- Email: [email protected]
Plasmodium falciparum:Naturally Occurring Rabbit Immunoglobulins Recognize Human Band 3 Peptide Motifs and Malaria-Infected Red Cells
JOBNAME: V82#1 96 PAGE: 1 SESS: 8 OUTPUT: Wed Jun 12 10:28:40 1996 /xypage/worksmart/tsp000/0752/20pu
EXPERIMENTAL PARASITOLOGY ARTICLE NO. 0006
82, 45–53 (1996)
Plasmodium falciparum: Naturally Occurring Rabbit Immunoglobulins Recognize Human Band 3 Peptide Motifs and Malaria-Infected Red Cells IAN CRANDALL, NEIL GUTHRIE,
AND IRWIN
W. SHERMAN
Department of Biology, University of California, Riverside, California 92521 U.S.A. CRANDALL, I., GUTHRIE, N., AND SHERMAN, I. W. 1996. Plasmodium falciparum: Naturally occurring rabbit immunoglobulins recognize human band 3 peptide motifs and malaria-infected red cells. Experimental Parasitology 82, 45–53 The reactivity to overlapping decapeptides based on sequences of human band 3 protein was determined for 24 rabbit serum samples. A high percentage of the sera recognized sequences of amino acid 650–670, as well as several other sequences representing putative exofacial regions of the band 3 protein. When sera were used to stain immunofluorescently human erythrocytes, some of which were infected with P. falciparum, those samples that predominantly reacted with amino acids 650–670 stained P. falciparum-infected red cells, whereas those that reacted with other regions of band 3 stained all erythrocytes. A positive correlation was found between anti-band 3 reactivity and the capacity of a serum to inhibit the cytoadherence of P. falciparum-infected erythrocytes to C32 amelanotic melanoma cells. Also, some rabbit sera immunoprecipitated a high-molecular-weight protein from extracts of surface iodinated P. falciparuminfected red cells © 1996 Academic Press, Inc.
identified as two sequences of amino acids in the human band 3 protein: HPLQKTY (residues 546–553) and YVKRVK (residues 824–829) (Crandall et al. 1993). The production of rabbit polyclonal antisera directed against human band 3 amino acid residues 546–553 and 824–829 has previously been reported (Crandall and Sherman 1994b). Subsequent attempts to determine which residues in these amino acid blocks were antigenic by the epitope mapping PEPSCAN method yielded an unexpected result: rabbit sera, prior to immunization, contained immunoglobulins directed against distinct regions of human band 3. In this report we have analyzed the time-dependent appearance of anti-band 3 responses in the original samples and have examined 20 additional rabbit sera to determine whether natural anti-band 3 responses were present.
INTRODUCTION The development of a practical and effective protective vaccine for human malaria—a disease that affects more than 200 million people and kills over 2 million annually (WHO Bulletin)—requires that there be some description of the antigens of the parasite and the infected cell. Approaches to antigenic analysis have involved antibodies either from infections in, or vaccination of, natural hosts such as humans and New World monkeys or sera from laboratory animals, such as mice, rats, and rabbits, that cannot be infected but have useful immunologic properties (Kwiatkowski 1994). We have studied the antigens present on the Plasmodium falciparum-infected erythrocyte using murine monoclonal antibodies and found several parasite-induced, host-derived antigens that are involved in the binding of the infected red cell to the endothelium (Winograd and Sherman 1989; Crandall and Sherman 1994 a, 1994b). The epitopes of some of these adhesion-related antigens (which are in an inactive conformation in uninfected red cells but are exposed with the intraerythrocytic development of the parasite (Guthrie et al. 1995)) have been
METHODS
AND
MATERIALS
PEPSCANs. Predicted exofacial regions of band 3 protein (Table I) were based on published data (summarized in Reithmeier 1993) and used as the pattern for overlapping decapeptides. All peptides were patterned on the human band 3 amino acid sequences reported by Lux et al. (1989) and Tanner et al. (1988). Epitope mapping of putative exo45 0014-4894/96 $18.00 Copyright © 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
JOBNAME: V82#1 96 PAGE: 2 SESS: 8 OUTPUT: Wed Jun 12 10:28:40 1996 /xypage/worksmart/tsp000/0752/20pu
46
CRANDALL, GUTHRIE, AND SHERMAN
facial regions was performed with decapeptides constructed with an offset of two amino acids using noncleavable peptides with acetylated N terminii synthesized on pins and referred to as PEPSCANs (Chiron Mimotope (Australia)). A single block of pins was used for all determinations with a 1:500 dilution of rabbit serum and a goat anti-rabbit IgG– alkaline phosphatase second antibody (Bio-Rad, Richmond, CA) used at 1:500 dilution. Periodic checks indicated that minimal changes in the reactivity of the peptide block were occurring during the assays. The epitope of an anti-P. falciparum-infected red cell murine monoclonal antibody, 4A3 (Winograd and Sherman 1989), was determined using a 1:400 dilution of ascites fluid and a goat anti-mouse IgG– alkaline phosphatase second antibody. Plates were developed in 0.5 mg/ml nitrophenol phosphate, 150 mM Tris base, 10 mM MgCl2 for 1 hr before being read in an ELISA plate reader (Model 450, Bio-Rad) at 405 nm with a reference filter of 655 nm. Results are expressed as the observed optical density value subtracted from the average absorbance of two wells that had the peptide bearing pins removed. Samples. Two groups of samples of sera were examined. Group 1 contained sera from four rabbits that had been used for anti-synthetic peptide polyclonal antibody production (Crandall and Sherman 1994b). Rabbit polyclonal sera against human band 3 amino acids 546–555 and 821–834 were produced commercially by immunizing pairs of rabbits with MAP octamers (Tam 1988) of the peptides (Research Genetics, Huntsville, AL). Each rabbit produced four samples (0, 4, 8, and 10 weeks) for a total of 16 samples. Group 2 consisted of 1-ml serum samples from 20 rabbits that had not received any immunizations (’preimmunes’) and were purchased from Research Genetics. Immunofluorescence. Immunofluorescent staining was carried out as described in Crandall and Sherman (1994a) using a 1:25 dilution of rabbit serum except that biotinylated goat anti-rabbit immunoglobulin (Calbiochem, La Jolla, CA) was used as the secondary antibody.
Parasites. The Gambian FCR-3 strain of P. falciparum was cultured according to Trager and Jensen (1976). Cultures were synchronized at the ring stage by sorbitol lysis of mature forms (Lambros and Vanderberg 1980). Cytoadherence assays. Cytoadherence assays were carried out as described previously (Udeinya et al. 1981; Crandall, et al., 1991) except that BTS buffer (50 mM Bis-Tris, 130 mM NaCl, pH 6.6) was substituted for BTC buffer (i.e., the Ca2+ was omitted from the cytoadherence assay buffer). Forty microliters of undiluted preimmune rabbit serum was added per 5 ml of incubation buffer. Four ’blanks’ (no rabbit sera added) were assayed at the same time. Immunoprecipitation of iodinated proteins. Human erythrocytes infected with the FCR-3 line of P. falciparum and uninfected erythrocytes were surface-iodinated in the presence of IODOBEADs (Pierce), using the technique of Markwell (1982). Immunoprecipitation was by the method of Howard et al. (1988) using 20 ml of rabbit serum per 1–2 ml. The precipitates were resuspended in sample buffer, loaded, and electrophoresed on a 7.5% polyacrylamide gel. Separated proteins were transferred to nitrocellulose membrane (0.45-mm pore size, MSI) following the method of Towbin et al. (1974). The dried membrane was exposed to X-ray film (Fuji) for 1 week prior to development.
RESULTS The sera from four rabbits that had been used for anti-synthetic peptide polyclonal antibody production against human band 3 amino acids 546–555 were used for PEPSCAN analysis (Fig. 1, columns a and b) or 821–834 (Fig. 1, columns c and d). One of the PEPSCAN series (column a) produced the result expected for an immunization of an animal with a synthetic peptide; that is, immunization of the rabbit with a
JOBNAME: V82#1 96 PAGE: 3 SESS: 8 OUTPUT: Wed Jun 12 10:28:40 1996 /xypage/worksmart/tsp000/0752/20pu
NATURALLY OCCURRING IMMUNOGLOBULINS
47
FIG. 1. PEPSCANs of four rabbits at four times. Polyclonal rabbit sera collected from four rabbits (columns 1–4) at four times (0 weeks (row 1), 4 weeks (row 2), 8 weeks (row 3), and 12 weeks (row 4)) were used for PEPSCAN analysis of decapeptides based on putative human band 3 exofacial regions (see Table I). The location of the synthetic peptide used to cahllenge the animal (peptide was introduced as a MAP octamer) is marked with a broken line.
MAP octamer consisting of human residues 546–555 resulted in an immune response to this region as well as several ’spikes’ in other parts of the PEPSCAN. A second animal (Fig. 1, column b) treated in the same way developed a weak response to the immunogen. However, a PEPSCAN analysis revealed that the animal was already producing a strong response (even in the preimmune) for residue block 650–670. The response to residues 650–670 diminished between 0 and 10 weeks. Rabbits immunized with human band 3 amino acids 821–834 also displayed unexpected results. One rabbit (Fig. 1, column c) prior to immunization had a minor response against residues 650–670 and a major response against residues 821–834. Immunization of this animal with a MAP octamer consisting of amino acids 821–834 resulted in the loss of response to this
region. A second rabbit (Fig. 1, column d) was immunized with the same MAP octamer and this rabbit produced an immune response to the synthetic peptide, several ’spikes’ in the region of external loop 4, and a significant response to amino acid residues 540–557. That rabbits, prior to immunization, could produce strong anti-band 3 reactions was unexpected. To determine whether reactivity against discrete regions of the human band 3 protein (and presumably against the rabbit sequence as well) was a common occurrence, sera from a larger group of rabbits were examined. The results obtained using this group are shown in Table II.{tabII} The majority (17/20) of the rabbit serum samples produced an anti-band 3 response (i.e., greater than twice background levels for at least three overlapping sequences) against specific segments of the protein. Thir-
JOBNAME: V82#1 96 PAGE: 4 SESS: 8 OUTPUT: Wed Jun 12 10:28:40 1996 /xypage/worksmart/tsp000/0752/20pu
48
CRANDALL, GUTHRIE, AND SHERMAN
teen of the 20 samples had relatively simple reactivity profiles with the intensity of a single region being at least twice that seen for the background. In most cases the strongest response was to amino acids 650–677. An example, RG 8836, is shown in Fig. 2a. Two other samples, RG 8821 and RG 8824, had a strong reaction to the amino acid block 624–639 (Figs. 2b and 2c). Immunofluorescent staining of samples from P. falciparum cultures using the rabbit polyclonal sera indicated that 17 of the 20 samples preferentially recognized P. falciparuminfected erythrocytes (Fig. 3a). However, three samples (8778, 8821, and 8824) recognized all erythrocytes (Fig. 3b). This suggests that in most cases (17/20) the human band 3 peptide motifs being recognized were in conformations absent from uninfected erythrocytes, but present in P. falciparum-infected erythrocytes. The
peptide motifs (amino acids 650–677) recognized by the majority of the rabbit sera corresponded to the epitope of the previously described murine monoclonal antibody, 4A3 (Winograd and Sherman 1989; Crandall et al. 1995), which specifically recognized infected erythrocytes and inhibited CD36-mediated cytoadherence of P. falciparum-infected erythrocytes to target cells (Winograd and Sherman 1989; Crandall et al. 1994). To determine whether rabbit sera had the same anti-adherent properties, each sample of rabbit serum was tested in a cytoadherence assay. Not only was there a relationship between anti-band 3 reactivity and cytoadherence inhibition (Fig. 4a), the reactivity with amino acids 650–677 was positively correlated with inhibitory activity (Fig. 4b). Polyclonal sera raised against P. falciparuminfected erythrocytes have been used to identify
JOBNAME: V82#1 96 PAGE: 5 SESS: 8 OUTPUT: Wed Jun 12 10:28:40 1996 /xypage/worksmart/tsp000/0752/20pu
NATURALLY OCCURRING IMMUNOGLOBULINS
49
DISCUSSION
FIG. 2. Sample PEPSCANs of sera from preimmune rabbits: (a) sample RG 8836; (b and c) samples 8821 and 8824.
potential adhesins on malaria-infected red cells. For example, the parasite-encoded, highmolecular-weight, Triton X-100-insoluble P. falciparum erythrocyte membrane protein 1 (PfEMP 1) was immunoprecipitated from radiolabeled infected red cells using rabbit, monkey, and human sera (van Schravendijk et al. 1991; Day et al. 1993; Howard et al. 1988) and sequestrin, the PfEMP 1-like adhesin, was identified in radiolabeled red cells by an anti-idiotype rabbit polyclonal serum (Ockenhouse et al. 1992). Our findings indicated that most rabbit sera produced what are effectively anti-P. falciparum-infected erythrocyte responses; therefore, we assayed rabbit sera for evidence of anti-PfEMP 1 activity as well. Four samples (RG8820, 8821, 8834, and 8839) immunoprecipitated a high-molecular-weight protein (PfEMP1?) from extracts of infected erythrocytes, but not from uninfected samples (Fig. 5).
The use of a vertebrate to produce a serum to be used as an antigen-specific reagent is based on the assumption that the animal is not already producing an immune response to the antigen in question and that conserved proteins are ’silent.’ The results of the PEPSCAN profiles obtained with sera from preimmunized rabbits indicate that the majority of rabbits produce immunoglobulins that recognize (albeit to differing degrees) human band 3 motifs and that the epitopes recognized by these immunoglobulins vary not only from animal to animal, but can and do change with time in a single animal. Therefore the assumption that ’preimmune’ rabbits do not, or cannot, produce anti-band 3 antibodies is invalid. A priori it would not be expected that rabbits that have never been exposed to human band 3 would have the capacity to produce a strong reaction to a human protein. However, the evidence presented here shows that such responses do occur. Due to the highly conserved nature of the band 3 protein it is likely that the response is directed against rabbit rather than human band 3. The amino acid sequence of rabbit band 3 has not been reported; however, the published sequences of band 3 from other species show a high degree of homology in the membrane spanning region of band 3 (Reithmeier 1993). Therefore, that human and rabbit bands 3 share a high degree of homology in the regions of human band 3 protein recognized by rabbit immunoglobulins is not entirely unexpected. The reason for rabbit sera containing antibodies that react with human (and presumably rabbit) band 3 peptide motifs is unknown; however, such responses appear to be widespread and are not restricted to single peptide motifs. The rabbits used as sample sources were healthy and were chosen at random; therefore, it appears that the presence of anti-band 3 immunoglobulins does not result in pathologic consequences. It should be noted that PEPSCANs detect the presence of antibodies that recognize a particular peptide sequence or sequences, but this
JOBNAME: V82#1 96 PAGE: 6 SESS: 8 OUTPUT: Wed Jun 12 10:28:40 1996 /xypage/worksmart/tsp000/0752/20pu
50
CRANDALL, GUTHRIE, AND SHERMAN
FIG. 3. Immunofluorescent staining of P. falciparum-infected human erythrocytes: (a) staining produced by 1:25 dilution of RG 8836; and (b) staining produced by a 1:25 dilution of RG 8824. (Left) A field observed under white light (phase contrast); (right) the same field observed using uv light with filters. Infected erythrocytes are indicated by black arrows.
method gives little or no information about the conformation that the peptide assumes when the antibody binds to it. This distinction is important in the case of a host protein since the strong reactivity observed in some of the samples may be directed against conformations of the band 3 peptides that do not normally occur in erythrocytes or these epitopes may be cryptic. However, after invasion and with intra-erythrocytic development of the malaria parasite there is exposure of these normally cryptic regions, presumably due to changes in the conformation of the band 3 protein (Crandall and Sherman 1994a, 1994b). It is of interest that the majority
of the rabbit samples displayed a stronger immunofluorescence with P. falciparum-infected erythrocytes than with uninfected erythrocytes; this suggests that the anti-band 3 responses present in the rabbit sera are directed against an altered conformation of the band 3 protein. It is possible that the anti-P. falciparuminfected erythrocyte response present in the majority of rabbit sera could be due to another (non-band 3) component of the polyclonal sera; however, this seems unlikely. The murine monoclonal antibody 4A3 shares a common epitope (650–677) with most rabbit serum samples (Crandall et al. 1995), has a similar
JOBNAME: V82#1 96 PAGE: 7 SESS: 8 OUTPUT: Wed Jun 12 10:28:40 1996 /xypage/worksmart/tsp000/0752/20pu
NATURALLY OCCURRING IMMUNOGLOBULINS
51
FIG. 4. Relationship between anti-band 3 reactivity and cytoadherence inhibition: the relative reactivity of either (a) the strongest reacting peptide residue in the scan (maximum observed value minus background) or (b) residues 656–668 (maximum observed value minus background) was plotted against adhesion (infected red cells/melanoma cell) in the presence of 40 ml of rabbit sera (1:125 dilution). Data points corresponding to no serum addition &hecirc;(&hb9;), and serum with negative (L), and positive (l) reactivity to band 3 peptides are indicated. The solid line indicates the best fit regression and the broken lines the 95% confidence intervals for the fit as determined by the computer program Sigma Plot (Jandel Scientific). The data points in (a) have an r2 value of 0.36 and an x coefficient of −2.8, whereas the r2 value for (b) is 0.51 with an x coefficient of −4.7. The higher r2 value and decreased x coefficient in (b) suggest that reactivity with specific regions of band 3 can be more effective in inhibiting cytoadherence.
immunofluorescent staining pattern, and has similar cytoadherence inhibiting properties (Winograd and Sherman 1989; Crandall and Sherman 1994a). Consistent with this is the positive correlation between the observed response against the amino acid sequence 650– 677 and the degree of inhibition of cytoadherence (Fig. 5b). Although the epitopes of MAb 4A3 and many of the rabbit samples are located farther along the amino acid sequence (650– 667) than the adhesin (546–553) (Crandall et al. 1994), it is possible that these regions are close enough in the tertiary structure of the adhesive version of the protein to affect the interaction of residues 546–553 and its receptor. The PEPSCANs obtained from group 2 (the 20 preimmunized rabbit samples) represents reactivities for animals that had not been challenged with a band 3 based immunogen. Sample group 1 contained rabbits that were challenged with band 3 peptide motifs and in one of the challenged animals (Fig. 1, column a) the animal appeared to be truly preimmune and the expected immune response was achieved. In two other animals immunization with a synthetic human band 3 peptide resulted in the loss
of a preexisting strong anti-band 3 responses (columns b and c) with the final sample containing low anti-band 3 reactivity. One of these animals is interesting because the PEPSCAN profile of the preimmune bleed indicates that the animal had a strong and specific response to the very peptide that was used to immunize the animal. Immunization of this animal resulted in the loss of reactivity to this sequence. The underlying reason for the loss of anti-820–834 upon challenge with the same sequence is unknown. The animal corresponding to column d was truly preimmune and generated the desired response to the immunogen; however, it also produced a strong response to another region (540–557) which coincidentally contains the other adhesive region of band 3 protein. It is not known whether this response reflects crossreactivity between two antigenically similar regions (a hypothesis that is not supported by similar amino acid sequences, but the ability of peptides from both regions to inhibit P. falciparum adherence (Crandall et al. 1994) suggests that there is some similarity) or whether this is the result of interplay between the introduced synthetic peptide immunogen, the host’s im-
JOBNAME: V82#1 96 PAGE: 8 SESS: 8 OUTPUT: Wed Jun 12 10:28:40 1996 /xypage/worksmart/tsp000/0752/20pu
52
CRANDALL, GUTHRIE, AND SHERMAN
FIG. 5. Immunoprecipitation of iodinated P. falciparuminfected blood using rabbit polyclonal sera: polyclonal sera samples used in the immunoprecipitations were (lane, sample): A1, no primary (rabbit) antibody added; A2, RG 8778; A3, RG 8820; A4, RG 8821; A5, RG 8822; A6, RG 8823; A7, RG 8824; A8, RG 8825; B1, RG 8826; B2, RG 8827; B3, RG 8828; B4, RG 8829; B5, RG 8830; B6, RG 8831; B7, RG 8832; B8, RG 8833; C1, RG 8834; C2, RG 8835; C3, RG 8836; C4, RG 8837; and C5, RG 8839. The distances to which prestained molecular weight standards (not shown) migrated through the 7.5% gel are indicated on the left.
mune system, and homologous regions of the host’s band 3 protein. The parasite encoded protein PfEMP 1 (Baruch et al. 1995) has been reported to be a potential ligand for adhesion of P. falciparuminfected erythrocytes. Typically, PfEMP 1 is identified in extracts of infected cells as a surface-iodinatable, Triton X-100-insoluble protein immunoprecipitated by strain-specific sera (reviewed by Howard et al. 1990). It is of interest that PfEMP 1, or a material with similar properties (molecular weight of approximately >200,000), is susceptible to exposure to 0.1%
trypsin for 5 min at room temperature (N.G., data not shown)) and can be immunoprecipitated from extracts of surface-iodinated infected cells (but not uninfected cell extracts) using four of the serm samples (RG 8820, RG 8821, RG 8834, and RG 8839) obtained from preimmune animals. Examination of the corresponding PEPSCAN profiles (based on 10 mers of putative exofacial regions), however, did not indicate a single common feature that was present in the four precipitating sera but absent from the 16 nonprecipitating samples. Therefore, it appears that if precipitation of PfEMP 1 was due to a cross–reaction with a band 3-related response then some other factor (such as conformational specificity) must have been involved. Notwithstanding our inability to accurately describe the factors that result in antigen precipitation it is noteworthy that in some instances immunoprecipitating sera may not result from exposure of the animal to the immunizing antigen. Because preimmune rabbit sera can have reactivity with an antigen, and because such reactivity can change (diminish) after immunization, the routine use of preimmune rabbit serum as a negative control may lead to erroneous conclusions as to antigen specificity. ACKNOWLEDGMENTS We thank Jacques Prudhomme for maintaining the parasite cultures and preparing the target cells. This research was supported in part by grants from the UNDP/World Bank/WHO Special Programme for Research and Training in Tropical Diseases and from the National Institute of Allergy and Infectious Diseases (National Institutes of Health).
REFERENCES BARUCH, D. I., PASLOSKE, B. L., SINGH, H. B., BI, X., MA, X. C., FELDMAN, M., TARASCHI, T. F., AND HOWARD, R. J. 1995. Cloning the Plasmodium falciparum gene encoding Pf EMP-1, a malarial variant antigen and adherence receptor on the surface of parasitized human erythrocytes. Cell 82, 77–87. CRANDALL, I., SMITH, H., AND SHERMAN, I. W. 1991. Plasmodium falciparum: The effect of pH and Ca2+ concentration on the in vitro cytoadherence of infected erythrocytes to a melanotic melanoma cells. Experimental Parasitology 73, 362–368. CRANDALL, I. E., AND SHERMAN, I. W. 1994a. Cytoadherence-related neoantigens on Plasmodium falciparum (hu-
JOBNAME: V82#1 96 PAGE: 9 SESS: 8 OUTPUT: Wed Jun 12 10:28:40 1996 /xypage/worksmart/tsp000/0752/20pu
NATURALLY OCCURRING IMMUNOGLOBULINS man malaria)-infected human erythrocytes result from the exposure of normally cryptic regions of the band 3 protein. Parasitology 108, 257–267. CRANDALL, I. E., AND SHERMAN, I. W. 1994b. Antibodies to synthetic peptides based on band 3 motifs react specifically with Plasmodium falciparum (human malaria)infected erythrocytes and block cytoadherence. Parasitology 108, 389–396. CRANDALL, I. E., GUTHRIE, N., AND SHERMAN, I. W. 1995.. Sera from individuals resident in The Gambia contain immunoglobulins that recognize human Band 3 peptide motifs. American Society of Tropical Medicine and Hygiene 52(5), 450–455. CRANDALL, I., LAND, K. M., AND SHERMAN, I. W. 1994. Plasmodium falciparum: Pfalhesin and CD36 form an adhesion/receptor pair that is responsible for the pH dependent portion of cytoadherence/sequestration. Experimental Parasitology 78, 203–209. CRANDALL, I., COLLINS, W. E., GYSIN, J., AND SHERMAN, I. W. 1993. Synthetic peptides based on motifs present in human band 3 protein inhibit cytoadherence/ sequestration of the malaria parasite Plasmodium falciparum. Proceedings of the National Academy of Sciences USA 90, 4703–4707. DAY, K. P., KARAMALIS, F., THOMPSON, J., BARNES, D. A., PETERSON, C., BROWN, H., AND KEMP, D. J. 1993. Genes necessary for expression of a virulence determinant and for transmission of Plasmodium falciparum are located on a 0.3 megabase region of chromosome 9. Proceedings of the National Academy of Sciences USA 90, 8292–8296. GEYSEN, H. M., MELOEN, R. H., AND BARTELING, S. J. 1984. Use of peptide synthesis to probe viral antigens for epitopes to a resolution of a single amino acid. Proceedings of the National Academy of Sciences USA 81, 3998–4002. GUTHRIE, N., CRANDALL, I. E., MARINI, S., AND SHERMAN, I. W. 1995. Monoclonal antibodies that react with human band 3 residues 542–555 recognize different conformations of this protein in uninfected and Plasmodium falciparum-infected erythrocytes. Molecular and Cellular Biochemistry 144, 117–123. HOWARD, R. J., BARNWELL, J. W., ROCK, E. P., NEEQUAYE, J., OFORI-ADJEI, D., MALOY, W. L., LYON, J. A., AND SAUL, A. 1988. Two approximately 300 kilodalton Plasmodium falciparum proteins at the surface membrane of infected erythrocytes. Molecular and Biochemical Parasitology 27, 207–224. HOWARD, R. J., HANDUNNETTI, S. M., HASLER, T., GILLADOGA, A., DE AGUIAR, J. C., PASLOSKE, B. L., MOREHEAD, K., ALBRECHT, G. R., AND VAN SCHRAVENDIJK, M. R. 1990. Surface molecules on Plasmodium falciparum-infected erythrocytes involved in adherence. American Journal of Tropical Medicine and Hygiene Suppl. 43(3), 15–29. KWAITKOWSKI, D. 1994. Prospects of an antidisease vaccine. In ’Molecular Immunological Considerations in Malaria
53
Vaccine Development’ Ed. (M. F Good and A. J. Saul, Ed). pp. 131–208. CRC Press, Boca Raton, FL. LAMBROS, C., AND VANDERBERG, J. P. 1980. Synchronization of Plasmodium falciparum erythrocytic stages in culture. Journal of Parasitology 65, 418–420. LUX, S. E., JOHN, K. M., KOPITO, R. R., AND LODISH, H. F. 1989. Cloning and characterization of band 3, the human erythrocyte anion-exchange protein (AE1). Proceedings of the National Academy of Sciences USA 86, 9089–9093. MARKWELL, M. A. K. 1982. A new solid-state reagent to iodinate proteins: Conditions for the efficient labelling of antiserum. Analytical Biochemistry 125, 427–432. OCKENHOUSE, C., KLOTZ, F. W., TANDON, N. N., AND JAMISON, G. A. 1992. Sequesrtrin, a CD36 recognition protein on Plasmodium falciparum malaria-infected erythrocytes identified by anti-idiotype antibodies. Proceedings of the National Academy of Sciences USA 88, 3175–3179. REITHMERIER, R. A. F. 1993. The erythrocyte anion transporter (band 3). Current Opinion in Structural Biology 3, 515–523. TAM, J. P. 1988. Synthetic peptide vaccine design: Synthesis and properties of a high-density multiple antigenic peptide system. Proceedings of the National Academy of Sciences USA 85, 5409–5413. TANNER, M. J. A., MARTIN, P. G., AND HIGH, S. 1988. The complete amino acid sequence of the human erythrocyte membrane anion-transport protein deduced from the cDNA sequence. Biochemical Journal 256, 703–712. TOWBIN, H., STAEHELIN, T., AND GORDON, J. 1974. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proceedings of the National Academy of Sciences USA 76, 4350–4354. TRAGER, W., AND JENSEN, J. B. 1976. Human malaria parasites in continuous culture. Science 193, 673–675. UDEINYA, I., SCHMIDT, J. A., AIKAWA, M., MILLER, L. H., AND GREEN, I. 1981. Falciparum malaria-infected erythrocytes specifically bind to cultured human endothelial cells. Science 213, 555–557. VAN SCHRAVENDIJK, M. R., ROCK, E. P., MARSH, K., ITO, Y., AIKAWA, M., NEEOUAYE, J., OFORI-ADEJEI, D., RODRIGUEZ, R., PATARROYO, M. E., AND HOWARD, R. J. 1991. Characterization and localization of Plasmodium falciparum surface antigens on infected erythrocytes from West African patients. Blood 78, 226–236. WHO Bulletin 1990. 68, 667–673. WINOGRAD, E., AND SHERMAN, I. W. 1989. Characterization of a modified red cell membrane protein expressed on erythrocytes infected with the human malaria parasite Plasmodium falciparum: Possible role as cytoadherent mediating protein. Journal of Cell Biology 108, 23–30. Received 8 September 1995; accepted 20 October 1995
EXPERIMENTAL PARASITOLOGY ARTICLE NO. 0006
82, 45–53 (1996)
Plasmodium falciparum: Naturally Occurring Rabbit Immunoglobulins Recognize Human Band 3 Peptide Motifs and Malaria-Infected Red Cells IAN CRANDALL, NEIL GUTHRIE,
AND IRWIN
W. SHERMAN
Department of Biology, University of California, Riverside, California 92521 U.S.A. CRANDALL, I., GUTHRIE, N., AND SHERMAN, I. W. 1996. Plasmodium falciparum: Naturally occurring rabbit immunoglobulins recognize human band 3 peptide motifs and malaria-infected red cells. Experimental Parasitology 82, 45–53 The reactivity to overlapping decapeptides based on sequences of human band 3 protein was determined for 24 rabbit serum samples. A high percentage of the sera recognized sequences of amino acid 650–670, as well as several other sequences representing putative exofacial regions of the band 3 protein. When sera were used to stain immunofluorescently human erythrocytes, some of which were infected with P. falciparum, those samples that predominantly reacted with amino acids 650–670 stained P. falciparum-infected red cells, whereas those that reacted with other regions of band 3 stained all erythrocytes. A positive correlation was found between anti-band 3 reactivity and the capacity of a serum to inhibit the cytoadherence of P. falciparum-infected erythrocytes to C32 amelanotic melanoma cells. Also, some rabbit sera immunoprecipitated a high-molecular-weight protein from extracts of surface iodinated P. falciparuminfected red cells © 1996 Academic Press, Inc.
identified as two sequences of amino acids in the human band 3 protein: HPLQKTY (residues 546–553) and YVKRVK (residues 824–829) (Crandall et al. 1993). The production of rabbit polyclonal antisera directed against human band 3 amino acid residues 546–553 and 824–829 has previously been reported (Crandall and Sherman 1994b). Subsequent attempts to determine which residues in these amino acid blocks were antigenic by the epitope mapping PEPSCAN method yielded an unexpected result: rabbit sera, prior to immunization, contained immunoglobulins directed against distinct regions of human band 3. In this report we have analyzed the time-dependent appearance of anti-band 3 responses in the original samples and have examined 20 additional rabbit sera to determine whether natural anti-band 3 responses were present.
INTRODUCTION The development of a practical and effective protective vaccine for human malaria—a disease that affects more than 200 million people and kills over 2 million annually (WHO Bulletin)—requires that there be some description of the antigens of the parasite and the infected cell. Approaches to antigenic analysis have involved antibodies either from infections in, or vaccination of, natural hosts such as humans and New World monkeys or sera from laboratory animals, such as mice, rats, and rabbits, that cannot be infected but have useful immunologic properties (Kwiatkowski 1994). We have studied the antigens present on the Plasmodium falciparum-infected erythrocyte using murine monoclonal antibodies and found several parasite-induced, host-derived antigens that are involved in the binding of the infected red cell to the endothelium (Winograd and Sherman 1989; Crandall and Sherman 1994 a, 1994b). The epitopes of some of these adhesion-related antigens (which are in an inactive conformation in uninfected red cells but are exposed with the intraerythrocytic development of the parasite (Guthrie et al. 1995)) have been
METHODS
AND
MATERIALS
PEPSCANs. Predicted exofacial regions of band 3 protein (Table I) were based on published data (summarized in Reithmeier 1993) and used as the pattern for overlapping decapeptides. All peptides were patterned on the human band 3 amino acid sequences reported by Lux et al. (1989) and Tanner et al. (1988). Epitope mapping of putative exo45 0014-4894/96 $18.00 Copyright © 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
JOBNAME: V82#1 96 PAGE: 2 SESS: 8 OUTPUT: Wed Jun 12 10:28:40 1996 /xypage/worksmart/tsp000/0752/20pu
46
CRANDALL, GUTHRIE, AND SHERMAN
facial regions was performed with decapeptides constructed with an offset of two amino acids using noncleavable peptides with acetylated N terminii synthesized on pins and referred to as PEPSCANs (Chiron Mimotope (Australia)). A single block of pins was used for all determinations with a 1:500 dilution of rabbit serum and a goat anti-rabbit IgG– alkaline phosphatase second antibody (Bio-Rad, Richmond, CA) used at 1:500 dilution. Periodic checks indicated that minimal changes in the reactivity of the peptide block were occurring during the assays. The epitope of an anti-P. falciparum-infected red cell murine monoclonal antibody, 4A3 (Winograd and Sherman 1989), was determined using a 1:400 dilution of ascites fluid and a goat anti-mouse IgG– alkaline phosphatase second antibody. Plates were developed in 0.5 mg/ml nitrophenol phosphate, 150 mM Tris base, 10 mM MgCl2 for 1 hr before being read in an ELISA plate reader (Model 450, Bio-Rad) at 405 nm with a reference filter of 655 nm. Results are expressed as the observed optical density value subtracted from the average absorbance of two wells that had the peptide bearing pins removed. Samples. Two groups of samples of sera were examined. Group 1 contained sera from four rabbits that had been used for anti-synthetic peptide polyclonal antibody production (Crandall and Sherman 1994b). Rabbit polyclonal sera against human band 3 amino acids 546–555 and 821–834 were produced commercially by immunizing pairs of rabbits with MAP octamers (Tam 1988) of the peptides (Research Genetics, Huntsville, AL). Each rabbit produced four samples (0, 4, 8, and 10 weeks) for a total of 16 samples. Group 2 consisted of 1-ml serum samples from 20 rabbits that had not received any immunizations (’preimmunes’) and were purchased from Research Genetics. Immunofluorescence. Immunofluorescent staining was carried out as described in Crandall and Sherman (1994a) using a 1:25 dilution of rabbit serum except that biotinylated goat anti-rabbit immunoglobulin (Calbiochem, La Jolla, CA) was used as the secondary antibody.
Parasites. The Gambian FCR-3 strain of P. falciparum was cultured according to Trager and Jensen (1976). Cultures were synchronized at the ring stage by sorbitol lysis of mature forms (Lambros and Vanderberg 1980). Cytoadherence assays. Cytoadherence assays were carried out as described previously (Udeinya et al. 1981; Crandall, et al., 1991) except that BTS buffer (50 mM Bis-Tris, 130 mM NaCl, pH 6.6) was substituted for BTC buffer (i.e., the Ca2+ was omitted from the cytoadherence assay buffer). Forty microliters of undiluted preimmune rabbit serum was added per 5 ml of incubation buffer. Four ’blanks’ (no rabbit sera added) were assayed at the same time. Immunoprecipitation of iodinated proteins. Human erythrocytes infected with the FCR-3 line of P. falciparum and uninfected erythrocytes were surface-iodinated in the presence of IODOBEADs (Pierce), using the technique of Markwell (1982). Immunoprecipitation was by the method of Howard et al. (1988) using 20 ml of rabbit serum per 1–2 ml. The precipitates were resuspended in sample buffer, loaded, and electrophoresed on a 7.5% polyacrylamide gel. Separated proteins were transferred to nitrocellulose membrane (0.45-mm pore size, MSI) following the method of Towbin et al. (1974). The dried membrane was exposed to X-ray film (Fuji) for 1 week prior to development.
RESULTS The sera from four rabbits that had been used for anti-synthetic peptide polyclonal antibody production against human band 3 amino acids 546–555 were used for PEPSCAN analysis (Fig. 1, columns a and b) or 821–834 (Fig. 1, columns c and d). One of the PEPSCAN series (column a) produced the result expected for an immunization of an animal with a synthetic peptide; that is, immunization of the rabbit with a
JOBNAME: V82#1 96 PAGE: 3 SESS: 8 OUTPUT: Wed Jun 12 10:28:40 1996 /xypage/worksmart/tsp000/0752/20pu
NATURALLY OCCURRING IMMUNOGLOBULINS
47
FIG. 1. PEPSCANs of four rabbits at four times. Polyclonal rabbit sera collected from four rabbits (columns 1–4) at four times (0 weeks (row 1), 4 weeks (row 2), 8 weeks (row 3), and 12 weeks (row 4)) were used for PEPSCAN analysis of decapeptides based on putative human band 3 exofacial regions (see Table I). The location of the synthetic peptide used to cahllenge the animal (peptide was introduced as a MAP octamer) is marked with a broken line.
MAP octamer consisting of human residues 546–555 resulted in an immune response to this region as well as several ’spikes’ in other parts of the PEPSCAN. A second animal (Fig. 1, column b) treated in the same way developed a weak response to the immunogen. However, a PEPSCAN analysis revealed that the animal was already producing a strong response (even in the preimmune) for residue block 650–670. The response to residues 650–670 diminished between 0 and 10 weeks. Rabbits immunized with human band 3 amino acids 821–834 also displayed unexpected results. One rabbit (Fig. 1, column c) prior to immunization had a minor response against residues 650–670 and a major response against residues 821–834. Immunization of this animal with a MAP octamer consisting of amino acids 821–834 resulted in the loss of response to this
region. A second rabbit (Fig. 1, column d) was immunized with the same MAP octamer and this rabbit produced an immune response to the synthetic peptide, several ’spikes’ in the region of external loop 4, and a significant response to amino acid residues 540–557. That rabbits, prior to immunization, could produce strong anti-band 3 reactions was unexpected. To determine whether reactivity against discrete regions of the human band 3 protein (and presumably against the rabbit sequence as well) was a common occurrence, sera from a larger group of rabbits were examined. The results obtained using this group are shown in Table II.{tabII} The majority (17/20) of the rabbit serum samples produced an anti-band 3 response (i.e., greater than twice background levels for at least three overlapping sequences) against specific segments of the protein. Thir-
JOBNAME: V82#1 96 PAGE: 4 SESS: 8 OUTPUT: Wed Jun 12 10:28:40 1996 /xypage/worksmart/tsp000/0752/20pu
48
CRANDALL, GUTHRIE, AND SHERMAN
teen of the 20 samples had relatively simple reactivity profiles with the intensity of a single region being at least twice that seen for the background. In most cases the strongest response was to amino acids 650–677. An example, RG 8836, is shown in Fig. 2a. Two other samples, RG 8821 and RG 8824, had a strong reaction to the amino acid block 624–639 (Figs. 2b and 2c). Immunofluorescent staining of samples from P. falciparum cultures using the rabbit polyclonal sera indicated that 17 of the 20 samples preferentially recognized P. falciparuminfected erythrocytes (Fig. 3a). However, three samples (8778, 8821, and 8824) recognized all erythrocytes (Fig. 3b). This suggests that in most cases (17/20) the human band 3 peptide motifs being recognized were in conformations absent from uninfected erythrocytes, but present in P. falciparum-infected erythrocytes. The
peptide motifs (amino acids 650–677) recognized by the majority of the rabbit sera corresponded to the epitope of the previously described murine monoclonal antibody, 4A3 (Winograd and Sherman 1989; Crandall et al. 1995), which specifically recognized infected erythrocytes and inhibited CD36-mediated cytoadherence of P. falciparum-infected erythrocytes to target cells (Winograd and Sherman 1989; Crandall et al. 1994). To determine whether rabbit sera had the same anti-adherent properties, each sample of rabbit serum was tested in a cytoadherence assay. Not only was there a relationship between anti-band 3 reactivity and cytoadherence inhibition (Fig. 4a), the reactivity with amino acids 650–677 was positively correlated with inhibitory activity (Fig. 4b). Polyclonal sera raised against P. falciparuminfected erythrocytes have been used to identify
JOBNAME: V82#1 96 PAGE: 5 SESS: 8 OUTPUT: Wed Jun 12 10:28:40 1996 /xypage/worksmart/tsp000/0752/20pu
NATURALLY OCCURRING IMMUNOGLOBULINS
49
DISCUSSION
FIG. 2. Sample PEPSCANs of sera from preimmune rabbits: (a) sample RG 8836; (b and c) samples 8821 and 8824.
potential adhesins on malaria-infected red cells. For example, the parasite-encoded, highmolecular-weight, Triton X-100-insoluble P. falciparum erythrocyte membrane protein 1 (PfEMP 1) was immunoprecipitated from radiolabeled infected red cells using rabbit, monkey, and human sera (van Schravendijk et al. 1991; Day et al. 1993; Howard et al. 1988) and sequestrin, the PfEMP 1-like adhesin, was identified in radiolabeled red cells by an anti-idiotype rabbit polyclonal serum (Ockenhouse et al. 1992). Our findings indicated that most rabbit sera produced what are effectively anti-P. falciparum-infected erythrocyte responses; therefore, we assayed rabbit sera for evidence of anti-PfEMP 1 activity as well. Four samples (RG8820, 8821, 8834, and 8839) immunoprecipitated a high-molecular-weight protein (PfEMP1?) from extracts of infected erythrocytes, but not from uninfected samples (Fig. 5).
The use of a vertebrate to produce a serum to be used as an antigen-specific reagent is based on the assumption that the animal is not already producing an immune response to the antigen in question and that conserved proteins are ’silent.’ The results of the PEPSCAN profiles obtained with sera from preimmunized rabbits indicate that the majority of rabbits produce immunoglobulins that recognize (albeit to differing degrees) human band 3 motifs and that the epitopes recognized by these immunoglobulins vary not only from animal to animal, but can and do change with time in a single animal. Therefore the assumption that ’preimmune’ rabbits do not, or cannot, produce anti-band 3 antibodies is invalid. A priori it would not be expected that rabbits that have never been exposed to human band 3 would have the capacity to produce a strong reaction to a human protein. However, the evidence presented here shows that such responses do occur. Due to the highly conserved nature of the band 3 protein it is likely that the response is directed against rabbit rather than human band 3. The amino acid sequence of rabbit band 3 has not been reported; however, the published sequences of band 3 from other species show a high degree of homology in the membrane spanning region of band 3 (Reithmeier 1993). Therefore, that human and rabbit bands 3 share a high degree of homology in the regions of human band 3 protein recognized by rabbit immunoglobulins is not entirely unexpected. The reason for rabbit sera containing antibodies that react with human (and presumably rabbit) band 3 peptide motifs is unknown; however, such responses appear to be widespread and are not restricted to single peptide motifs. The rabbits used as sample sources were healthy and were chosen at random; therefore, it appears that the presence of anti-band 3 immunoglobulins does not result in pathologic consequences. It should be noted that PEPSCANs detect the presence of antibodies that recognize a particular peptide sequence or sequences, but this
JOBNAME: V82#1 96 PAGE: 6 SESS: 8 OUTPUT: Wed Jun 12 10:28:40 1996 /xypage/worksmart/tsp000/0752/20pu
50
CRANDALL, GUTHRIE, AND SHERMAN
FIG. 3. Immunofluorescent staining of P. falciparum-infected human erythrocytes: (a) staining produced by 1:25 dilution of RG 8836; and (b) staining produced by a 1:25 dilution of RG 8824. (Left) A field observed under white light (phase contrast); (right) the same field observed using uv light with filters. Infected erythrocytes are indicated by black arrows.
method gives little or no information about the conformation that the peptide assumes when the antibody binds to it. This distinction is important in the case of a host protein since the strong reactivity observed in some of the samples may be directed against conformations of the band 3 peptides that do not normally occur in erythrocytes or these epitopes may be cryptic. However, after invasion and with intra-erythrocytic development of the malaria parasite there is exposure of these normally cryptic regions, presumably due to changes in the conformation of the band 3 protein (Crandall and Sherman 1994a, 1994b). It is of interest that the majority
of the rabbit samples displayed a stronger immunofluorescence with P. falciparum-infected erythrocytes than with uninfected erythrocytes; this suggests that the anti-band 3 responses present in the rabbit sera are directed against an altered conformation of the band 3 protein. It is possible that the anti-P. falciparuminfected erythrocyte response present in the majority of rabbit sera could be due to another (non-band 3) component of the polyclonal sera; however, this seems unlikely. The murine monoclonal antibody 4A3 shares a common epitope (650–677) with most rabbit serum samples (Crandall et al. 1995), has a similar
JOBNAME: V82#1 96 PAGE: 7 SESS: 8 OUTPUT: Wed Jun 12 10:28:40 1996 /xypage/worksmart/tsp000/0752/20pu
NATURALLY OCCURRING IMMUNOGLOBULINS
51
FIG. 4. Relationship between anti-band 3 reactivity and cytoadherence inhibition: the relative reactivity of either (a) the strongest reacting peptide residue in the scan (maximum observed value minus background) or (b) residues 656–668 (maximum observed value minus background) was plotted against adhesion (infected red cells/melanoma cell) in the presence of 40 ml of rabbit sera (1:125 dilution). Data points corresponding to no serum addition &hecirc;(&hb9;), and serum with negative (L), and positive (l) reactivity to band 3 peptides are indicated. The solid line indicates the best fit regression and the broken lines the 95% confidence intervals for the fit as determined by the computer program Sigma Plot (Jandel Scientific). The data points in (a) have an r2 value of 0.36 and an x coefficient of −2.8, whereas the r2 value for (b) is 0.51 with an x coefficient of −4.7. The higher r2 value and decreased x coefficient in (b) suggest that reactivity with specific regions of band 3 can be more effective in inhibiting cytoadherence.
immunofluorescent staining pattern, and has similar cytoadherence inhibiting properties (Winograd and Sherman 1989; Crandall and Sherman 1994a). Consistent with this is the positive correlation between the observed response against the amino acid sequence 650– 677 and the degree of inhibition of cytoadherence (Fig. 5b). Although the epitopes of MAb 4A3 and many of the rabbit samples are located farther along the amino acid sequence (650– 667) than the adhesin (546–553) (Crandall et al. 1994), it is possible that these regions are close enough in the tertiary structure of the adhesive version of the protein to affect the interaction of residues 546–553 and its receptor. The PEPSCANs obtained from group 2 (the 20 preimmunized rabbit samples) represents reactivities for animals that had not been challenged with a band 3 based immunogen. Sample group 1 contained rabbits that were challenged with band 3 peptide motifs and in one of the challenged animals (Fig. 1, column a) the animal appeared to be truly preimmune and the expected immune response was achieved. In two other animals immunization with a synthetic human band 3 peptide resulted in the loss
of a preexisting strong anti-band 3 responses (columns b and c) with the final sample containing low anti-band 3 reactivity. One of these animals is interesting because the PEPSCAN profile of the preimmune bleed indicates that the animal had a strong and specific response to the very peptide that was used to immunize the animal. Immunization of this animal resulted in the loss of reactivity to this sequence. The underlying reason for the loss of anti-820–834 upon challenge with the same sequence is unknown. The animal corresponding to column d was truly preimmune and generated the desired response to the immunogen; however, it also produced a strong response to another region (540–557) which coincidentally contains the other adhesive region of band 3 protein. It is not known whether this response reflects crossreactivity between two antigenically similar regions (a hypothesis that is not supported by similar amino acid sequences, but the ability of peptides from both regions to inhibit P. falciparum adherence (Crandall et al. 1994) suggests that there is some similarity) or whether this is the result of interplay between the introduced synthetic peptide immunogen, the host’s im-
JOBNAME: V82#1 96 PAGE: 8 SESS: 8 OUTPUT: Wed Jun 12 10:28:40 1996 /xypage/worksmart/tsp000/0752/20pu
52
CRANDALL, GUTHRIE, AND SHERMAN
FIG. 5. Immunoprecipitation of iodinated P. falciparuminfected blood using rabbit polyclonal sera: polyclonal sera samples used in the immunoprecipitations were (lane, sample): A1, no primary (rabbit) antibody added; A2, RG 8778; A3, RG 8820; A4, RG 8821; A5, RG 8822; A6, RG 8823; A7, RG 8824; A8, RG 8825; B1, RG 8826; B2, RG 8827; B3, RG 8828; B4, RG 8829; B5, RG 8830; B6, RG 8831; B7, RG 8832; B8, RG 8833; C1, RG 8834; C2, RG 8835; C3, RG 8836; C4, RG 8837; and C5, RG 8839. The distances to which prestained molecular weight standards (not shown) migrated through the 7.5% gel are indicated on the left.
mune system, and homologous regions of the host’s band 3 protein. The parasite encoded protein PfEMP 1 (Baruch et al. 1995) has been reported to be a potential ligand for adhesion of P. falciparuminfected erythrocytes. Typically, PfEMP 1 is identified in extracts of infected cells as a surface-iodinatable, Triton X-100-insoluble protein immunoprecipitated by strain-specific sera (reviewed by Howard et al. 1990). It is of interest that PfEMP 1, or a material with similar properties (molecular weight of approximately >200,000), is susceptible to exposure to 0.1%
trypsin for 5 min at room temperature (N.G., data not shown)) and can be immunoprecipitated from extracts of surface-iodinated infected cells (but not uninfected cell extracts) using four of the serm samples (RG 8820, RG 8821, RG 8834, and RG 8839) obtained from preimmune animals. Examination of the corresponding PEPSCAN profiles (based on 10 mers of putative exofacial regions), however, did not indicate a single common feature that was present in the four precipitating sera but absent from the 16 nonprecipitating samples. Therefore, it appears that if precipitation of PfEMP 1 was due to a cross–reaction with a band 3-related response then some other factor (such as conformational specificity) must have been involved. Notwithstanding our inability to accurately describe the factors that result in antigen precipitation it is noteworthy that in some instances immunoprecipitating sera may not result from exposure of the animal to the immunizing antigen. Because preimmune rabbit sera can have reactivity with an antigen, and because such reactivity can change (diminish) after immunization, the routine use of preimmune rabbit serum as a negative control may lead to erroneous conclusions as to antigen specificity. ACKNOWLEDGMENTS We thank Jacques Prudhomme for maintaining the parasite cultures and preparing the target cells. This research was supported in part by grants from the UNDP/World Bank/WHO Special Programme for Research and Training in Tropical Diseases and from the National Institute of Allergy and Infectious Diseases (National Institutes of Health).
REFERENCES BARUCH, D. I., PASLOSKE, B. L., SINGH, H. B., BI, X., MA, X. C., FELDMAN, M., TARASCHI, T. F., AND HOWARD, R. J. 1995. Cloning the Plasmodium falciparum gene encoding Pf EMP-1, a malarial variant antigen and adherence receptor on the surface of parasitized human erythrocytes. Cell 82, 77–87. CRANDALL, I., SMITH, H., AND SHERMAN, I. W. 1991. Plasmodium falciparum: The effect of pH and Ca2+ concentration on the in vitro cytoadherence of infected erythrocytes to a melanotic melanoma cells. Experimental Parasitology 73, 362–368. CRANDALL, I. E., AND SHERMAN, I. W. 1994a. Cytoadherence-related neoantigens on Plasmodium falciparum (hu-
JOBNAME: V82#1 96 PAGE: 9 SESS: 8 OUTPUT: Wed Jun 12 10:28:40 1996 /xypage/worksmart/tsp000/0752/20pu
NATURALLY OCCURRING IMMUNOGLOBULINS man malaria)-infected human erythrocytes result from the exposure of normally cryptic regions of the band 3 protein. Parasitology 108, 257–267. CRANDALL, I. E., AND SHERMAN, I. W. 1994b. Antibodies to synthetic peptides based on band 3 motifs react specifically with Plasmodium falciparum (human malaria)infected erythrocytes and block cytoadherence. Parasitology 108, 389–396. CRANDALL, I. E., GUTHRIE, N., AND SHERMAN, I. W. 1995.. Sera from individuals resident in The Gambia contain immunoglobulins that recognize human Band 3 peptide motifs. American Society of Tropical Medicine and Hygiene 52(5), 450–455. CRANDALL, I., LAND, K. M., AND SHERMAN, I. W. 1994. Plasmodium falciparum: Pfalhesin and CD36 form an adhesion/receptor pair that is responsible for the pH dependent portion of cytoadherence/sequestration. Experimental Parasitology 78, 203–209. CRANDALL, I., COLLINS, W. E., GYSIN, J., AND SHERMAN, I. W. 1993. Synthetic peptides based on motifs present in human band 3 protein inhibit cytoadherence/ sequestration of the malaria parasite Plasmodium falciparum. Proceedings of the National Academy of Sciences USA 90, 4703–4707. DAY, K. P., KARAMALIS, F., THOMPSON, J., BARNES, D. A., PETERSON, C., BROWN, H., AND KEMP, D. J. 1993. Genes necessary for expression of a virulence determinant and for transmission of Plasmodium falciparum are located on a 0.3 megabase region of chromosome 9. Proceedings of the National Academy of Sciences USA 90, 8292–8296. GEYSEN, H. M., MELOEN, R. H., AND BARTELING, S. J. 1984. Use of peptide synthesis to probe viral antigens for epitopes to a resolution of a single amino acid. Proceedings of the National Academy of Sciences USA 81, 3998–4002. GUTHRIE, N., CRANDALL, I. E., MARINI, S., AND SHERMAN, I. W. 1995. Monoclonal antibodies that react with human band 3 residues 542–555 recognize different conformations of this protein in uninfected and Plasmodium falciparum-infected erythrocytes. Molecular and Cellular Biochemistry 144, 117–123. HOWARD, R. J., BARNWELL, J. W., ROCK, E. P., NEEQUAYE, J., OFORI-ADJEI, D., MALOY, W. L., LYON, J. A., AND SAUL, A. 1988. Two approximately 300 kilodalton Plasmodium falciparum proteins at the surface membrane of infected erythrocytes. Molecular and Biochemical Parasitology 27, 207–224. HOWARD, R. J., HANDUNNETTI, S. M., HASLER, T., GILLADOGA, A., DE AGUIAR, J. C., PASLOSKE, B. L., MOREHEAD, K., ALBRECHT, G. R., AND VAN SCHRAVENDIJK, M. R. 1990. Surface molecules on Plasmodium falciparum-infected erythrocytes involved in adherence. American Journal of Tropical Medicine and Hygiene Suppl. 43(3), 15–29. KWAITKOWSKI, D. 1994. Prospects of an antidisease vaccine. In ’Molecular Immunological Considerations in Malaria
53
Vaccine Development’ Ed. (M. F Good and A. J. Saul, Ed). pp. 131–208. CRC Press, Boca Raton, FL. LAMBROS, C., AND VANDERBERG, J. P. 1980. Synchronization of Plasmodium falciparum erythrocytic stages in culture. Journal of Parasitology 65, 418–420. LUX, S. E., JOHN, K. M., KOPITO, R. R., AND LODISH, H. F. 1989. Cloning and characterization of band 3, the human erythrocyte anion-exchange protein (AE1). Proceedings of the National Academy of Sciences USA 86, 9089–9093. MARKWELL, M. A. K. 1982. A new solid-state reagent to iodinate proteins: Conditions for the efficient labelling of antiserum. Analytical Biochemistry 125, 427–432. OCKENHOUSE, C., KLOTZ, F. W., TANDON, N. N., AND JAMISON, G. A. 1992. Sequesrtrin, a CD36 recognition protein on Plasmodium falciparum malaria-infected erythrocytes identified by anti-idiotype antibodies. Proceedings of the National Academy of Sciences USA 88, 3175–3179. REITHMERIER, R. A. F. 1993. The erythrocyte anion transporter (band 3). Current Opinion in Structural Biology 3, 515–523. TAM, J. P. 1988. Synthetic peptide vaccine design: Synthesis and properties of a high-density multiple antigenic peptide system. Proceedings of the National Academy of Sciences USA 85, 5409–5413. TANNER, M. J. A., MARTIN, P. G., AND HIGH, S. 1988. The complete amino acid sequence of the human erythrocyte membrane anion-transport protein deduced from the cDNA sequence. Biochemical Journal 256, 703–712. TOWBIN, H., STAEHELIN, T., AND GORDON, J. 1974. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proceedings of the National Academy of Sciences USA 76, 4350–4354. TRAGER, W., AND JENSEN, J. B. 1976. Human malaria parasites in continuous culture. Science 193, 673–675. UDEINYA, I., SCHMIDT, J. A., AIKAWA, M., MILLER, L. H., AND GREEN, I. 1981. Falciparum malaria-infected erythrocytes specifically bind to cultured human endothelial cells. Science 213, 555–557. VAN SCHRAVENDIJK, M. R., ROCK, E. P., MARSH, K., ITO, Y., AIKAWA, M., NEEOUAYE, J., OFORI-ADEJEI, D., RODRIGUEZ, R., PATARROYO, M. E., AND HOWARD, R. J. 1991. Characterization and localization of Plasmodium falciparum surface antigens on infected erythrocytes from West African patients. Blood 78, 226–236. WHO Bulletin 1990. 68, 667–673. WINOGRAD, E., AND SHERMAN, I. W. 1989. Characterization of a modified red cell membrane protein expressed on erythrocytes infected with the human malaria parasite Plasmodium falciparum: Possible role as cytoadherent mediating protein. Journal of Cell Biology 108, 23–30. Received 8 September 1995; accepted 20 October 1995