Anal. Chem. 2007, 79, 4690-4695
Detection of Malaria Infection via Latex Agglutination Assay Duangporn Polpanich,† Pramuan Tangboriboonrat,*,† Abdelhamid Elaissari,‡ and Rachanee Udomsangpetch§
Department of Chemistry, Faculty of Science, Mahidol University, Rama VI Rd, Phyathai, Bangkok 10400, Thailand, Unite´ Mixte CNRS-bioMe´ rieux, UMR-2714, Ecole Normale Supe´ rieure de Lyon, 46 alle´ e d’Italie, 69364 Lyon, France, and Department of Pathobiology, Faculty of Science, Mahidol University, Rama VI Rd, Phyathai, Bangkok 10400, Thailand
A rapid test for malaria diagnosis based on specific agglutination of sensitive polystyrene particles containing carboxylic acid with antigen or antibody molecules in the presence of their corresponding antibody or antigen in human plasma has been achieved. The particle-malaria antigen conjugate (PAgC), particle-monoclonal IgG antibody to Plasmodium falciparum heat shock protein 70 conjugate (PmAbC), and particle-polyclonal IgG antibody to P. falciparum malaria conjugate (PpAbC) were prepared via adsorption process. The higher affinity of the malaria antigen adsorption onto particles was observed compared to that of the antibodies. Immunoagglutination of sensitive latex particles was monitored by measuring the change in turbidity, and the aggregate’s formation was clearly observed under optical microscope. The efficacy in malaria diagnosis of the conjugated particles evaluated at an outpatient malaria clinic (Mae Sod, Thailand) indicated a success detection of antibody or antigen. Sensitivity of PAgC, PmAbC, and PpAbC for P. falciparum was 84%, 90%, and 90%, respectively, while specificity for malaria disease was 70% for PAgC and 80% for PmAbC and PpAbC. The rapid agglutination-based latex particles assay developed herein showed a great potential for diagnosis of malaria infection. Malaria remains one of the world’s most prevalent and lethal parasitic diseases1 infected by Plasmodium falciparum. Prompt and accurate diagnosis is the key to effective disease management with regards to both endemic and the problem of drug resistance. Traditionally, the malaria parasites have to be microscopically diagnosed.2 However, even for a skilled microscopist, the process is not straightforward but has proved to be time-consuming and labor intensive.1,3 Simple and rapid diagnostic tests (RDTs) have, therefore, been developed especially for field use. Among the RDTs, an immunochromatographic assay based on the detection * To whom correspondence should be addressed. Phone: +662 201 5135. Fax: +662 354 7151. E-mail:
[email protected]. † Department of Chemistry, Mahidol University. ‡ Unite ´ Mixte CNRS-bioMe´rieux, Ecole Normale Supe´rieure de Lyon. § Department of Pathobiology, Mahidol University. (1) World Health Organization. Report of a Joint WHO/USAID Informal Consulation; WHO/MAL/2000.1091; World Health Organization: Geneva, 2000. (2) Moody, A. Clin. Microbiol. Rev. 2002, 15, 66-78. (3) Hanscheid, T. Clin. Lab. Haematol. 1999, 21, 235-245.
4690 Analytical Chemistry, Vol. 79, No. 12, June 15, 2007
of malaria antigen in blood, in the form of dipstick or test strip bearing monoclonal antibody, has, frequently, been employed. The commercial test kits can be divided into two groups according to P. falciparum-specific antigen targets, i.e., histidine-rich protein II (HRP II) and parasite lactate dehydrogenase enzyme (pLDH).4 In spite of their uniformly high specificity, false-positive results are reported in blood from patients with rheumatoid factor,5,6 while the presence or absence of pLDH is a comparatively reliable diagnostic target,4 the HRP II test can remain positive for 7-14 days after parasite clearance. Due to the RDTs’ high cost, the latex agglutination test which involves in vitro aggregation of latex particles coated with antigen (or antibody), called latex-antigen (or -antibody) conjugate, in the presence of specific antibody (or antigen) is of great interest especially in biomedical applications.7,8 This test is claimed to be the quickest and easiest method which is portable and suitable for use in the field. However, it might be highly susceptible to nonspecific adsorption of proteins, resulting in nonspecific agglutination, when highly hydrophobic latex particle, e.g., polystyrene (PS), was employed.9 To overcome this problem, hydrophilic particles such as poly(St/acrylamide) (St/AAm),10 poly(St/sodium styrene sulfonate) (St/NaSS),11 and poly(St/acrylic acid) (St/AA)10 were used instead. Although the adsorption of various proteins onto a polymeric surface is an irreversible process and rapidly takes place,12 there has been no report concerning the adsorption behavior of malaria protein onto the latex particle surface. Due to the well-known chemistry of the carboxyl group,12 the PS particles containing carboxylic acid, synthesized and well characterized in our previous work,13 were immobilized with malaria antigen to prepare particlemalaria antigen conjugate (PAgC). To obtain the better meaning (4) Wongsrichanalai, C. Trends Parasitol. 2001, 17, 307-309. (5) Grobusch, M. P.; Alpermann, U.; Schwenke, S.; Jelinek, T.; Warhurst, D. C. Lancet 1999, 353, 297. (6) Iqbal, J.; Sher, A.; Rab, A. J. Clin. Microbiol. 2000, 38, 1184-1186. (7) Martin, C. R.; Mitchell, D. T. Anal. Chem. 1998, 70, 322A-327A. (8) Park, J.; Kurosawa, S.; Watanabe, J.; Ishihara, K. Anal. Chem. 2004, 76, 2649-2655. (9) Singer, J. M.; Plotz, C. M. Am. J. Med. 1956, 21, 888-892. (10) Suzawa, T.; Shirahama, H. Adv. Colloid Interface Sci. 1991, 35, 139-172. (11) Yoon, J. Y.; Kim, J. H.; Kim, W. S. Colloids Surf., B 1998, 12, 15-22. (12) Molina-Bolivar, J. A.; Galisteo-Gonzalez, F. Latex Immunoagglutination Assays. In Colloidal Biomolecules, Biomaterials, and Biomedical Applications; Elaissari, A., Ed; Marcel Dekker, Inc.: New York, 2004; pp 53-101. (13) Polpanich, D.; Tangboriboonrat, P.; Elaissari, A. Colloid Polym. Sci. 2005, 284, 183-191. 10.1021/ac070502w CCC: $37.00
© 2007 American Chemical Society Published on Web 05/19/2007
Figure 1. TEM micrograph of PS particles containing carboxylic acid.
of malaria diagnosis, two types of antibody to P. falciparum malaria, i.e., mouse monoclonal IgG antibody to P. falciparum heat shock protein 70 (HSP70) and human polyclonal IgG antibody to P. falciparum malaria, were used instead of antigen to prepare particle-monoclonal IgG antibody to P. falciparum HSP70 conjugate (PmAbC) and particle-polyclonal IgG antibody to P. falciparum malaria conjugate (PpAbC), respectively. Incubation time and protein concentration influence on the adsorbed amount were investigated. The agglutination of the conjugated particles in the presence of P. falciparum-positive plasma was also explored by using turbidimetry and an optical microscope (OM). The efficacy in malaria diagnosis of each conjugated latex particles was evaluated via naked eyes in the field. EXPERIMENTAL SECTION Latex Preparation. The preparation of monodispersed PS particles containing carboxylic acid (Figure 1) via a batch soapfree emulsion polymerization using ammonium persulfate as an initiator was described elsewhere.13 The hydrodynamic diameter (Dh) of the particles measured by using a Coulter counter (Coulter Electronics, Hialeah) was 349 ( 21 nm. Conductometric backtitration was performed to determine the surface charge density (σ) of the particles by the following equation
F σ ) Dh[Acid]NAe 6
(µC/cm2)
where F is the density of the polymer (1.045 g/cm3 for PS), [Acid] is the number of microequiv of acid groups per g, NA is Avogadro’s number (6.022 × 1023/mol), and e is the electronic charge in coulombs (1.602 × 10-19 C). The σ value of the particles was -271 ( 9 µC/cm2. Preparation and Characterization of Malaria Antigen. P. falciparum (strain AMB47) was cultured in human RBC group O at 5% hematocrit in RPMI-1640 medium (pH 7.2), gentamicin (40 µg/mL), HEPES (25 mM) (GIBCOBRL), and human serum (10%). The parasites were incubated (Heraeus, Hera cell) at 37 °C and
CO2 atmosphere (5%) until reaching mature stages (>3% parasitemia). The mature parasites were enriched by gradient centrifugation (800g, 25 min) using Percoll (60%) (Sigma). After sonication for 25 s and then centrifugation in order to remove insoluble pigments, protein concentration of the enriched soluble fraction was determined by using the Bradford method.14 Enzyme-linked immunosorbent assay (ELISA) was applied for characterization of antigen recognized by the malaria patient’s plasma. A PS immunoplate (nunc-Immuno Plater Maxisorp, Denmark) was coated with either parasite antigen or uninfected RBCs lysate (0.5 µg/well) diluted in phosphate-buffered saline (PBS) (pH 7.4) and incubated overnight at 4 °C. The plate was blocked with blocking buffer (0.5% casein in PBS and 0.1% Tween 20) for 2 h and then washed with PBS-Tween. Plasma (1:100 dilution) (50 µL) was added into duplicate wells and incubated at 37 °C for 2 h. The following steps were carried out as described elsewhere.15 Finally, the optical density (OD) at 405 nm was measured by an automated microplate reader (Perkin-Elmer, Germany). Preparation and Characterization of Antibody to P. falciparum Malaria. Mouse monoclonal IgG antibody to P. falciparum HSP70 and human polyclonal IgG antibody to P. falciparum malaria were purified by passing hybridoma cell culture supernatant and a pool of serum from P. falciparum-infected patients, respectively, through a ProPur Protein G column (Midi spin column, nunc). Before loading to the column, the sample was concentrated by using saturated ammonium sulfate solution and then dialyzed. Purification steps were then performed according to the manufacturer’s instruction. Concentrations of the antibodies were examined by the Bradford method. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was conducted on a Bio-Rad electrophoresis system to explore the purity of the antibodies after passing through the column. In which, the sample was solubilized in a Laemmli buffer containing 2-mercaptoethanol as a reducing buffer. Broad-range prestained standards were concomitantly run with the sample on polyacrylamide gel (12%) at 100 V for 150 min. When the process was complete, the gel was stained with coomassie blue solution (0.1% w/v) overnight. The Rf value (ratio of distance moved by protein to that moved by dye) used for determination of the molecular mass of protein was calculated from TotalLab 1D Gel Analysis V1.0 software. The isoelectric point (pI) of the antibodies investigated by using the isoelectric focusing (IEF) technique was performed on Phast System equipment with Phast Gel IEF polyacrylamide media (Amersham Biosciences) that covered pH 3-9. After staining with coomassie blue, the pI of each band was evaluated by the similar software as that for SDS-PAGE. Preparation of PAgC, PmAbC, and PpAbC. A given amount of the latex and malaria antigen or antibody to P. falciparum malaria was added into an Eppendorf tube (0.5 mL) and then made up to 500 µL with phosphate buffer (10 mM). The mixture was incubated, while being gently stirred at 25 °C for 2 h before centrifugation (Jouan, MR 22) at 12 000 rpm for 20 min, and resuspended in phosphate buffer (10 mM, pH 7.0) mixed with (14) Bradford, M. M. Anal. Biochem. 1976, 72, 248-254. (15) Jangpatarapongsa, K.; Sirichaisinthop, J.; Sattabongkot, J.; Cui, L.; Montgomery, S. M.; Looareesuwan, S.; Troye-Blomberg, M.; Udomsangpetch, R. Microbes Infect. 2006, 8, 680-686.
Analytical Chemistry, Vol. 79, No. 12, June 15, 2007
4691
BSA (0.5 mg/mL). The concentration of the residual protein in the supernatant was measured. Detection of Immunoagglutination. The absorbance (i.e., turbidity) change of the conjugated latex (4.86 × 1012 particle number/L, 950 µL) after mixing with various concentrations of P. falciparum-positive plasma or malaria naive control (50 µL) at time 300 s was examined by UV-vis spectrophotometer at 570 nm (Jasco, V-530). All experiments were carried out in triplicate measurements. An OM was also applied to investigate the occurrence of the immunoagglutinated particles. Latex Agglutination Test. The field test was conducted at Malaria Clinic, Mae Sod, Tak province, Thailand. For each patient who gave a P. falciparum slide-positive result, a finger-prick blood was drawn into a haparinized capillary tube. Plasma (2 µL) was mixed with the conjugated latex (1% w/v) (9.5 µL) on a glass slide and, subsequently, mixed with a disposable tip. After being gently shaken for 2 min, sample was considered negative if no agglutination was visually detected within 2 min. A score of 3+ was given when agglutination took place within 30 s, 2+ when agglutination was observed during 30 s and 1 min, and 1+ when agglutination became visible between 1 and 2 min. A healthy adult living in Bangkok, the nonmalaria-infected region, was used as naive control. Parameters measured were sensitivity and specificity. The sensitivity was calculated as true-positive/(true-positive + falsenegative) and specificity as true-negative/(true-negative + falsepositive). All patients or their attendant relatives gave fully informed written consent to blood sampling. This study was approved by the Committee on Human Rights Related to Human Experimentation, Mahidol University, and by the Ministry of Health, Thailand. Statistical Analysis. Data were analyzed using SPSS for Windows (version 10.0, Chicago, IL). Statistical significance of antibody level between malaria patient and naive control was evaluated by parametric method, two-tailed t test. Correlation was determined by Spearman rank correlation analysis. RESULTS AND DISCUSSION Malaria Antigen. ELISA was carried out to verify malaria antigen by using patient’s plasma. The level of anti-P. falciparum antibody detected by median OD values for P. falciparum patients, P. vivax patients, and naive controls were 0.196, 0.056, and 0.043, respectively. The level of anti-P. falciparum antibody in P. falciparum plasma was significantly higher than that in naive control (P < 0.05, two-tailed t test). In contrast, the OD value of P. vivax plasma containing the low level of antibody was not significantly different from that of naive control (P ) 0.152, twotailed t test). Results, therefore, indicated that the crude P. falciparum antigen was capable of recognizing antibody in P. falciparum-infected patients and was specific to P. falciparum species. Antibody to P. falciparum Malaria. The mouse monoclonal IgG antibody to P. falciparum HSP70 and human polyclonal IgG antibody to P. falciparum malaria were purified by using a protein G column, which has a very high affinity for the Fc region of IgG. Their purity monitored by SDS-PAGE indicated that both monoclonal IgG and polyclonal IgG antibodies were highly pure. The pI of the monoclonal IgG antibody to P. falciparum HSP70 (6.31 ( 0.04) was slightly acidic than that of polyclonal IgG antibody to P. falciparum malaria (6.83 ( 0.01). 4692 Analytical Chemistry, Vol. 79, No. 12, June 15, 2007
Figure 2. Adsorption isotherms of PAgC (2) (incubation time 2 h, pH 6.8, buffer concentration ) 10 mM), PmAbC (×), and PpAbC (9) (incubation time 2 h, pH 6.0, buffer concentration ) 10 mM).
Adsorption Kinetics. From the amount of protein adsorbed onto the particle (Γads) as a function of incubation time, it was observed that the adsorption kinetics of PAgC and of PpAbC were similar at the investigated pH, i.e., they drastically increased in the initial period and then approached the plateau of 2.8 and 2.7 mg/m2, within only 20 min, respectively.16 It is well-known that the adsorption of protein from aqueous solution onto the solid substrate is entropy driven.17,18 In this case, the increase in entropy is a result of freedom gained by water molecules released from the partial dehydration of protein and substrate’s surface. It was believed that the hydrophobic and hydrophilic interactions simultaneously took place resulted in the immobilization process of the protein molecule to the surface via orientations, side-on and endon. Moreover, the number of contact points of the protein molecules onto the substrate increased with incubation time. Due to its difficulty to detach from the surface of substrate, the protein adsorption became an irreversible process. Since protein in solution could not be stable within the longer time,19 the incubation time of 2 h was selected for further experiments. Adsorption Isotherms. The adsorption isotherms, i.e., the plot of Γads versus the equilibrium protein concentration (Ceq), of PAgC, PmAbC, and PpAbC were studied and the curves are presented in Figure 2. The initial slope (at low Ceq) of the curve indicated the high affinity of protein to the substrate surface, i.e., almost all proteins in the system were immobilized onto the particle surface.12 It could be observed that, for PAgC, the Γads greatly increased up to the plateau value where the total particle surface was saturated with the malaria antigen (2.8 mg/m2), whereas the less steep slope of both antibodies than that of the antigen indicated the lower affinity of IgG antibodies to the particle surface. The plateau values of the monoclonal IgG antibody and polyclonal IgG antibody isotherms which were 2.1 and 5.5 mg/m2, respectively, could be explained by considering pI values. Since the pI of the polyclonal IgG antibody was less acidic than that of monoclonal IgG antibody, the polyclonal antibody at pH 6.0 carrying more positive charges could enhance the attractive electrostatic interaction to the negative particle. After the adsorption process, PAgC, PmAbC, and PpAbC were redispersed in the phosphate buffer at the same pH and ionic (16) Polpanich, D. Latex-Peptide Conjugates for Diagnosis of Malaria Infection. Ph.D. Thesis, Mahidol University, 2007. (17) Norde, W.; Lyklema, J. J. Biomater. Sci., Polym. Ed. 1991, 2, 183-202. (18) Haynes, C. A.; Norde, W. Colloids Surf., B 1994, 2, 517-566. (19) Serra, J.; Puig, J.; Martin, A.; Galisteo, F.; Galvez, Ma. J.; Hidalgo-Alvarez, R. Colloid Polym. Sci. 1992, 270, 574-583.
Table 1. Agglutination of PAgC Tested with P. falciparum- and P. vivax-Infected Plasma for Sensitivity and Species Specificity with Comparison to Malaria Naive Plasma degree of agglutination no. of positive cases (% cases)a no. of no. parasites/µL positive no. of mean ( SD cases cases (ranges) (%)a
1+
2+
3+
P. falciparum
19
70 789 ( 80 833 (5000-355 000)
16 (84.2)
2 8 6 (12.5) (50.0) (37.5)
P. vivax
10
17 250 ( 6917 (2500-25 000)
10 (100.0)
1 4 5 (10.0) (40.0) (50.0)
malaria naive
10
0
3 (30.0)
3 (100.0)
0
0
a Result was considered positive when the agglutination appeared within 2 min.
Figure 3. (a) Immunoprecipitin curves showing the absorbance change of the PAgC when the latexes were mixed with various concentrations of P. falciparum plasma (solid line) and naive control (dashed line). (b) At the equivalent zone, the larger agglutinated particles (average diameter of 0.9 mm) of PAgC (1% w/v) (bII) after mixing with P. falciparum plasma were clearly observed by OM compared with naive control (bI).
strength as used in the adsorption experiment. The amount of desorbed antigen or antibody was determined by the Bradford method, and it was worth noting that no any desorption was observed in all cases. The adsorption process was, therefore, irreversible, i.e., the interaction between the particles and the antigen or antibody molecules were strong. Detection of Immunoagglutinated Latex Particles. The unstable conjugated latex or latex agglutination in buffer medium containing specific antibody took place when bridging flocculation was induced by specific interaction between antigen-containing particles and free-targeted antibody. Immunoagglutinated particles might be large enough for reducing the intensity of transmitted light. Immunoprecipitin curves, or the plots of the absorbance change of PAgC (Γads ) 2.8 mg/m2) mixed with various concentrations of malaria naive control and P. falciparum-infected patient’s plasma, for 300 s are reported in Figure 3. The typical bell-shaped curves were clearly observed in Figure 3, i.e., the absorbance change initially increased with increasing plasma concentration, passed through a maximum, and then decreased with excess plasma concentration.20,21 The explanation was based on the Heidelberger-Kendall curve, which divided the immunoprecipitin curves of malaria antigen, interacted with antibody, into three zones,22 i.e., antigen excess zone, equivalent zone, and antibody excess zone. When the number of antibody in individual plasma was equivalent to the amount of antigen adsorbed onto the latex surface, the antibody and antigen effectively bound resulting in the large size of the immune (20) Stoll, S.; Lanet, V.; Pefferkorn, E. Colloid Interface Sci. 1993, 157, 302311. (21) Ouali, L.; Stoll, S.; Pefferkorn, E.; Elaissari, A.; Lanet, V.; Pichot, C.; Mandrand, B. Polym. Adv. Technol. 1995, 6, 541-546. (22) Steward, M.; Male, D. Immunological Techniques. In Immunology; Roitt, I.; Brostoff, J.; Male, D., Eds; 6th ed.; Mosby: Edinburgh; 2001.
complexes (i.e., clusters of particles) and, consequently, the maximum absorbance change. The optimum value of PAgC was reached to 4.90 µg/mL, while a detection limit of PAgC in P. falciparum plasma was 0.34 µg/mL. In the antigen and antibody excess zones, the size of immunoagglutinated particles remained small and no distinct absorbance change could be detected. A large difference of absorbance change between infected and naive control was observed, which indicated a high immunoreactivity of the conjugated particles. The clusters of immune complexes at the optimum condition of PAgC investigated under OM are reported in Figure 3b. The agglutinated sensitive particles, Figure 3bII, having an average size close to 1 mm were noticed when the PAgC was incubated with P. falciparum-infected plasma, whereas the size and quantity of aggregates in the naive control could not be greatly detected, Figure 3bI. Field Study. The latex agglutination test of PAgC, PmAbC, and PpAbC observed via naked eyes was performed in the field. The results of the test are summarized in Tables 1 and 2. In Table 1, among 19 total cases of P. falciparum-infected plasma, 16 cases were considered as positive due to their agglutination within 2 min. The remaining three negative cases did not agglutinate with parasitemia of 5000, 15 000, and 90 000 parasites/µL in this period. When compared to the microscopic study, our results showed the sensitivity of PAgC of 84% which was similar to that of the latex agglutination detected by laser light scattering (88%) and malaria detecting immunosensor (86%) as previously reported.23,24 From 16 positive cases of P. falciparum-infected plasma, the degree of agglutination could be classified into three levels, i.e., 1+, mild agglutination within 1 to 2 min or fine clumping when carefully observed; 2+, strong agglutination within 30 s to 1 min or big clumps; and 3+, very strong agglutination within 30 s or very big clumps. It was noticed that the numbers of positive cases of P. falciparum-infected plasma at the degree of agglutination of 1+, 2+, and 3+ were 2, 8, and 6, respectively. Spearman rank (23) Mya, M. M.; Roy, A.; Saxena, R. K.; Roy, K. B. Jpn. J. Infect. Dis. 2002, 55, 150-156. (24) Mya, M. M.; Saxena, R. K.; Roy, A.; Roy, K. B. Parasitol. Res. 2003, 89, 371-374.
Analytical Chemistry, Vol. 79, No. 12, June 15, 2007
4693
Table 2. Agglutination of the PmAbC and PpAbC Tested with P. falciparum- and P. vivax-Infected Plasma for Sensitivity and Species Specificity with Comparison to Malaria Naive Plasma PmAbC
PpAbC
degree of agglutination no. of positive cases (% cases)a
degree of agglutination no. of positive cases (% cases)a
no. of cases
no. parasites/µL mean ( SD (ranges)
no. of positive cases (%)a
1+
2+
3+
no. of positive cases (%)a
P. falciparum
20
68 750 ( 79 205 (5000-355 000)
18 (90.0)
1 (5.5)
5 (27.8)
12 (66.7)
P. vivax
10
17 250 ( 6917 (2500-25 000)
10 (100.0)
0
1 (10.0)
malaria naive
10
0
2 (20.0)
1 (50.0)
1 (50.0)
a
1+
2+
3+
18 (90.0)
0
6 (33.3)
12 (66.7)
9 (90.0)
10 (100.0)
0
0
10 (100.0)
0
2 (20.0)
0
2 (100.0)
0
See footnote of Table 1.
correlation analysis was then applied for determination of the relationship between degree of agglutination and parasitemia. The results showed the positive correlation but not statistical significance (Spearman rank correlation coefficient (R) ) 0.339, significant value (P) > 0.05). On the contrary, ELISA, performed by using the same P. falciparum plasma, exhibited negative correlation (R ) -0.337, P > 0.05) between parasitemia and OD value. The latter was in accordance with the principle stated that, during P. falciparum infection, the parasite-specific immune response was suppressed.25 It could be assumed that the higher degree of agglutination with the higher parasitemia was induced by residual antibody circulating in blood stream. Unexpectedly, the false-positive results were observed in all 10 P. vivax plasma, of which the degree of agglutination of 1+, 2+, and 3+ were 1, 4, and 5, respectively. The self-aggregation of the particles might be neglected because the PAgC well dispersed after mixing with medium. It was presumed that the nonspecific binding between P. falciparum antigen and anti-P. vivax antibody might occur in the latex agglutination test in spite of the absence of cross-reaction between malaria P. falciparum antigen used and anti-P. vivax antibody in ELISA. This might be due to the dissimilar surface hydrophilic, i.e., antigen’s conformation attached on the PS immunoplate and the carboxylated particle surfaces were probably difference.26,27 It was also indicated in Table 1 that 3 of 10 malaria naive controls gave false-positive results at a degree of agglutination of 1+, possibly due to the antibody against other febrile illness, if presented, in the control bound nonspecifically to malaria antigen.28 Moreover, it was found that specificity to malaria disease of PAgC was 70%. Table 2 showed that the agglutination results obtained from the PmAbC and PpAbC were identical (sensitivity 90%) although the adsorbed amount of monoclonal antibody (Γads 2.2 mg/m2) onto the particles was lower than that of polyclonal antibody (Γads
3.2 mg/m2). This indicated that 100% of the adsorbed monoclonal antibody molecules onto the latex particle were immunologically active and were able to totally link with the antigen, compared to those rendered by polyclonal complexes.29 Among 20 samples, 18 of P. falciparum-infected plasma cases were positive and the rest with parasitemia of 15 000 and 90 000 parasites/µL were negative. The two false-negative results were detected similar to the data recorded for commercial dipstick assay (ICT Malaria P.f. and ParaSight) based on the detection of P. falciparum HRP II.2,30 It was also reported that the number of false-negative results of the commercial dipstick test increased with decreasing level of parasitemia.2,5,30 However, the false-negative interpretation in sample with high-level parasitemia, e.g., higher than 100 000 parasites/µL, occasionally occurred in those commercial tests.31 As also shown in Table 2, the number of positive cases of P. falciparum-infected plasma at the degree of agglutination of 1+, 2+, and 3+ were 1, 5, and 12 samples in the PmAbC and 0, 6, and 12 cases for the PpAbC, respectively. Correlation analysis between the degree of agglutination and parasitemia was, therefore, conducted by applying Spearman rank correlation analysis. Results showed that the degree of agglutination tested by the PmAbC and PpAbC did not significantly correlate with parasitemia (R ) 0.009 and 0.097, respectively, P > 0.05). However, the positive correlation indicated that the higher parasitemia, i.e., the more antigen released, was efficient to rapidly form larger aggregate leading to the higher degree of agglutination. Similar to the case of the PAgC, the positive results of PmAbC and PpAbC with all 10 P. vivax infections were of concern. It might be explained based on the previous work which reported that HSP70 antigen was expressed in P. vivax.32 This was possibly responsible for the false-positive result of the PmAbC by the P. vivax sample. The heterogeneous composition of polyclonal antibody, consisting of malaria and nonmalaria antibodies, prob-
(25) Good, M. F.; Xu, H.; Wykes, M.; Engwerda, C. R. Annu. Rev. Immunol. 2005, 23, 69-99. (26) Esser, P. Principles in Adsorption to Polystyrene; Nunc Bulletin No. 6; Nunc Laboratories, 1988. (27) Okubo, M.; Yamamoto, Y.; Uno, M.; Kamei, S.; Matsumoto, T. Colloid Polym. Sci. 1987, 265, 1061-1066. (28) Funk, M.; Schlagenhauf, P.; Tschopp, A.; Steffen, R. Trans. R. Soc. Trop. Med. Hyg. 1999, 93, 268-272.
(29) Ortega-Vinuesa, J. L.; Bastos-Gonzalez, D. J. Biomater. Sci., Polym. Ed. 2001, 12, 379-408. (30) Makler, M. T.; Palmer, C. J.; Ager, A. L. Ann. Trop. Med. Parasitol. 1998, 92, 419-433. (31) Stow, N. W.; Torrens, J. K.; Walker, J. Trans. R. Soc. Trop. Med. Hyg. 1999, 93, 519-520. (32) Bianco, A. E.; Crewther, P. E.; Coppel, R. L.; Stahl, H. D.; Kemp, D. J.; Anders, R. F.; Brown, G. V. Am. J. Trop. Med. Hyg. 1988, 38, 258-267.
4694
Analytical Chemistry, Vol. 79, No. 12, June 15, 2007
ably caused the nonspecific aggregation of PpAbC with P. vivax antigen. In addition, 2 of 10 naive control showing the false-positive test might be caused by the delayed diagnosis and unsuitable treatment of other important causes of fever such as typhoid or dengue.1,28 From this data, the specificity to malaria disease of PmAbC and PpAbC was 80%. However, the correlation between the clinical history of patients, i.e., age and fever, and degree of agglutination for each conjugated particles, was not significant. To gain better interpretation, the number of sample should be increased. CONCLUSIONS Particles conjugated with malaria antigen or antibody to P. falciparum malaria (monoclonal IgG antibody to HSP70 and polyclonal IgG antibody to P. falciparum malaria) were successfully prepared via adsorption process. The adsorption affinity of the antibodies toward the latex particles was lower than that of the malaria antigen. The behavior observed has been discussed in terms of the electrostatic interaction. The agglutination of PAgC in the presence of specific antibodies measured by monitoring the turbidity change was sensitive to the concentration of P. falciparum-infected plasma of 0.34 µg/mL. In the field, the latex agglutination test was simple to perform and the agglutination results were obtained within only 2 min. PAgC tested with P. falciparum-infected plasma exhibited a sensitivity of 84%, while the specificity for malaria disease was 70%. Nonetheless, while not useful as a diagnostic tool for malaria disease due to persistence of antibody for a year after parasite clearance, the antibody detection may play an important role in epidemiologic study of the disease or in screening blood donors.
In the case of antigen detection, which provided a better meaning of diagnosis, PmAbC and PpAbC gave a higher sensitivity and specificity to malaria disease of 90% and 80% compared to those of PAgC. To further understanding of the correlation between degree of agglutination and clinical history of patients, we are currently performing the mass screening test of the antibody-coated particles. Even though the limitation of the latex agglutination test in our study was the lack of reliable discrimination between P. falciparum and P. vivax infections, the results herein lightened up an expectation that the synthesized latex particles could be applied for malaria diagnosis in the near future. Additionally, due to the fact that the majority of malaria cases are found in countries where cost-effectiveness is an important factor, the latex agglutination test, which is a cheap diagnostic test, is likely to be of great benefit in those areas. ACKNOWLEDGMENT The Research Grant from The Thailand Research Fund (TRF) to P.T. and the scholarships to D.P. under the TRF Royal Golden Jubilee Ph.D. Program and French Embassy in Thailand are gratefully acknowledged. The authors thank the BioMe´rieux Company for research support through the chemicals and equipment, and the staff of Malaria Clinic, Mae Sod, Tak province, Thailand for collection of plasma sample. The authors also thank Dr. F. Zavala and Dr. J. Sacci for providing the hybridoma cells and Ms. P. Chootong for monoclonal antibody production. Received for review March 12, 2007. Accepted April 16, 2007. AC070502W
Analytical Chemistry, Vol. 79, No. 12, June 15, 2007
4695
