the Americas { [2] #257} and where 3392 VL cases were reported in 2012 { [3], #258}. In the last decades, VL is becoming more prevalent in urban centers of Brazil, especially in expanding cities with different patterns of economic and social development { [4] #259}. Patent VL is classically characterized by the presence of irregular fever, paleness, and splenomegaly and, in the absence of treatment; it is a progressive infection with fatal outcome { [3] #279}. However, a large number of infections remain asymptomatic, in which individuals are infected by Leishmania but display an apparent healthy condition (rev. in { [5] #260}). There are concerns that asymptomatic individuals remain carriers of the infection and, as such, may play an important role in the epidemiology of VL, especially in urban concentrations where the disease is present { [6] #261}.
Herein, our goal was to determine the prevalence of anti-Leishmania antibodies among blood donors of Salvador, Bahia, situated in northeastern Brazil.

Ethical considerations
This study was approved by the research ethics committee of CPqGM-FIOCRUZ-BA (No. 215/2010). Individuals involved in this study were required to sign the Informed Consent Form.

Study design and population
The purpose of this study was to identify asymptomatic infection in blood donors attending the Hematology and Hemotherapy Foundation of Bahia State (HEMOBA/ SESAB), situated in Salvador, BA, Brazil. The necessary sample size of approximately 220 individuals was estimated according to the following parameter: prevalence of an asymptomatic infection rate of 9%, based on a study conducted in Natal { [19] #37}, a VL endemic area { [26] #254}. Blood was obtained from 700 donors attending HEMOBA/SESAB and blood donations were screened over a period of nine months (January to September 2010). Donors resided in Salvador and other municipalities from BA.

Sample collection
Whole blood was collected in quadruple blood packs (Fresenius/KAB), using the top-and-bottom system. Aliquots of 5 ml of peripheral blood were obtained following blood collection and were placed into Vacutainer® tubes (Becton Dickinson) containing Heparin as anticoagulant. Serum was separated by centrifugation and stored at −20°C for subsequent serologic studies. Aliquots of 200 μl of blood were used for DNA extraction.
Soluble Leishmania antigen SLA was prepared from L. infantum MCAN/BR/00/ BA262) and from L. amazonensis (MHOM/BR/1985/ BA32) promastigotes maintained in Schneider (LGC Scientific) medium (Sigma) supplemented with 10% inactivated fetal bovine serum, 100 U/ml penicillin and 100 ug/ml streptomycin (Life Technologies). The parasites were initially submitted to 10 alternating cycles of freezing in liquid nitrogen and thawing in a water bath and were then centrifuged at 1600 × g, 4°C for 15 min. The supernatant containing SLA was collected and protein content was quantified using the Micro BCA TM Protein Reagent Kit assay (Pierce).

ELISA to detect anti-Leishmania antibodies
Anti-Leishmania serology was performed by ELISA, as described { [27] #20}. Briefly, 96-well plates (Linbro/ Titertek) were coated with 10 μg/mL of antigen (L. infantum or L. amazonensis SLA) and were incubated overnight at 4°C. After three washes with PBS/0.5% Tween, plates were blocked for 1 hour at 37°C with PBS/Tween 0.5% plus 1% fetal bovine serum (FBS) (Life Technologies). Plasma samples, diluted 1:100 in PBS/ Tween 0.5% plus 1% FBS were added in duplicate and plates were incubated for 2 hours at room temperature. After three rounds of washing, wells were incubated with anti-human IgG conjugated to alkaline phosphatase (SIGMA) at a 1:2500 dilution in PBS/Tween 0.5% plus 1% FBS for 1 hour at room temperature. Plates were then washed and were further incubated for 30 minutes with a chromogenic solution of p-nitrophenyl phosphate (SIGMA) in sodium carbonate buffer pH 9.6 containing 1 mg/mL of MgCl 2. In all experiments, the values obtained were subtracted from those obtained in the background. ELISA experiments were repeated twice yielding similar results. The ELISA cut-off value for L. infantum and L. amazonensis SLA was established using ROC curves (please see Statistical analysis below). In this case, we employed a panel of positive sera (n = 20), obtained from VL patients [defined by clinical and laboratory signs { [28] #280} and a positive bone marrow aspirate]. Control sera (n = 20) were obtained from negative individuals, living in areas free of L. infantum. These sera were randomly selected from a serum bank (LIP-CPqGM-FIOCRUZ).

DNA extraction
DNA was extracted from 200 μL of whole blood using Illustra Genomic Prep Blood Mini Spin Kit™ (GE Healthcare), following the manufacturer's instructions. DNA was eluted into 200 μL of Elution buffer and stored at −20°C. Purity of genetic material was assessed by absorbance at 260/280 nm.

Real-Time PCR
Detection of Leishmania DNA by Real Time PCR was performed as described by { [23] #24}. Briefly, real time PCR was conducted in an ABI PRISM® 7500 Sequence Detection System (Applied Biosystems) employing SYBR-Green® PCR Master Mix (Invitrogen), 50 ng DNA and 200 nM of each forward and reverse primers. Thermal cycle conditions consisted of a denaturation step at 95°C for 10 minutes followed by 40 cycles at 95°C for 15 seconds and 60°C for 1 minute. Oligonucleotides (Fwd G6PD 5′-CCCGAGGGCAGCACTTG-3′ and Rev G6PD 3′-CCAC CGGTCGTTGTTGATG-5′) were designed using Primer-Express3.0 (Applied Biosystems) and were based on the L. infantum G6PD gene (glucose-6-phosphate dehydrogenase) (GeneBank: DQ212794.1). For parasite quantification, a standard curve was constructed based on DNA extracted from serial dilutions of L. infantum, ranging from 10 6 to 10 1 parasites. Analysis and acquisition of data were performed with SDS software (Applied Biosystems).

Statistical analysis
Statistical analysis was performed using Prism v.5.0 (Graph Pad Software). ROC (Receiver Operating Characteristic) curves were calculated based on ELISA assays performed with SLA obtained from L. amazonensis and L. infantum. The cut-off, as determined by high sensitivity and specificity, from the highest probability of discrimination established by the curve, was therefore determined. The performance of each SLA was established by the parameters obtained from ROC curves values (AUC, p values and likelihood ratio). Nonparametric Mann-Whitney test was used to compare the frequency of seropositive individuals with a positive PCR result or a negative PCR result. The p value was considered significant when < 0.05. Agreement between PCR assays targeting Leishmania ITS and kDNA were determined by the kappa index.

Results
Anti-L. infantum serology in blood donors recruited at HEMOBA/SESAB Initially, we determined the sensitivity and specificity of the Soluble Leishmania Antigen (SLA) preparations [obtained from L. infantum or from L. amazonensis (the latter an etiological agent of Cutaneous Leishmaniasis)] when employed as the capture antigen in ELISA assays. With the results obtained, we constructed two ROC curves (Figure 1). L. infantum SLA performed better (AUC: 0.99, p < 0.0001, Likelihood ratio: 18.0) (Figure 1) when compared to L. amazonensis SLA (AUC 0.87; p < 0.0001 and Likelihood Ratio 15.0). Therefore, SLA from L. infantum presented higher sensitivity and specificity when compared to SLA from L. amazonensis and was selected for subsequent experiments. The cut off value, based on higher sensitivity and specificity of L. infantum SLA, was established as 0.01685.

Cross-reaction among Leishmania and other blood borne pathogens
Next, we investigated the presence of anti-Leishmania antibodies in the sera from 700 blood donors (74.5% male) attending HEMOBA/SESAB. The median age was 42 years: 31.7% individuals had between 19 and 28 years, 32.8% between 28 and 38 years, and 34% were older than 38 years of age. The reactivity of blood donors' sera to L. infantum SLA was significantly higher compared to that obtained with negative controls (Figure 2). As expected, the reactivity of sera from confirmed VL patients was higher (p < 0.0001) compared to both blood donors and negative controls (Figure 2). Among the blood donors, we identified 38 individuals (5.4%) (81.5% male) that reacted above the established cut-off value (0.01685) and were thus considered seropositive against L. infantum. These 38 samples were also tested for cross-reactivity to other agents, following the routine blood screening investigation performed at HEMOBA/SESAB. Among the 38 Leishmania seropositive samples, 1/38 (0.02%) reacted positive for Chagas' disease whereas another sample (1/38) reacted positively for Syphilis. Of note, among the 38 blood donors that were seropositive for Leishmania, 26/38 (68%) currently live in Salvador whereas 8/38 (21%) live in adjacent municipalities (address information was missing from four samples).

Detection of Leishmania DNA in blood donors' samples
We then investigated the parasite load in the 38 (5.4%) Leishmania seropositive samples ( Figure 2). To do so, DNA extracted from the 38 blood samples was submitted to quantitative PCR (qPCR) targeting the single copy G6PD gene { [30] #243}. A standard curve (threshold cycle versus logarithm of the amplicon copy number) was constructed using DNA extracted from L. infantum, ranging 10 6 to 10 1 parasites (Figure 3). The standard curve showed a good linear correlation (R 2 = 0.971) when plotted against the DNA equivalent to the number of parasites and the efficiency of the assay was 97%. We did not detect any level of interference due to the presence of human DNA as a similar efficiency of amplification was obtained when standard curves were generated from parasite DNA spiked with human DNA or from DNA extracted from a mixture of parasites plus human DNA (data not shown). Using the G6PD qPCR assay, we were not able to detect Leishmania DNA in any of the 38 blood donors that were seropositive by ELISA (Table 1). We then investigated whether end-point PCR assays targeting multiple copy genes would improve detection of parasite DNA. Targeting the region located between the small and the large subunit of the Leishmania rRNA gene array, including the Internal Transcribed Spacers (ITS) and the 5.8S rRNA { [24] #252}, we detected positive amplification in 8/38 (21.0%) samples (Table 1). In another assay, we amplified the conserved region of minicircles, present in 10,000-20,000 copies, located within the Leishmania kinetoplast { [25] #253}. This assay was positive in 26/38 (68.4%) samples (Table 1), raising the detection level by 47,4% when compared to ITS PCR ( Table 1). The concordance between kDNA and ITS PCR was fair (kappa = 0.219) ( Table 2). Lastly, positivity in either PCR assays, ITS or kDNA amplification, was dissociated from serology    Figure 4A and B, respectively).

Discussion
Asymptomatic Leishmania infantum infection can be detected initially with antibodies { [31] #275}; antibody responses decline with time in parallel to the development of a cell mediated immune response { [32] #282}. Therefore, serologic responses to Leishmania indicate acute infection, in the presence of symptoms or not. Reports of Leishmania transmission through the use of contaminated blood products { [14] #7; [10] #9} have raised the question whether asymptomatic infection among blood donors poses a problem for the blood supply, especially in regions where VL is endemic. Presently, we investigated the anti-Leishmania serologic response among blood donors from Salvador, northeastern Brazil. Additionally, we employed three PCR-based assays to determine the presence of Leishmania DNA in samples with positive serology.
Among the 700 blood donors recruited at HEMOBA/ SESAB, the major blood donation center in Salvador, we detected 38 individuals (5.4%) with a positive serology to Leishmania. This finding is similar to that reported in Montes Claros (5.5%) { [33] #265} but lower than that observed in Natal (9%) { [20] #10} and in Campo Grande (15.6%) { [21] #262}, both of which are endemic regions for VL. The lower prevalence detected in our study may be due to the fact that Salvador is not a VL endemic area per se, differently from the other three municipalities. Additionally, differences in the sensitivity of the tests employed herein (SLA-based ELISA) versus those used in Natal (FML-ELISA) { [20] #10}, Montes Claros and Campo Grande (IFAT) { [33] #265; [21] #262} may contribute for the differing results. In our study, an ELISA assay employing L. amazonensis SLA was less sensitive when compared to L. infantum SLA (Figure 1), suggesting that the choice of antigen impacts on the results    [43] #14}. These results are in accordance with our observation that the higher the number of repetitions of a given amplification target, such as the ITSs or kDNA, the higher the sensitivity of the assay { [45] #278}. In our hands, kDNA amplification also proved a useful tool for the diagnosis of Cutaneous { [29] #264} and Mucocutaneous Leishmaniasis { [46] #281}. The fact that either ITS or kDNA amplification was negative in 12/38 samples (31.6%) with a positive Leishmania serology could again be explained by the presence of very low parasite numbers and, hence, few targets were available for PCR amplification. Another hypothesis is the presence of parasite remnants within phagocytic cells, rendering a positive amplification of Leishmania DNA by PCR.
In the current setting, blood screening for the presence of VL would decrease blood supply by 5%, negatively impacting on the blood supply provided by HEMOBA/SESAB. Hamsters transfused with blood from an experimentally infected animal developed disease as shown by presence of ascites, cachexia and death { [47] #44}. Foxhounds that received packed red blood cell transfusions from seropositive donors also tested positive for Leishmania antibodies { [48] #43}, indicating transmission of Leishmania to recipient dogs by blood transfusion. In a study conducted in Bangladesh, 1.195 blood donors were screened, antibodies were found in three individuals but none developed VL during a 6 month follow-up period { [11] #250}, suggesting that the number of circulating parasites is below the threshold necessary to initiate symptoms.

Conclusions
Given that positive serology and/or presence of parasite DNA does not guarantee transfusional risk, the importance of our results lies in the recognition of the presence of cryptic Leishmania infection among blood donors from Salvador. Leishmania testing is not mandatory in blood banks in Brazil but knowledge of positive serology would impose blood elimination. Since VL is expanding and control measures are not efficient, a broader discussion is needed regarding the need to implement specific screening strategies and, possibly, the risk of Leishmania transmission by blood transfusion.