Malaria Polyclonal B Cell Activator Activation of the Memory

Increased B Cell Survival and Preferential Activation of the Memory Compartment by a Malaria Polyclonal B Cell Activator

This information is current as Daria Donati, Bobo Mok, Arnaud Chêne, Hong Xu, Mathula of October 2, 2021. Thangarajh, Rickard Glas, Qijun Chen, Mats Wahlgren and Maria Teresa Bejarano J Immunol 2006; 177:3035-3044; ; doi: 10.4049/jimmunol.177.5.3035 http://www.jimmunol.org/content/177/5/3035 Downloaded from

References This article cites 35 articles, 15 of which you can access for free at: http://www.jimmunol.org/content/177/5/3035.full#ref-list-1 http://www.jimmunol.org/

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication by guest on October 2, 2021 *average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2006 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Increased B Cell Survival and Preferential Activation of the Memory Compartment by a Malaria Polyclonal B Cell Activator1

Daria Donati,2† Bobo Mok,* Arnaud Cheˆne,*† Hong Xu,† Mathula Thangarajh,‡ Rickard Glas,† Qijun Chen,§ Mats Wahlgren,* and Maria Teresa Bejarano*†

Chronic malaria infection is characterized by polyclonal B cell activation, hyperglobulinemia, and elevated titers of autoantibodies. We have recently identified the cysteine-rich interdomain region 1␣ (CIDR1␣)ofthePlasmodium falciparum erythrocyte membrane protein 1 as a T cell-independent polyclonal B cell activator and Ig binding protein. Here, we show that, although the binding affinity of CIDR1␣ to human IgM and IgG is relatively low, B cell activation still proceeds. CIDR1␣ rescues tonsillar B cells from apoptosis, and increases the proportion of cycling cells. Comparison of the impact on naive and memory B cell compartment indicated that CIDR1␣ preferentially activates memory B lymphocytes. Analysis of the gene expression profiles Downloaded from induced by CIDR1␣ and anti-Ig activation using a cDNA microarray demonstrated a low degree of homology in the signatures imposed by both stimuli. The microarray data correlate with the functional analysis demonstrating that CIDR1␣ activates various immunological pathways and protects B cells from apoptosis. Together, the results provide evidence for a role of malaria in preferentially activating the memory B cell compartment. The polyclonal B cell activation and augmented survival induced by CIDR1␣ is of relevance for understanding the mechanisms behind the increased risk of Burkitt’s lymphoma in malaria endemic areas. The Journal of Immunology, 2006, 177: 3035–3044. http://www.jimmunol.org/

berrant immune activation induced by chronic infections sent ϳ40% of the splenocytes. Thus, IE and their constituent Ags with Plasmodium falciparum leads to polyclonal B cell could interact in the spleen with B cells displaying a variety of A activation characterized by the presence of hyperglobu- surface phenotypes, Ag-binding repertoires and signaling proﬁles linemia (1), elevated titers of autoantibodies (2, 3), and frequent (7). Among malarial Ags, the P. falciparum erythrocyte membrane occurrence of Burkitt’s lymphoma (4) and splenic lymphoma (5). protein 1 (PfEMP1) family of proteins often display Ig binding The mechanisms that lead to this polyclonal B cell activation are properties (8, 9). The Ig binding activity of the PfEMP1, cloned

poorly understood. from two different P. falciparum strains, resides in two different by guest on October 2, 2021 The marked effect of malaria infection on B cells is related both variable domains, the Duffy binding-like domain 2␤ (DBL2␤) and to the biology of the infection, and to the nature of the malarial the cysteine-rich interdomain region 1␣ (CIDR1␣) (8). The latter Ags. P. falciparum-infected erythrocytes (IE)3 have the potential domain has been identified as a polyclonal B cell activator and an to directly interact with B cells in different anatomical sites and to Ig binding protein (IBP) (10) with a binding pattern similar to that induce B cell proliferation and differentiation into Ab-secreting of another microbial IBP, the protein A of Staphylococcus aureus cells. We have shown that a large proportion (83%) of fresh iso- (8, 10, 11). Microbial IBPs are produced by protozoa, viruses, and lates of IE bind nonimmune Igs (6), suggesting that in the periph- bacteria (12), and play important physiological roles (13). During eral blood IE could interact with B cells through their surface Igs. an infectious process, IBPs may act as an evasion mechanism to Moreover, bloodborne Ags (and thus malarial Ags related to the divert specific Ab responses (14, 15). CIDR1␣ binds to and acti- erythrocytic phase) are trapped in the spleen where B cells repre- vates purified B lymphocytes in vitro, an interaction partially mediated through the binding to surface Ig (10). To further understand the impact of CIDR1␣ on the immune *Microbiology and Tumorbiology Center, Karolinska Institutet, †Center for Infec- system, we analyzed its effect on the dynamics of the B cell com- tious Medicine, Department of Medicine, ‡Division of Neurology R54, Karolinska § partment and compared the gene expression profiles during acti- Institutet, Karolinska University Hospital Huddinge, and Swedish Institute for In- ␣ fection Disease Control, Stockholm, Sweden vation induced by CIDR1 and the triggering of the BCR via ␣ Received for publication May 24, 2005. Accepted for publication May 25, 2006. anti-Ig. The data show that CIDR1 preferentially induces the activation of the memory B cell compartment and that this activation The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance seems to be different from the one imposed by anti-Ig treatment. with 18 U.S.C. Section 1734 solely to indicate this fact. 1 This work was supported by grants from the Swedish International Development Cooperation Agency, Barncancerfonden, the Swedish Research Council, the Swedish Materials and Methods Foundation for Strategic Research, and the Karolinska Institutet. Production of recombinant Ags 2 Address correspondence and reprint requests to Dr. Daria Donati, Center for Infec- ␣ tious Medicine, Karolinska Institutet, Karolinska University Hospital Huddinge, F59, CIDR1 , of the cloned strain FCR3S1.2var1, was cloned in the pGEX-4T SE-141 86 Stockholm, Sweden. E-mail address: [email protected] plasmid (Amersham Biosciences) and expressed in Escherichia coli 3 (BL21) as previously described (8). The CIDR1␣-GST fusion-protein, re- Abbreviations used in this paper: IE, infected erythrocyte; PfEMP1, Plasmodium ␣ falciparum erythrocyte membrane protein 1; IBP, Ig binding protein; CIDR1␣, cys- ferred to as CIDR1 , was expressed and purified according to the instruc- teine-rich interdomain region 1␣; PI, propidium iodide; XIAP, X-linked inhibitor of tions of the manufacturer. GST produced by the empty vector was used as apoptosis; APRIL, a proliferation-inducing ligand; GC, germinal center; CT, cycle control. Henceforth, this is referred to as GST. The purity was determined threshold. by SDS-PAGE and Western blot as described (16) (data not shown).

B cells and cell culture tions. Real-time PCR was performed using predesigned assays (Applied Biosystems) for bcl-xL and X-linked inhibitor of apoptosis (XIAP) and a Buffy coats from blood of healthy individuals, never exposed to malaria, custom-designed assay for a proliferation-inducing ligand (APRIL). The were obtained from the blood bank of the Karolinska Hospital. Mononu- GAPDH gene was used as an endogenous control. Gene-specific PCR clear cells were isolated by centrifugation over Lymphoprep (Nycomed ϩ products were measured using an ABI PRISM 7700 (Applied Biosystems) Pharma). CD19 B cells were isolated by positive selection using the sequence detection system and analyzed with ABI PRISM 7000 SDS soft- MACS cell separation system (Miltenyi Biotec) according to the manu- ware. With the help of a standard curve, cycle threshold (CT) values were facturer’s instructions; the B cell purity varied between 94 and 99%. used to determine the corresponding mRNA quantities in each sample. Tonsils were obtained from patients undergoing routine tonsillectomy at Ն Samples with a CT 35 were excluded from the analysis. Results were the Karolinska University Hospital. Lymphocyte suspensions were pre- normalized for GADPH gene expression and therefore expressed as rela- pared by mincing the tissues and suspending the cells in complete RPMI tive mRNA expression. 1640. Isolated mononuclear cells were depleted of T cells by two rounds of rosette formation with amino ethyl isothiouronium bromide-treated SRBC DNA microarray analysis on ice. Rosettes were removed by centrifugation over Lymphoprep (17). The tonsillary B cell purity was Ͼ95% as revealed by FACS analysis after Microarray slides (KTH HUM 29.8k) were obtained from The Royal In- staining with the pan-T cell Ab (CD3) and the monocyte marker CD14. stitute of Technology, Sweden (͗www.biotech.kth.se/molbio/microarray/ Cultures were maintained in RPMI 1640 (Invitrogen Life Technologies) index.html͘ and ͗www.ebi.ac.uk/arrayexpress/͘; accession no. A-MEXP- supplemented with 10% FCS, 100 U/ml penicillin, and 2 mM glutamine. 114). Differential gene expression was investigated using two-color Purified B cells were cultured either in medium alone, or medium contain- hybridization scheme in three independent experiments performed with B ␣ Ј ␮ cells from different donors. B cells were cultured for 24 h in medium alone ing GST, CIDR1 , or anti-Ig F(ab )2 (10 g/ml) (Jackson ImmunoRe- or supplemented with recombinant CIDR1␣, GST, or anti-Ig, and RNA search Laboratories), and incubated up to 48 h at 37°C and 5% CO2. Unless otherwise specified, the final concentration of GST and CIDR1␣ was 100 was extracted using TRIzol reagent (Invitrogen Life Technologies) accord- ␮g/ml. ing to the manufacturer’s instructions. Residual DNA was removed by

Approval for these studies was obtained from the Karolinska Institutet treatment with DNA-free (Ambion). The RNA quality was checked with an Downloaded from Ethical Committee. Agilent 2100 Bioanalyser. RNA (ϳ2 ␮g) was reverse transcribed into cDNA with Superscript III RNase HϪ reverse transcriptase (Invitrogen Phenotypic analysis Life Technologies), sonicated for 30 s, purified using Microcon filters (YM-30; Millipore), and subsequently labeled with Cy3-dCTP or Cy5- Saturating concentrations of PE- or avidin-conjugated monoclonal anti- dCTP by random priming using the Klenow fragment (40 U/ml; Invitrogen human Abs (mAb) (BD Biosciences) were added to cell pellets collected Life Technologies) at 37°C for 2.5 h. The reactions were stopped by the after 48 h of culture and incubated for 30 min at 4°C. PE- and FITC- addition of 0.5 M EDTA (pH 8.0), and unincorporated nucleotides were conjugated isotype-matched mAbs (BD Biosciences) with irrelevant spec- removed using Microcon filters (YM-30). The two fluorescent probes were http://www.jimmunol.org/ ificities were used as negative controls. Fluorescence intensity was mea- mixed with hybridization buffer, denatured at 95°C for 2 min, and vacuum- sured with a FACSCalibur flow cytometer, and analyzed using the concentrated for 1 min. Hybridization and washings were performed using CellQuest software (BD Biosciences). HS400 hybridization station (TECAN); thereafter, the slides were scanned B cell survival and cell cycle analysis with a GenePix 4000 A scanner (Axon Instruments); at least three hybrid- izations (with at least one dye-swap experiment) were performed. Expres- DNA profiles were obtained by staining cells with propidium iodide (PI) sion data from replicate microarray hybridization showed a high degree of (Sigma-Aldrich). Cultured tonsillar B lymphocytes (0.5 ϫ 106/ml) were reproducibility with correlation coefficients ϳ0.8 (data not shown). Fluo- washed with ice-cold PBS and resuspended in 100 ␮l of PBS containing PI (50 rescent spot and local background intensities were quantified using Gene- ␮g/ml), and 0.1% (v/v) Triton X-100. Cells were incubated for 6–8 h at 4°C pix Pro 5.1 software (Axon Instruments). Spots that passed the quality ϩ before being analyzed on a FACSCalibur. The rescue from cell death was controls (i.e., 55% of pixel intensities should be greater than background by guest on October 2, 2021 calculated from the percentage of cell death in different cultures, using the 1 SD and the signal intensity Ͻ3% of the saturation level for each channel), following formula: percentage of rescue ϭ 1 Ϫ (percentage of cell death in together with visual inspection, were analyzed using GeneSpring 6.1 soft- CIDR1 cultures/percentage of cell death in control cultures) ϫ 100. ware (Silicon Genetics). Local background was subtracted from spot sig-

The distribution of the cells in the sub G0-G1 fraction and in the different nals, and fluorescence ratios were calculated. To compensate for unequal phases of the cell cycle was analyzed by the CellQuest software (BD dye incorporation or any effect of the amount of template, data normaliza- Biosciences). tion was done by LOWESS, an intensity-dependent normalization ap- proach. The complete data set is available at ͗www.ebi.ac.uk/arrayex- Immunoblot analysis press/͘ (accession no. E-MEXP-316). The data were selected using three different filters: 1) by flags removing genes that were absent in any repli- Lysates of B lymphocytes were separated by SDS-PAGE, and analyzed by cate (no signal detected), 2) by expression level to remove those genes that immunoblotting. To control for an equal loading of cell lysate, protein were deemed to be unchanging between log values 0.5 and 2.0 (Ͼ2-fold concentration was determined by BCA assay (Pierce Biotechnology) ac- difference), and 3) by confidence using a one-sample t test against the cording to the manufacturer’s instruction, using BSA as a standard. SDS- baseline value of 1. Hierarchical clustering of gene expression profiles was PAGE resolved samples were transferred to a nitrocellulose membrane and performed using the Pearson correlation. Analysis of gene expression pro- probed with primary (anti-Bcl2 and Bcl-xL; DakoCytomation) and HRP- file was conducted on three groups corresponding to different experimental conjugated secondary Ab (DakoCytomation), followed by ECL reagent conditions: CIDR1␣ vs GST, CIDR1␣ vs medium, and anti-Ig vs medium. (ECL) detection (Amersham Biosciences). A first unsupervised gene expression analysis was made on all three com- ␣ ␣ Ig binding affinity determination by competition ELISA binations, given the similarity between CIDR1 vs GST, CIDR1 vs medium signatures, the supervised analysis was conducted only on CIDR1␣ Increasing amounts of CIDR1␣ protein (3 ϫ 10Ϫ8 to 5 ϫ 10Ϫ6 M) were vs GST and anti-Ig vs medium. For comparison between CIDR1␣ vs GST incubated for2hatroom temperature with a constant concentration of IgG and anti-Ig vs medium, the genes up- or down-regulated in CIDR1␣ vs or IgM (5 ␮g/ml) (Jackson ImmunoResearch Laboratories) in PBS-Tween GST were selected and compared with the expression of the same genes in 0.05%. The mixture was thereafter added to wells precoated with 1.2 ␮gof anti-Ig vs medium. Because the number of replicates of each comparison CIDR1␣. Following incubation at room temperature for 1 h, alkaline phos- is limited, we applied the Cross-Gene Error Model (provided in the Gene- phatase-conjugated anti-human IgM or anti-human IgG were added, and Spring software), which provides a more accurate estimate for the precision the plates were further incubated for 1 h. The concentration of Ag which of a expression data by combining measurement variation and between- gave half-maximum absorption at 405 nm was determined by linear re- sample variation information. gression and regarded as the value of the dissociation constant (K ) that d Semiquantitative RT-PCR corresponds to an association constant (Ka)of1/Kd. Values were computed graphically using four-parameter sigmoidal curve fitting. To verify the microarray results, semiquantitative RT-PCR was performed, RNA preparation and real-time PCR to measure gene expression of XCL1, GPS1, CDK4, and POLE4 mRNAs on B cell cultured for 24 h. RNA extraction, DNase treatment, and reverse Total RNA was isolated from frozen B cell pellets after 12 h of culture in transcription were performed as previously described. The PCR was per- medium alone or containing GST or CIDR1␣. The RNA was extracted formed using Platinum Taq polymerase (Invitrogen Life Technologies) with Qiagen’s RNeasy mini-kit and reverse-transcribed using TaqMan RT with oligonucleotide primer pairs specific for XCL1 (forward, 5Ј-GCA reagents (Applied Biosystems) according to the manufacturer’s instruc- GAATCAAGACCTACACCATCAC-3Ј; reverse, 5Ј-ATTGCTGGGTTC The Journal of Immunology 3037

CTGTTGGC-3Ј), GSP1 (forward, 5Ј-CACAGGTCCGAGACATCAT CTTC-3Ј; reverse, 5Ј-CCTTGCCCATCAACAGAGACTTC-3Ј), CDK4 (forward, 5Ј-AGGCTTTTGAGCATCCCAATG-3Ј; reverse, 5Ј-CCAC CACTTGTCACCAGAATGTTC-3Ј), POLE4 (forward, 5Ј-CCATCT TCATTCTGGCACG-3Ј; reverse, 5Ј-GCATTATCCAAGTCTCTCCT CTG-3Ј), ␤-actin (forward, 5Ј-ACTGTGCCCATCTACGAGGGG TAT-3Ј; reverse: 5Ј-TCCTTAATGTCACGCACGATTTCC-3Ј). The PCR were denaturated at 95°C/3 min and temperature cycled at 95°C/30 s, 56°C/30 s and 68°C/45 s for 28–30 cycles. All data were normalized to an internal housekeeping ␤-actin control, and the linear amplification range for each gene was tested on the adjusted cDNA. Quantification was done using the Quantity One Software (Bio-Rad). Statistical analysis A Student paired two-tailed t test was performed. A value of p Ͻ 0.05 was considered significant. Results CIDR1␣ preferentially activates the memory B cell compartment To determine the effect of CIDR1␣ on the composition of different B lymphocyte subsets in peripheral blood, we analyzed and compared the proportion and expression of activation markers in the Downloaded from memory and naive compartments 48 h following CIDR1␣ stimulation. The proportion of CD27ϩ (memory B cells) and IgDϩCD27Ϫ (naive B cells) varied among different donors (data not shown). Within the memory B cell compartment, the relative proportion of cells expressing the activation markers CD70 (CD27ϩCD70ϩ) and CD95 (CD27ϩCD95ϩ) increased between 2 http://www.jimmunol.org/ and 4%, and 8 and 10%, respectively, following CIDR1␣ stimulation (Fig. 1). These changes are modest, but both consistent and statistically significant ( p Ͻ 0.05). In contrast, CIDR1␣ did not significantly affect the proportion of CD27Ϫ naive B cells, inde- pendently from their activation status (Fig. 1). However, in all experiments, CIDR1␣ stimulation induced a low but consistent FIGURE 1. CIDR1␣ preferentially activates memory (CD27ϩ) B cells. increase in the relative proportion of B cells expressing both IgD ϩ and CD27; such cells seem to play a crucial in the secondary im- CD19 B cells were cultured for 48 h in medium alone, or containing GST ␣ by guest on October 2, 2021 mune response by producing high-affinity IgM (18) (Fig. 1). or CIDR1 , and stained with FITC- or PE-conjugated anti-human Abs. A, Density plot analysis of the proportion of activated memory B cell subsets. Analysis of the levels of expression of activation markers ϩ ␣ Data show one representative of eight independent experiments. B, Histo- showed that, in the memory subpopulation (CD27 ), CIDR1 en- gram of the median values (ϮSD) of the proportion of naive (CD27Ϫ) and hanced the expression of the activation markers CD70, CD95, memory (CD27ϩ) B cells expressing different markers (CD70, CD95, CD69, the costimulatory molecule CD86, and HLA-DR (Fig. 2A) IgD). Ⅺ and f, Data relative to the GST and CIDR1␣-treated B cells, without affecting the expression of CD23, CD27, CD54, CD80, respectively. Statistical significance was determined by a two-tailed paired and HLA-ABC molecules (data not shown). In the naive subpopu- Student t test on data collected from eight independent experiments with .(p Ͻ 0.05 ,ء) lation (CD27Ϫ), CIDR1␣ increased the expression of HLA-DR eight donors and induced down-regulation of HLA-ABC expression, whereas the level of expression of IgD and the activation molecules CD95, CD70 were not significantly altered (Fig. 2B). The level of the dergo spontaneous apoptosis when cultured (17). At time 0, im- activation molecules CD23 and CD69 and of the costimulatory mediately after isolation, the level of cell death was negligible. At molecules CD54, CD80, and CD86 were not affected (data not the earliest time point analyzed, 24 h, the survival was higher in shown). It is not likely that the difference in responsiveness to the CIDR1␣ cultures compared with GST and medium control CIDR1␣ obeys a different kinetic between naive and memory B cultures (Fig. 3). cells. Naive B cells show a very modest activation at 24 h (see Fig. In six experiments, the median percentage Ϯ SD of dead cells Ͻ ␣ Ϯ 2B) that with the time (48 h) disappears. Initial kinetic studies ( G0/G1) in the CIDR1 culture at 24 h was 39 15%, whereas performed on total B cells did not reveal a change in the response, in the medium and in GST cultures, it was 50 Ϯ 16 and 55 Ϯ 18%, or a biphasic response when the cultures were analyzed for up to respectively (Fig. 4). The median CIDR1␣-induced relative pro- 5 days. Furthermore, naive and memory B cells have different tection from cell death was 22 and 29% when compared with me- thresholds of activation (19, 20). dium and GST, respectively ( p Ͻ 0.05). Treatment with anti-Ig, Together, the data indicate that CIDR1␣ preferentially activates known to rescue B cells from death, induced a median protection the memory B cell compartment. of 31%. By 48 h, the effect of CIDR1␣ on cell survival was no longer evident, although the rescue mediated by anti-Ig was still ␣ CIDR1 increases B cell survival and drives cells into cell cycle seen (data not shown). To analyze the effects of CIDR1␣ on the survival and cell cycle We conclude that CIDR1␣ mediates a degree of protection from progression of primary B cells, tonsillar B lymphocytes were cul- apoptosis in B lymphocytes, equal to, although more transient tured in the presence of GST or CIDR1␣ for 24–48 h, and their than, that induced by anti-Ig. Along with the rescue from cell DNA content was measured by PI staining. Tonsillar B cell pop- death, CIDR1␣ induced cell cycle progression with a modest al- ulations contain germinal center (GC) cells that are prone to un- though consistent increase in the proportion of cycling cells 3038 MALARIA AND POLYCLONAL B CELL ACTIVATION Downloaded from http://www.jimmunol.org/

FIGURE 2. Phenotypic changes of memory and naive B lymphocytes FIGURE 3. CIDR1␣ rescues tonsillary B cells from spontaneous cell ϩ stimulated with CIDR1␣. CD19 B cells were cultured for 48 h in medium death. Unstimulated B cells were cultured for 24 h in medium alone or alone, or containing GST or CIDR1␣, and stained with FITC- or PE-con- containing GST, CIDR1␣, anti-Ig, or PMA plus ionomycin. Histograms jugated anti-human Abs. Histograms show median values of the mean in- illustrate the DNA content measured by PI incorporation. The numbers Ϯ Ͻ tensity ﬂuorescence (MFI SD) of the markers analyzed on the following: represent the percentage of cell death ( G0/G1) and cells in S, G2/M ϩ Ϫ the memory population, CD27 (A); the naive population, CD27 (B). phases are indicated as cycling cells. The percentage of rescue from cell Statistical signiﬁcance was determined by a two-tailed paired Student t test death was calculated as follows: 1 Ϫ (percentage of cell death in CIDR -p Ͻ 0.05). cultures/percentage of cell death in control cultures) ϫ 100. Data are rep ,ء) on data collected from eight independent experiments

resentative of one of six independent experiments. by guest on October 2, 2021 ϩ ␣ (S G2/M). In the CIDR1 -treated cultures, the percentage of cy- Ϯ Ϯ cling cells was 6.6% (3 2% in S phase, 3 2% in G2/M phase), ␣ whereas in the GST and medium cultures the values were 4% (2 Ϯ M) for IgG. These Ka values place the CIDR1 among the low- Ϯ 2% in S phase, 2 2% in G2/M phase) (Fig. 4). It could be argued affinity Ags, although able to allow BCR triggering (21). that the enhanced survival seen in CIDR1␣ cultures reflects an ␣ increased number of viable cells due to more cell division rather The impact of CIDR1 on genes that regulate the apoptotic than reflecting protection from death. This assumption is unlikely pathway because the proportion of cells that have undergone cell division at Given the capacity of CIDR1␣ to protect B cells from cell death, 24 h is negligible. we investigated the effect of CIDR1␣ on the expression of some

The proportions of resting cells in G0/G1 phase were as follows: genes that regulate cell survival and apoptosis such as bcl2, bcl-xL, 67 Ϯ 19% in CIDR1␣-treated cells, 41 Ϯ 15% in the GST, and APRIL, and XIAP (22). We quantified and compared by quantita- Ϯ 46 18% in medium control cultures (Fig. 4). B cells treated with tive real-time PCR the mRNA expression levels of bcl-xL, APRIL, ϳ ␣ anti-Ig and PMA plus ionomycin had 2-fold increase in the and XIAP. CIDR1 treatment did not affect bcl-xL mRNA levels amount of cells that progressed toward the G2/M phase compared and produced a modest decrease of APRIL mRNA, significantly with the medium control. The differences between CIDR1␣ treat- increasing XIAP mRNA ( p Ͻ 0.05) (Fig. 6A). At the protein level, ␣ ment and medium or GST control, regarding the proportion of cells CIDR1 produced down-modulation of Bcl-xL with a slight down- Ͻ in S, G2/M, and G0/G1 phase, were statistically significant ( p regulation of Bcl2 expression (Fig. 6B). Thus, the protection from 0.05). cell death induced by CIDR1␣ may be mediated by inducing the expression of the antiapoptotic regulator XIAP without affecting CIDR1␣ has a low binding affinity for Igs the bcl2, bcl-xL intrinsic antiapoptotic pathway. This protective We have previously shown that CIDR1␣ binds to and activates B effect appears to be different from the one mediated by the anti-Ig. cells, an interaction mediated, at least in part, through the binding ␣ to surface IgM and IgG (10). To further analyze the binding of CIDR1 and Ig activation generate a different gene expression CIDR1␣ to human Ig and obtain an estimation of the affinity of the profile interaction, we performed competition ELISA experiments. Fig. 5 In our previous studies, competition experiments demonstrated shows results relative to the percentage of IgM and IgG binding to that the CIDR1␣ induced B cell activation is only partially medi- CIDR1␣. The affinity values obtained by calculation in competing ated by its binding to Igs (10). Thus, it became of interest to an- Ϸ ϫ 6 Ϫ1 ϭ Ϯ ϫ Ϫ7 conditions were Ka 3.7 10 M (Kd 2.68 0.30 10 alyze and compare the gene expression profiles induced by Ϸ ϫ 6 Ϫ1 ϭ Ϯ ϫ Ϫ7 ␣ M) for IgM and Ka 1.2 10 M (Kd 8.57 0.89 10 CIDR1 and anti-Ig. To this end, three comparative analyses were The Journal of Immunology 3039

FIGURE 4. CIDR1␣ promotes cell cycle entry. Percentage of B cells in the different phases of the cell cycle after 24 h of culture in medium alone or containing the following: CIDR1␣, GST, anti-Ig, or PMA plus ionomycin. Cell cycle analysis was performed after PI staining. Median values of six independent experiments are shown.

FIGURE 5. Evaluation of CIDR1␣ affinity for human IgG and IgM by competition ELISA. Inhibition of human Ig binding (IgG and IgM) to Downloaded from performed: CIDR1␣ vs GST, CIDR1␣ vs medium, and anti-Ig vs immobilized CIDR1␣ was evaluated in a competition ELISA as described medium. First, we conducted an unsupervised analysis of the in Materials and Methods. Values were computed graphically using four- CIDR1␣ vs GST profile, compared with the CIDR1␣ vs medium parameter sigmoidal curve fitting. Curves shown are representative of six and identified a high degree of homology (95%) (data not shown). independent experiments. This degree of homology between the two profiles confirmed that, in our experimental conditions, the effect of GST on B cell re- ␣ Up-regulated genes induced by CIDR1 includes several genes http://www.jimmunol.org/ sponses is minor; hence, we will only refer to the comparisons previously described as mitogen-activated, involved in pathways between CIDR1␣ vs GST and anti-Ig vs medium. To compare the that control cell growth/apoptosis, transcription/translation, and gene expression signatures induced by CIDR1␣ and anti-Ig, we that are normally induced during immune responses. The up-reg- performed a supervised analysis and selected those genes up- or ulation of both TRAF3 and TRAF4 suggests activation of the down-regulated by CIDR1␣ (CIDR1␣ vs GST). The same set of NF-␬B pathway. The increased expression of two of the most im- genes was then analyzed in anti-Ig vs medium, and the two ex- portant protein kinases, PKC and the PKA related forms pression profiles were compared (Fig. 7A). (PRKAG3, PRKX), is in line with the observed CIDR1␣-mediated The immunological signatures of CIDR1␣ and anti-Ig showed a up-regulation of genes involved in different signaling pathways; high number of differentially expressed genes. Although the ma- among them: the MEK/ERK pathway (EIF2B5), the MEKK/JNK by guest on October 2, 2021 jority (97%) of the genes listed in Fig. 7A were differentially ex- pathway (TRAF3, GPS1), and the MKK/MAPK pathways pressed in the two groups (CIDR1␣ vs GST and anti-Ig vs me- (MAPK4K5, IQGAP1, IQGAP2). These signaling pathways lead to dium), only a small number of genes (n ϭ 8) were common. Interestingly, these eight genes were all up-regulated (Fig. 7B, Ta- ble I). Among them, U2AF2, ZNF135, and SLC39A3 indicate activation-induced gene transcription. GSTM2 is commonly associated with inhibition of cell death and activation of different signaling pathways, such as the MEK pathway. The other four genes do not as yet appear to have known function or apparent relation to the activation process.

CIDR1␣-treated B cells display an activated gene expression profile To further understand the cellular processes altered by CIDR1␣, we analyzed and compared the gene expression profile of CIDR1␣ and GST control-treated B cells. We examined the expression of ϳ29,800 genes using the human microarray KTH HUM 29.8k. Analysis of the data established that a total of 377 genes exhibited different and significantly altered gene expression between CIDR1␣ and GST cultured B cells. These included 105 known genes, of which 81 were up-regulated (Table II) and 24 down- regulated (Table III). A cluster analysis of known genes made according to their cellular function (determined by the gene infor- FIGURE 6. Effect of CIDR1␣ on Bcl , Bcl-x , APRIL, and XIAP ex- mation: SwissProtID (ExpASy Proteomics Server; ͗www.expasy- 2 L ͘ pression. B cells were cultured for 12–24 h in medium alone or containing .org ) and PubMed search), allowed classification into five groups: ␣ GST, CIDR1 , or anti-Ig. A, Expression of the mRNA levels of bcl-xL, cell growth/apoptosis; immune response; transcription and trans- APRIL, and XIAP was performed by quantitative real-time PCR after 12 h lation; immunological signaling; and cell communication (Tables of culture. Data are expressed as relative to the expression of GAPDH II and III). The largest groups were represented by genes involved mRNA. B, Expression of the antiapoptotic proteins Bcl2 and Bcl-xL was in immunological signaling pathways, cell growth and apoptosis, evaluated by Western blot after 24 h of culture. Statistical significance was .(p Ͻ 0.05 ,ء) and transcription. determined by a two-tailed paired Student t test 3040 MALARIA AND POLYCLONAL B CELL ACTIVATION Downloaded from http://www.jimmunol.org/

FIGURE 7. Gene expression profile comparison between CIDR1␣ and Ig stimulation on B lymphocytes. Cluster analysis of gene expression profile of B cells in two different experimental conditions: CIDR1␣ vs GST (CG) and anti-Ig vs medium (IM) (see Materials and Methods). A, Tree view display of the results of 377 differentially expressed genes under CG condition (with the positive expression values appearing in all two conditions) as described in Materials and Methods. Each row represents a gene; each column represents a sample. The level of expression of each gene in each sample is represented Ͼ using a red-blue color scale (red, high expression). B, Enlargement of the Tree view display of the genes showing up-regulation ( 2-fold change) in both by guest on October 2, 2021 conditions. The data were processed with the GeneSpring 6.1 software. The color scale is based on the signal ratio in between each condition, and ranges between 0 (blue) and 6 (red), where red indicates up-regulation and blue indicates down-regulation. C, Validation of the microarray results by semiquantitative PCR. Transcription level of two up-regulated genes (XCL1 and GPS1) and two down-regulated genes (CDK4 and POLE4) were verified by semiquantitative PCR, to validate the microarray analysis (left panel). The expected sizes of XCL1, GPS1, CDK4, POLE4, and ␤-actin RT-PCR products were 186, 363, 265, 118, and 158 bp, respectively. The histograms in the right panel show the relative gene expression levels after normalization against the housekeeping ␤-actin gene (bottom). Expressions relative to CIDR␣-treated B cells are shown in dark gray (the numbers above the significance bar show .(indicates the consistency of CIDR1␣ vs GST compared with the array data (ء) the fold increase/decrease in CIDR1␣ vs GST, and the asterisk

activation and survival are involved in B cell immune responses. important Ca2ϩ buffer molecule and early mediator of cellular ac- CIDR1␣ may inhibit the JAK/STAT pathway as indicated by an tivation known to inhibit numerous apoptotic stimuli. Later medi- increased expression in the negative regulators PIAS2 and SOCS1. ators are DAPK2, DNAJC7 also known as Hsp40, and HRK or The coordinated regulation of these different pathways results in B hara-kiri whose overexpression can either induce or suppress ap- cell activation as reflected by the activated phenotype (Fig. 2). optosis. FBXO6 and FPGS genes are involved in the cell cycle CIDR1␣ stimulation also increased the gene expression of nu- control and proliferation. merous transcription factors including DOTL-1, Ets-2 (ETV1), A small number of genes with known function were down-reg- TLE4, and many members of the ZNF family (ZNF36, ZNF135, ulated by CIDR1␣; few of them have been described in association ZNF318, ZNF263) (Table II). The concerted action of these tran- with B lymphocyte functions. Genes functionally clustered in the scription factors is crucial for the processes that lead to differen- cell growth/apoptosis were CDK4, important for cell cycle pro- tiation of mature B cells into Ig-secreting plasma cells and memory gression; RP9, described in relation to B cell proliferation; and B cells (23–25). Up-regulated genes strictly correlated to immune SAV1 and TNFSF8 (CD30). SAV1 promotes exit from cell cycle responses included: chemokine and chemokine receptors such as and together with TNFSF8 induces apoptosis. Two genes were the lymphoactin XCL1 and CCRL2, often expressed during B cell classified as part of the immune response group, CD36 and activation and migration to secondary lymph nodes; the IL-7R HLA-A: the latter for which we had noted down-regulation of sur- (IL7R) expressed during different phases of B cell development; face protein expression (Fig. 2B). To confirm the array data, we the insulin-like growth factor binding protein 5 (IGFBP5); and the run a semiquantitative RT-PCR on two up-regulated (XCL1, IL-18 expressed on naive, memory, and GC B cells upon activa- GPS1) and two down-regulated (CDK4, POLE4) genes. The data tion. The up-regulation of the HLA-DQ is indicative of B cell showed an increase/decrease of gene expression consistent with activation. Other up-regulated genes are involved in early and late the one seen in the array (Fig. 7C). Taken together, the microarray events of cell growth and apoptosis. CALB1 or calbindin is an data are consistent with the phenotypic and functional analysis The Journal of Immunology 3041

Table 1. Genes up-regulated by both CIDR1␣ and anti-Ig

Fold Accession No. Change HUGO IDa Functional Catagories Description

AA142971 2.5 GSTM2 Cell grow/apoptosis GST M2

N67262 2.2 ZNF135 Transcription/translation Zinc ﬁnger protein 135

N59534 2.2 SLC39A3 Immunological signaling Solute carrier family 39 (zinc transporter), member 3 AA620672 2.7 U2AF2 U2 (RNU2) small nuclear RNA auxiliary factor 2

AA057620 2.5 LOC91689 Unknown AA779352 2.5 PLAC4 Placenta-speciﬁc 4 AA608893 2.1 H17551 6.4

a Annotation based on HUGO Gene Nomenclature Committee: Ͻwww.ensembl.org/Ͼ.

demonstrating that CIDR1␣ activates human B lymphocytes. specific Ab responses (14, 15) is in line with the lack of specific Ig However, the signaling pathways induced by CIDR1␣ do not seem memory observed in children from malaria endemic areas (27). Downloaded from to be the same as those induced by BCR (anti-Ig signaling) as CIDR1␣ protects a significant proportion of B cells from spon- reflected by the different gene expression profiles. taneous cell death, an outcome that is accompanied by the up- regulation of the antiapoptotic factor XIAP. The contention that Discussion the two observations are linked is supported by the up-regulation Although the impact of P. falciparum malaria on B cell-mediated of numerous genes involved in the activation of the ERK/MAPK immunity has been recognized for ϳ40 years (1), little is known pathway (Table II), known to be associated with suppression of http://www.jimmunol.org/ about the identity of the Ags and the mechanisms that underlie the proapoptotic signaling and sustaining XIAP levels (28). Following P. falcipa- ␣ production of large amounts of nonspecific Igs during CIDR1 activation, a large proportion of B cells stay in the G0/G1 rum malaria infection. Our recent identification of the CIDR1␣ phase with concomitant protection from spontaneous apoptosis. domain of PfEMP1 as a polyclonal B cell activator and as an IBP However, the given signal is sufficient to allow a small proportion led us to investigate further its effects on the B cell compartment. of cells to proceed beyond the S phase restriction point leading to The data presented in this paper extend our previous results and, proliferation. Polyclonal activators stimulate rapid G entry, pro- ␣ 1 importantly, demonstrate that CIDR1 preferentially activates tection from apoptosis, and S phase entry, which is dependent on memory B cells and protects B cells from apoptosis. B cell differentiation state and the Ag affinity (29–31). A signal by guest on October 2, 2021 ␣ It could be argued that because CIDR1 fusion protein was generated through surface Ig can lead to proliferation or apoptosis, produced in E. coli, the activation observed could be due to con- depending on the maturation stage of the responding B cell, and on taminating endotoxin. We can rule out this possibility because the additional signals provided by costimulation and T cell help. In proliferative response to the recombinant Ag preparations was not vivo, the antiapoptotic and B cell activation properties of CIDR1␣ affected by the presence of polymyxin B or LPS. Furthermore, may be potentiated by other not-yet-identified polyclonal B cell human B cells, in contrast to murine B cells, do not respond to activators present in P. falciparum-infected cells or by additional endotoxin or LPS, because they do not express the TLR-2 and -4, B cell stimulatory factors such as IL-10, IgE, known to be pro- essential components of the LPS receptor signaling complex (26). duced during malaria infection (32). The occurrence of protracted The preferential activation of the memory compartment may malaria episodes is an additive factor that may enhance and per- relate to their lower threshold for activation, and is in line with the petuate the B cell stimulation that finally leads to malignant results of Bernasconi et al. (19), who demonstrated that memory B development. lymphocytes selectively proliferate and differentiate into plasma CIDR1␣ effect, as polyclonal B cell activator, is partially me- cells in vitro in response to polyclonal stimuli in the absence of ␣ BCR triggering; whereas naive B cells require specific BCR trig- diated through binding to surface Ig (10). In vitro, CIDR1 also gering. These results led to the hypothesis that one of the mech- binds to CD36, and PECAM-1/CD31 (6, 8, 16); thus, the effect of ␣ anisms for the maintenance of serological memory involves con- CIDR1 on the cell cycle may involve not only the interaction tinuous activation of memory B cells by polyclonal B cell with surface Ig but also the binding to other receptors that affect activators (19). Most likely, CIDR1␣, as a polyclonal B cell acti- the decision to proceed to activation. vator, contributes in a similar fashion to the hyperglobulinemia The binding affinity of an Ag is important for the outcome of ␣ that characterizes chronic malaria infection (1). It should be noted activation. The binding affinity of CIDR1 to IgG and IgM clas- that, in vivo, the consequence of CIDR1␣ interacting with B lym- sifies it as a low-affinity Ag. However, the same level of affinity, phocytes may be potentiated by costimulation and bystander T cell has been demonstrated to be sufficient to drive B cell proliferation help (19). During a malaria episode, IE are trapped in the spleen and differentiation during T-independent immune responses (21). Ϸ 5 Ϫ1 where splenic B cells may be activated by CIDR1␣ expressed on Indeed, Ags with very low affinity for the BCR (Ka 10 M ) the IE surface, and presented as soluble Ag, or as immune com- can transduce activation signals (33–35). During T-independent Ϸ 5 6 plexes, by dendritic cells in the presence of cytokines, T cell help, immune responses, large differences in affinity (Ka 10 –10 Ϫ and costimulatory signals. In contrast, the polyclonal memory B M 1) produce only small variations in the intrinsic ability of B cell activation induced by CIDR1␣ may impair the maintenance of cells to mount a productive response to Ag (21). The in vivo Ag-specific memory B cells. Thus, the suggested role of microbial impact of an Ag results from a combination of its affinity and IBPs, and thereby of CIDR1␣, as an evasion mechanism to divert persistence. A low antigenic clearance of CIDR1␣, due to high 3042 MALARIA AND POLYCLONAL B CELL ACTIVATION

Table II. Genes up-regulated by CIDR1␣ (CIDR vs GST)

Fold Accession No. Change SD HUGO IDa Functional Categories Description

AA989473 2.8 0.61 BNIP1 Cell growth/apoptosis BCL2/adenovirus E1B 19-kDa interacting protein 1 H88329 2.2 0.35 CALB1 Calbindin 1 AA284072 2.8 0.59 CDKN3 Cyclin-dependent kinase inhibitor 3 AI362933 3.0 0.68 DAPK2 Death-associated protein kinase 2 AA626252 2.1 0.25 DNAJC7 DnaJ (Hsp40) homolog AA142971 2.5 0.19 GSTM2 GST M2 AI083676 3.0 0.46 HRK Harakiri, BCL2 interacting protein W47183 2.4 0.20 FBXO6 F-box only protein 6 R44864 2.1 0.06 FPGS Folylpolyglutamate synthase AI017433 2.3 0.34 GGA2 Golgi-associated ␥ adaptin ear containing, ARF binding protein 2 AI344518 2.1 0.13 H2AFV H2A histone family, member V AA171899 2.4 0.20 HSPC121 Butyrate-induced transcript 1 AA481057 3.7 0.50 KBTBD6 Kelch repeat and BTB (POZ) domain 6 AA452147 3.2 0.58 PSMD5 Proteasome (prosome, macropain) 26S subunit, non-ATPase, 5 AI360342 2.2 0.11 RAB33A RAB33A, member RAS oncogene family AA463642 2.2 0.43 RAB5B RAB5B, member RAS oncogene family H20138 2.6 0.44 RAB6A RAB6A, member RAS oncogene family AA937215 2.3 0.38 CHAD Chondroadherin

AI288845 2.8 0.10 CCRL2 Immune response Chemokine (COC motif) receptor-like 2 T63324 2.0 0.33 HLA-DQA2 MHC class II

T52830 2.5 0.42 IGFBP5 Insulin-like growth factor binding protein 5 Downloaded from AI129421 2.6 0.63 IL18 IL-18 AA485865 2.1 0.31 IL7R IL-7R AI298976 2.4 0.33 XCL1 Chemokine (C motif) ligand 1 W60968 3.6 0.79 EVA1 Epithelial V-like Ag 1 AI334914 2.4 0.38 ITGA2B CD41B

AA045320 2.0 0.26 AADAC Transcription/translation Arylacetamide deacetylase (esterase) AA917506 2.0 0.19 DOT1L DOT1-like, histone H3 methyltransferase AA406285 2.3 0.06 DRAP1 DR1-associated protein 1 http://www.jimmunol.org/ AA191548 2.7 0.28 ELL2 Elongation factor, RNA polymerase II AA486753 2.1 0.12 ETV1 Ets variant gene 1 AA206497 2.2 0.32 PSMA4 Proteasome (prosome, macropain) subunit, ␣ type, 4 AA620580 2.7 0.33 PSMB3 Proteasome (prosome, macropain) subunit, ␤ type, 3 T61445 2.2 0.36 TLE1 Transducin-like enhancer of split 1 (E(sp1) homolog) AI302139 2.5 0.30 TLE4 BCE-1 protein N67262 2.2 0.00 ZNF135 Zinc finger protein 135 AA421783 2.6 0.14 ZNF263 Zinc finger protein 263 AI298194 2.2 0.16 ZNF318 Zinc finger protein 318 AI292232 2.5 0.12 ZNF36 Zinc finger protein 36

H29322 3.1 0.38 CAMK1 Immunological signaling Calcium/calmodulin-dependent protein kinase I by guest on October 2, 2021 R53966 2.4 0.18 CHN1 Chimerin 1 AI018624 2.3 0.23 EIF2B5 Eukaryotic translation initiation factor 2B AA476508 2.5 0.24 ENPP2 Ectonucleotide pyrophosphatase/phosphodiesterase 2 (autotaxin) AA521025 3.5 0.16 GPS1 G protein pathway suppressor 1 AA504624 2.9 0.31 INPP5D Inositol polyphosphate-5-phosphatase AA598496 2.4 0.23 IQGAP1 IQ motif containing GTPase activating protein 1 W32272 3.3 0.49 IQGAP2 IQ motif containing GTPase activating protein 2 T99645 2.1 0.03 KCTD5 Potassium channel tetramerisation domain containing 5 AA485959 2.5 0.33 KRT7 Keratin 7 AA214574 2.4 0.46 MAP4K5 Mitogen-activated protein kinase kinase kinase kinase 5 W48815 2.0 0.43 MGC35521 Pellino 3␣ AI361112 2.8 0.31 MMP19 Matrix metalloproteinase 19 R09548 2.7 0.48 MOBKL2B MOBI, Mps One Binder kinase activator-like 2B AA146773 2.2 0.47 OAS1 2Ј,5Ј-Oligoadenylate synthetase 1 AA504468 3.3 0.32 PAFAHIB2 Platelet-activating factor acetylhydrolase AA521431 2.2 0.08 PFN1 Proﬁlin 1 AA682903 2.4 0.40 PIAS2 Protein inhibitor of activated STAT2 AI214426 2.3 0.21 PIR51 RAD51-interacting protein W32763 2.2 0.39 PPP1R12A Protein phosphatase 1, regulatory subunit 12A AA002166 2.5 0.01 PRKAG3 Protein kinase, AMP-activated, ␥3 noncatalytic subunit AA236171 2.0 0.34 PRKCA Protein kinase C␣ AI264246 2.6 0.30 PRKX Protein kinase, X-linked AA701448 2.4 0.30 PTPLA Protein tyrosine phosphatase-like, member a R93006 2.5 0.39 PTPLB Protein tyrosine phosphatase-like, member b AA448303 2.6 0.33 SH2BP1 SH2 domain binding protein 1 N21654 2.5 0.26 SLC16A6 Solute carrier family 16, member 6 AI302205 2.9 0.27 SLC16A7 Solute carrier family 16, member 7 AI004377 2.4 0.38 SLC30A7 Solute carrier family 30 (zinc transporter), member 7 AA490891 2.1 0.11 SLC35B1 Solute carrier family 35, member B1 AA705032 2.3 0.44 SLC38A4 Solute carrier family 38, member 4 N59534 2.2 0.45 SLC39A3 Solute carrier family 39 (zinc transporter), member 3 AA404709 2.1 0.03 SLC43A3 Solute carrier family 43, member 3 AA280137 2.2 0.36 SOCS1 Suppressor of cytokine signaling 1 AA504259 2.1 0.16 TRAF3 TNFR-associated factor 3 AA598826 2.5 0.34 TRAF4 TNFR-associated factor 4

W20462 3.6 0.74 COL23A1 Cell communication Collagen, type XXIII, ␣1 AA857098 2.4 0.17 COL5A2 Collagen, type V, ␣2 AI264291 2.5 0.17 ARHGAP25 Rho GTPase activating protein 25 AI139498 2.6 0.56 SGCD Sarcoglycan, ␦ AA481028 2.8 0.46 TRIP11 Thyroid hormone receptor interactor 11

* Annotation based on HUGO Gene Nomenclature Committee: Ͻwww.ensembl.org/Ͼ. The Journal of Immunology 3043

Table III. Genes down-regulated by CIDR1␣ (CIDR vs GST)

Fold Accession No. Change SD HUGO IDa Functional Categories Description

R93124 2.4 0.32 AKR1C1 Cell growth/apoptosis Aldo-keto reductase family 1, member C1 (dihydrodiol dehydrogenase 1; 20␣(3-␣)- hydroxysteroid dehydrogenase) H94857 2.6 0.09 BLOC1S1 Biogenesis of lysosome-related organelles complex-1, subunit 1 R45413 2.2 0.04 CDK4 Cyclin-dependent kinase 4 AA455476 2.1 0.04 DNCI2 Dynein, cytoplasmic, intermediate polypeptide 2 R89567 2.1 0.14 GC Group-speciﬁc component (vitamin D binding protein) T67442 2.4 0.60 HIC2 Hypermethylated in cancer 2 R42041 2.2 0.42 KCNMB4 Potassium large conductance calcium- activated channel, subfamily M, ␤member 4 AA088745 2.4 0.13 RAB6A RAB6A, member RAS oncogene family AA463625 2.3 0.38 RHEB GTP-binding protein Rheb (Ras homolog enriched in brain) AA961190 2.3 0.27 RP9 Retinitis pigmentosa 9 (autosomal dominant) (Pim-1 ass. protein) target Downloaded from protein for the PIM1 kinase AA489498 2.5 0.22 SAV1 Salvador homolog 1 (Drosophila) AA287261 2.2 0.59 TNFSF8 TNF (ligand) superfamily, member 8 AA856703 2.1 0.37 VPS35 Vacuolar protein sorting 35 (yeast)

N34028 2.4 0.04 FLJ11088 Transcription/translation GGA binding partner

AA400317 6.4 0.13 POLE4 Polymerase (DNA-directed), ⑀ 4 (p12 http://www.jimmunol.org/ subunit) AA464421 2.4 0.05 RNF110 Ring ﬁnger protein 110 N39584 2.2 0.29 SOX7 SRY (sex determining region Y)-box 7 AA426227 2.1 0.40 UMPS Uridine monophosphate synthetase (orotate phosphoribosyl Transferase and orotidine-5Ј-dicarboxylase)

AA285128 2.7 0.37 GNAS Immunological signaling Guanine nucleotide-binding protein G(S), ␣subunit (Adenylate cyclase-stimulating G␣ protein) by guest on October 2, 2021 AA521358 2.9 0.25 NCOA5 Nuclear receptor coactivator 5 AA600190 2.2 0.25 NUP205 Nucleoporin 205 kDa

N39161 2.1 0.18 CD36 Cell communication CD36 Ag collagen type 1 receptor, thrombospondin receptor W92417 5.0 0.17 COL21A1 Collagen, type XXI, ␣1 AA644657 5.4 0.88 HLA-A MHC class I, A

aAnnotation based on HUGO Gene Nomenclature Committee: Ͻwww.ensembl.org/Ͼ. malaria endemicity, could result in a long persistence and therefore lymphoma may relate to its capacity to augment the survival of GC augment its impact on B cell responses. B cells that carry translocations. Despite its ability to bind Igs, characterization and comparison of the gene expression profile induced by both CIDR1␣ and anti-Ig Acknowledgments activation demonstrated that there was a large difference in the We thank Drs. Alf Grandien and Lyda M. Osorio for interesting discus- signatures imposed by both stimuli. These results indicate that ei- sions and critical review of this manuscript. We also thank Mia Lo¨wbeer ther the signaling by CIDR1␣ proceeds via receptor(s) other than for expert technical assistance. Ig, or concomitantly through Ig with additional receptors. Alter- Disclosures natively, the different affinities of CIDR1␣ and anti-Ig, or possibly The authors have no financial conflict of interest. epitopes engaged, may lead to different signaling profiles. In conclusion, the combination of gene expression profiling with References phenotypic data and cell cycle analysis has enabled us to gain a 1. Abele, D. C., J. E. Tobie, G. J. Hill, P. G. Contacos, and C. B. Evans. 1965. better understanding of the mechanisms that drive malaria poly- Alterations in serum proteins and 19S antibody production during the course of clonal B cell activation. The observed activation of memory B induced malaria infections in man. Am. J. Trop. Med. Hyg. 14: 191–197. ␣ 2. Adu, D., D. G. Williams, I. A. Quakyi, A. Voller, Y. Anim-Addo, cells mediated by CIDR1 , has to be considered for a rational A. A. Bruce-Tagoe, G. D. Johnson, and E. J. Holborow. 1982. Anti-ssDNA and design of a malaria vaccine, because it could lead to inappropriate antinuclear antibodies in human malaria. Clin. Exp. Immunol. 49: 310–316. Ab production and hyperglobulinemia. The polyclonal B cell ac- 3. McGregor, I. A., H. M. Giles, J. H. Walter, and A. H. Davies. 1956. Effects of heavy and repeated malaria infections of Gambian infants and children. Br. tivation and augmented survival induced by CIDR1␣ is of rele- Med. J. 2: 686–691. vance for understanding the mechanisms behind the increased risk 4. Greenwood, B. M., and R. M. Vick. 1975. Evidence for a malaria mitogen in human malaria. Nature 257: 592–594. of Burkitt’s lymphoma in malaria endemic areas (36). A possible 5. Bates, I., and G. Bedu-Addo. 1997. Chronic malaria and splenic lymphoma: clues role of malaria infection in increasing the risk of endemic Burkitt’s to understanding lymphoma evolution. Leukemia 11: 2162–2167. 3044 MALARIA AND POLYCLONAL B CELL ACTIVATION

6. Scholander, C., J. Carlson, P. G. Kremsner, and M. Wahlgren. 1998. Extensive 21. Shih, T. A., M. Roederer, and M. C. Nussenzweig. 2002. Role of antigen receptor immunoglobulin binding of Plasmodium falciparum-infected erythrocytes in a affinity in T cell-independent antibody responses in vivo. Nat. Immunol. 3: group of children with moderate anemia. Infect. Immun. 66: 361–363. 399–406. 7. MacLennan, I. C. 1994. Germinal centers. Annu. Rev. Immunol. 12: 117–139. 22. Rathmell, J. C. 2004. B-cell homeostasis: digital survival or analog growth? 8. Chen, Q., A. Heddini, A. Barragan, V. Fernandez, S. F. Pearce, and M. Wahlgren. Immunol. Rev. 197: 116–128. 2000. The semiconserved head structure of Plasmodium falciparum erythrocyte 23. Glimcher, L. H., and H. Singh. 1999. Transcription factors in lymphocyte devel- membrane protein 1 mediates binding to multiple independent host receptors. opment—T and B cells get together. Cell 96: 13–23. J. Exp. Med. 192: 1–10. 24. Clevers, H. C., and R. Grosschedl. 1996. Transcriptional control of lymphoid 9. Flick, K., C. Scholander, Q. Chen, V. Fernandez, B. Pouvelle, J. Gysin, and development: lessons from gene targeting. Immunol. Today 17: 336–343. M. Wahlgren. 2001. Role of nonimmune IgG bound to PfEMP1 in placental 25. Fitzsimmons, D., and J. Hagman. 1996. Regulation of gene expression at early malaria. Science 293: 2098–2100. stages of B-cell and T-cell differentiation. Curr. Opin. Immunol. 8: 166–174. 10. Donati, D., L. P. Zhang, Q. Chen, A. Chene, K. Flick, M. Nystrom, M. Wahlgren, and M. T. Bejarano. 2004. Identification of a polyclonal B-cell activator in Plas- 26. Muzio, M., D. Bosisio, N. Polentarutti, G. D’Amico, A. Stoppacciaro, modium falciparum. Infect. Immun. 72: 5412–5418. R. Mancinelli, C. van’t Veer, G. Penton-Rol, L. P. Ruco, P. Allavena, and 11. Romagnani, S., M. G. Giudizi, G. del Prete, E. Maggi, R. Biagiotti, A. Mantovani. 2000. Differential expression and regulation of Toll-like receptors F. Almerigogna, and M. Ricci. 1982. Demonstration on protein A of two distinct (TLR) in human leukocytes: selective expression of TLR3 in dendritic cells. immunoglobulin-binding sites and their role in the mitogenic activity of Staph- J. Immunol. 164: 5998–6004. ylococcus aureus Cowan I on human B cells. J. Immunol. 129: 596–602. 27. Dorfman, J. R., P. Bejon, F. M. Ndungu, J. Langhorne, M. M. Kortok, 12. Widders, P. R. 1990. Fc Receptors and the Pathogenesis of Bacterial Infections B. S. Lowe, T. W. Mwangi, T. N. Williams, and K. Marsh. 2005. B cell memory in Animals. Academic, Orlando, FL. to 3 Plasmodium falciparum blood-stage antigens in a malaria-endemic area. 13. Langone, J. J. 1982. Protein A of Staphylococcus aureus and related immuno- J. Infect. Dis. 191: 1623–1630. globulin receptors produced by streptococci and pneumonococci. Adv. Immunol. 28. Hu, P., Z. Han, A. D. Couvillon, and J. H. Exton. 2004. Critical role of endog- 32: 157–252. enous Akt/IAPs and MEK1/ERK pathways in counteracting endoplasmic retic- 14. Leonetti, M., J. Galon, R. Thai, C. Sautes-Fridman, G. Moine, and A. Menez. ulum stress-induced cell death. J. Biol. Chem. 279: 49420–49429. 1999. Presentation of antigen in immune complexes is boosted by soluble bac- 29. Green, D. R., and D. W. Scott. 1994. Activation-induced apoptosis in lympho- terial immunoglobulin binding proteins. J. Exp. Med. 189: 1217–1228. cytes. Curr. Opin. Immunol. 6: 476–487. Downloaded from 15. Daniel-Ribeiro, C., J. de Oliveira-Ferreira, D. M. Banic, and B. Galvao-Castro. 30. Yi, A. K., M. Chang, D. W. Peckham, A. M. Krieg, and R. F. Ashman. 1998. 1989. Can malaria-associated polyclonal B-lymphocyte activation interfere with CpG oligodeoxyribonucleotides rescue mature spleen B cells from spontaneous the development of anti-sporozoite specific immunity? Trans. R. Soc. Trop. Med. apoptosis and promote cell cycle entry. J. Immunol. 160: 5898–5906. Hyg. 83: 289–292. 31. Piatelli, M. J., D. Tanguay, T. L. Rothstein, and T. C. Chiles. 2003. Cell cycle 16. Chen, Q., A. Barragan, V. Fernandez, A. Sundstrom, M. Schlichtherle, A. Sahlen, control mechanisms in B-1 and B-2 lymphoid subsets. Immunol. Res. 27: 31–52. J. Carlson, S. Datta, and M. Wahlgren. 1998. Identification of Plasmodium fal- 32. Wahlgren, M., J. S. Abrams, V. Fernandez, M. T. Bejarano, M. Azuma, M. Torii, ciparum erythrocyte membrane protein 1 (PfEMP1) as the rosetting ligand of the M. Aikawa, and R. J. Howard. 1995. Adhesion of Plasmodium falciparum-in- malaria parasite P. falciparum. J. Exp. Med. 187: 15–23. fected erythrocytes to human cells and secretion of cytokines (IL-1-␤, IL-1RA, 17. Pound, J. D., and J. Gordon. 1997. Maintenance of human germinal center B cells http://www.jimmunol.org/ IL-6, IL-8, IL-10, TGF␤, TNF␣, G-CSF, GM-CSF). Scand. J. Immunol. 42: in vitro. Blood 89: 919–928. 626–636. 18. Shi, Y., K. Agematsu, H. D. Ochs, and K. Sugane. 2003. Functional analysis of human memory B-cell subpopulations: IgDϩCD27ϩ B cells are crucial in sec- 33. Bell, G. I. 1974. Model for the binding of multivalent antigen to cells. Nature ondary immune response by producing high affinity IgM. Clin. Immunol. 108: 248: 430–431. 128–137. 34. Armitage, R. J., and M. R. Alderson. 1995. B-cell stimulation. Curr. Opin. Im- 19. Bernasconi, N. L., E. Traggiai, and A. Lanzavecchia. 2002. Maintenance of se- munol. 7: 243–247. rological memory by polyclonal activation of human memory B cells. Science 35. George, J., S. J. Penner, J. Weber, J. Berry, and J. L. Claflin. 1993. Influence of 298: 2199–2202. membrane Ig receptor density and affinity on B cell signaling by antigen: impli- 20. Tangye, S. G., D. T. Avery, E. K. Deenick, and P. D. Hodgkin. 2003. Intrinsic cations for affinity maturation. J. Immunol. 151: 5955–5965. differences in the proliferation of naive and memory human B cells as a mech- 36. Greenwood, B. M., J. H. Playfair, and G. Torrigiani. 1970. Burkitt lymphoma and anism for enhanced secondary immune responses. J. Immunol. 170: 686–694. malaria. Lancet 2: 418–422. by guest on October 2, 2021