Seroprevalence of anti-Encephalitozoon antibodies in Spanish immunocompetent subjects.

Sera

Sera from 406 blood donors from Madrid (Spain) were studied. Hyperimmune rabbit serum developed against E. intestinalis (CDC:V297) spores was used as a control in the ELISA and IFI methods.

Microorganisms

E. intestinalis (CDC: V297) was cultured on E6 monolayers and the spores were Percoll purified to be used in the preparation of antigen and as control [1].

Antigens

Eight batches of antigen were obtained by different methods and labeled from A to H. Three methods for spore disruption were assayed. In the first case, 4.4 × 10⁹ spores were resuspended in 10 ml PBS/distilled water (1:3) and sonicated (Ultraschallprozessor UP 200S, dr Heilscher GmbH) on ice. as described [27]. The supernatant was labeled A and the sediment was sonicated again, increasing the disruption time to 80 min. One fraction of the supernatant was labeled B and the other freeze-dried overnight, resuspended in 1.5 ml of PBS and labeled C. As a second option, glass beads (425–600 μm; O.lg/100μl suspension) were used for disruption of 2.4 × 10⁹ spores in PBS. Five cycles of 5 min of vortex at maximum speed were applied, with 5 min breaks on ice. After centrifuging (2500g/lmin), the supernatant was labeled D and the sediment was reshaken under the same conditions and its supernatant labeled E. Under the same conditions of disruption, but suspending the spores in 2.5%SDS-2-ME 10%, batch F was obtained. Finally, 1.2 × 10⁹ spores suspended in PBS or 2.5%SDS-2-ME 10% were glass bead disrupted, using the FastPrepTM FpI20 (Bio 101, Savant). Five shakes of 45 sec at 6.5m/s were applied with breaks of 5 min on ice. Batches were labeled G and H respectively. The protein content was determined in all cases using the Bradford method (BIO-RAD^R).

SDS-PAGE and immunoblotting

SDS-polyacrylamide gel electrophoresis (SDS-PAGE) was used to compare protein profiles of antigen batches with higher yield with the spores from cultures. One μg of protein or 10⁷ spores were suspended in the sample buffer (10%SDS, 9M urea, 0.1 M Tris-CIH) and heated 15 min at 65°C. Samples were loaded onto each lane of 4–20% linear resolving gels (Mini-gels: Ready Gel. BIO-RAD®) and subjected to electrophoresis. A constant voltage of 200 was applied during 45 min beginning with 60 mA per gel. A continuous buffer system was used (95 mM Glycine, 12 mM Tris, 20 mM SDS). The separated proteins were either stained with silver [26] or electrophoretically transferred to polyvinylidene difluoride (PVDF) membranes (0.2 μm, BIO-RAD) following the manufacturer's intructions for Western blot (BIO-RAD). The transfer buffer contained 25 mM Tris. 192 mM glycine and 20% (v/v) methanol, pH 8.3 The protein contents, loaded onto gels, which were subsequently used for the transfer of proteins to PVDF membranes, were increased to nearly three times that used for gels that were stained with silver. The membranes were subsequently reacted for 1 hr at room temperature with a 1:400 dilution of rabbit anti-E.intestinalis serum or a 1:200 dilution of blood donor serum. After appropriate washes, the membranes were treated at room temperature for 1 hr with a 1:3000 dilution of goat anti-rabbit peroxidase conjugated IgG (SIGMA) or a 1:1000 dilution of anti-human peroxidase conjugated IgG (SIGMA). Hydrogen peroxidase (30%) and diaminobenzidine (0.005%) were used as the substrate and chromogen, respectively.

ELISA

An ELISA method was designed using the E.intestinalis antigens. In a first step, three coating concentrations of selected batches of the antigen were compared (1.6; 0.8 and 0.6 μg/ml) using the rabbit hyperimmune serum as a positive control. For the Seroprevalence study. 96 microtiter plates (Nunc-immunoplate) were coated overnight by the addition of 100 μl of antigen (0.8 μg/ml, batch H) in 0.1M carbonate buffer, pH9.6 at 4°C. Between assay steps, plates were washed 3 times in 100 mM PBS-0.05% Tween 20 (PBS-T). Plates were blocked with 250 μl/well of 0.1% bovine serum albumin (BSA) in PBS at 37°C for 1 h. Then 100 μl dilutions (1/800, 1/1600, 1/3200) of each sera, in duplicate, in PBS-T plus 0.1% BSA (PBS-T-BSA) were added and incubated for 1h 30 min at 37°C. As a control of specific binding, duplicates of sera dilutions were assayed with plates coated with the blocking solution. For conjugate control, four wells were filled with PBS-T-BSA in all plates. Appropriately diluted peroxidase conjugated goat anti-rabbit Ig (100μl/well, DAKO) in PBS-T-BSA was added and incubated for 1h at 37°C, followed by 100 μl/well of o-phelnylene-diamine (OPD, Sigma) with 0.04% H₂O₂. The reaction was stopped with 3N H₂SO₄ and plates were read at 492 nm. The titer of a serum sample was determined as being the reciprocal of the highest dilution showing optical density values of 2 standard deviations (2SD) over the media (m) [7].

IIF test

For the indirect imnumofluorescence test (IIF), spores of E. intestinalis were used, and the method was performed as described. Serum samples were diluted two fold in PBS (1:50–1:3200) [28].

Criteria for seropositivity

A serum specimen was considered as seropositive for Encephalitozoon sp. if the ELISA titer was ≥800 and IIF titer ≥80, following the criteria of Van Gool el al. [27].

Statistical analysis

For the data analysis, the Spearman Correlation Coefficient was calculated for the ELISA and IIF in blood donors.

Antigen preparation and characterization

Considerable differences in the protein yield were observed in the eight antigen batches prepared (Table 1). Mechanical disruption of the spores with glass beads showed higher efficiency as did the utilisation of SDS-2ME solution as a medium for the spore disruption. In this step, batches F, G and II were selected for further analysis.

SDS-PAGE protein analysis showed differentiable protein patterns ranging from 8 to 251 kDa (Fig 1). In spite of the complexity of the protein profile, a number of shared bands showed that batches F and H. treated with SDS-2ME had similar protein patterns. However, only H had a band higher than 200 kDa. also visible in the spore extract sample. Furthermore F and H had profiles closer to the spore extract than G.

In table 2 an estimation of molecular weights of the more prominent bands of the three antigenic batches is shown compared with the spore extract sample. Batch G showed more proteins of low molecular weight and less common banding with the spore extract than F and H. Due to the similar characteristics of batches F and H, we decided to select H to continue the comparative analyses with G, because of greater efficiency in its preparation.

The characterization of E.intestinalis by electrophoresis and immunoblot has been carried out by different authors [1,4,16,19], identifying differentiable protein/antigenic profiles in relation to the other human Encephalitozoon species. Although the methods of spore disruption, protein extraction, and the range of electrophoretic gradient used directly affect the profile obtained, some of the fractions have been detected by several authors. In this line, and bearing in mind the approximate estimation of molecular weights, proteins of 16, 19, 22, 23, 28, 32, 35, 40, 50, 55, 64 and 125 shown in this study have also been described by others [1,4,16,19]. However, recent studies with monoclonal antibodies have clearly shown that one band may contain two or more proteins of different origin, as is the case of the 60 kDa band described by Pigneau et al [19], corresponding to a polar tube as well as to a spore wall protein. Moreover, an intraspecific diversity among the spore wall proteins has been described [9,10,19].

In order to know which antigenic extract was more efficiently recognized by specific antibodies, a comparative ELISA using batches H and G for coating the plates at 0.8 μg/ml, and hyperimmune rabbit sera was performed (Fig2). A higher signal was obtained with batch H, which was selected for the seroprevalence study. A further assay was performed to optimize the coating concentration in ELISA. The 1.6, 0.8 and 0.4 μg/ml concentrations were studied, selecting 0.8 μg/ml to continue the study.

Seroprevalence of anti-Encephalitozoon antibodies in Spanish blood donors

Among 406 blood donors, 1 1 had high ELISA titers (≥3200), which represents 2.7%; and 22 (5.4%) fulfilled the criterion of seropositivity (ELISA≥800 and IIF≥100). Mean optical density (OD) values are shown in table 3. In the IFF test, only one of the positive sera had a titer of 100, 2 of 200, 7 of 400, 4 of 800, and 2 sera of 1600. To estimate the correlation between both immunologic techniques 94 sera were selected: 17 showed very high OD values (OD≥m+2SD), 10 very low OD (< m), and 67, intermediate OD values. The Spearman Coefficient (r=0.5) showed a positive result indicating a relative relationship between both techniques. In reference to the reactivity of blood donors' sera in the immunoblot technique, a correlation was observed with the other tests. Sera with low ELISA and IIF titers didn't react at all in immunoblot, and the positive sera mainly recognized bands of medium or high molecular weight. The antigenic fraction recognized in most cases was in the 45–66 kD range.

Most of previous studies of seroprevalence of microsporidia have been carried out using E.cuniculi as antigen, and positives were found in visitors from tropical countries but not, or only sporadically, in healthy blood donors [5,14,24]. To our knowledge, there is only one study on the seroprevalence of E. intestinalis in immunocompetent individuals carried out in healthy Dutch donors and pregnant French women which showed 8% and 5% of positivity respectively [27]. In both studies, a soluble antigen obtained from disrupted spores was used, which proved specificity at the genus level [27]. The data obtained correlates with the present study and furthers the idea that immunocompetent individuals are frequently exposed to Encephalitozoon infection. To date, E. intestinalis pathology in immunocompetent individuals is mainly associated with limited diarrhea episodes [15,18,20,22,25] while other species have also been related to renal and neurologic disorders [30]. This is particularly notable because Encephalitozoon parasites are capable of crossing the placenta and they are known to cause severe congenital infection in animals [23,30]. Thus, future studies are needed to evaluate the relation between Encephalitozoon infection and symptomatic disease in immunocompetent individuals.