Identifying broadly neutralizing antibodies against infectious agents such as influenza A viruses, HIV, and Plasmodium falciparum that display impressive degrees of antigenic variation is a major focus of investigators developing therapeutics and vaccines for pathogens of importance in public health (Corti and Lanzavecchia, 2013).  In a previous post, I discussed one study (Klein et al., 2013) illustrating the sorts of unanticipated types of mutations found for broadly neutralizing antibodies against HIV.  Lanzavecchia and colleagues have now identified antibodies reactive with antigens encoded by different isolates of Plasmodium falciparum and expressed on infected erythrocytes (Nature, 2015).  They find an unexpected source for the heavy chain variable domain amino acid sequences that confer the broad anti-malarial reactivity against proteins in the RIFIN family.

These investigators began their search for antibodies with broad reactivity for malaria antigens by developing a clever screening approach that mixes old and new technologies.  They mixed plasma pools, each of which was obtained from five individuals living in a malaria-endemic region of Kenya, with a pool of erythrocytes from patients infected by three different malaria isolates.  Each of the erythrocyte samples was stained with a different DNA dye displaying a different color.  Individual plasma samples from pools causing multi-color agglutinates were then further tested.

Two donors, C and D, whose plasma could form mixed agglutinates of erythrocytes infected with as many as eight different malaria isolates were then selected to isolate and immortalize IgG+ memory B cells.  Monoclonal antibodies (mAbs) secreted by these immortalized B cells were then tested in the mixed agglutination assay.  Most of these antibodies from both donors were reactive with multiple parasite isolates, but a smaller number were able to react with only one isolate.  These reactivities were confirmed using flow cytometric staining of erythrocytes infected by different malaria isolates.

Next the structural basis of the broad reactivity of the antibodies from these two donors was investigated by determining the amino acid sequences of the variable domains from the corresponding B cells.  This analysis revealed a major surprise.

The antibodies with narrow reactivities for malaria erythrocyte antigens exhibited the standard organization of the heavy chain V domain with standard sizes for those portions encoded by the V, D, and J gene segments.  However, 14 antibodies from one donor and 13 from the second donor displayed an unprecedented insertion of more than 100 amino acids between the V and the DJ segments.  Equally amazing, the evidence strongly suggests that the insert originated in a sequence from a gene residing on chromosome 19.  Since the immunoglobulin heavy chain locus is located on chromosome 14, this insertion, from the portion of the LAIR1 gene encoding the extracellular collagen-binding domain represents an interchromosomal transfer of sequence into a rearranged heavy chain gene.  The mechanisms by which antibodies were believed to diversify in humans have not previously included insertions from non-immunoglobulin genes on other chromosomes.

The LAIR1 inserts exhibited mutations with two key functional effects.  One group of mutations eliminated the collagen binding activity of the original LAIR1 donor sequence.  The second group of mutations conferred increased binding for malaria-infected red cells.

Also noted by the authors, the donor D antibodies also included short stretches of amino acid sequence apparently derived from chromosome 13.  While the broadly reactive donor C antibodies used three different light chain V domains, the donor D antibodies all used the same light chain V domain.  The authors were able to create clear phylogenies for the B cell clones that produced broadly reactive antibodies in both donors C and D enabling the appreciation of the sequential somatic mutations in the heavy and light chain variable domains including those in the inserted sequences derived from the LAIR1 gene.

Comparison of proteins immunoprecipitated by one of the broadly reactive mAbs (MGD21; originating in donor D) from erythrocytes reactive or not reactive with the mAb revealed that two particular members (PF3D7_1400600 and PF3D7_1040300) of the RIFIN family were recognized by the mAb.  The authors confirmed these reactivities for MGD21 by inserting genes for both proteins into Chinese hamster ovary (CHO) cells and demonstrating binding by the mAb to these CHO cells but not to CHO cells displaying an irrelevant member of the RIFIN family not demonstrated to bind MGD21.

In order to assess the roles of the various components of the heavy and light chain variable domains to recognition of the malaria proteins from different isolates, portions of the MGD21 amino acid sequence were replaced with heavy chain V, J, or light chain amino acid sequences derived from antibodies exhibiting unrelated binding specificities.  None of these replacements abrogated binding to infected erythrocytes as long as the LAIR1 region, modified by the mutations previously noted, was intact.  Deleting the LAIR1 sequence or reverting it to the original genomic sequence found on chromosome 19 that prevailed prior to somatic hypermutation eliminated binding to the malaria antigens.

While MGD21 and another broadly reactive antibody from donor C did not directly prevent infection of erythrocytes by malaria parasites or prevent expression of the RIFIN family proteins, they did effectively agglutinate and opsonize infected erythrocytes for monocytes.  These mechanisms could potentially contribute to destruction of infected erythrocytes and to immunity against malaria.

Further research will be necessary to determine if the RIFIN family proteins recognized by the antibodies from donors C and D are potential targets for active or passive vaccination.  However, given that there are up to 150 genes encoding members of the RIFIN family of variant malaria antigens and that these can be expressed clonally in erythrocytes, it is not clear whether eliciting antibodies like MGD21 through immunization or generating them via gene expression vectors will be sufficient for controlling malaria infection.

The authors provide solid evidence from their analysis so far that the lymphocyte-specific enzymes typically involved in rearrangement of antibody gene segments, called recombination activating gene-1 (RAG-1) and recombination activating gene-2 (RAG-2), are not responsible for this case of interchromosomal transfer of genetic information.  In particular, the persistence of two allelic copies of LAIR1 in the B cells synthesizing the broadly reactive antibodies containing mutated LAIR1 sequences strongly suggests that a “cut and paste” mechanism of the sort associated with RAG-1 and RAG-2 is not relevant.  Based on the evidence in hand, the authors suggest the possibility that reverse transcription could be involved in generating the LAIR1 insertion, a mechanism that would leave the genomic LAIR1 donor sequence intact.  Such a mechanism has not previously been found to contribute to antibody gene diversification in humans. It seems fair to conclude that the precise mechanistic details of the transfer of genetic information from chromosome 19 (or 13), to chromosome 14 in the heavy chain locus remain to be established through additional investigation.

From the broader perspective of evolution, we once again see a powerful illustration of Principle of Radical Evolutionary Indifference.  Under the intense selection pressure imposed by P. falciparum on both humans as organisms and on their B lymphocytes, an exceptional genetic event, interchromosomal donation of a large nucleotide sequence to a rearranged heavy chain gene that confers unique antigen-binding capability, has been captured and preserved.  Perhaps this sort of event has occurred before, but certainly it is noteworthy that this mechanism of antibody amino acid sequence diversity generation had not been previously documented in humans in the many decades of intense study of antibody gene diversity generation.  And, in this instance, the transferred sequence did not merely influence the antigen recognition function of the affected antibodies but was central to it. Thus, as argued by Godfrey-Smith in his recently published book on the philosophy of biology, it appears that the combination of “random” mutation and selective (i.e. exceptionally non-random) preservation can be impressively creative in the generation of new phenotypes (Godfrey-Smith, 2013).

References

Corti D, Lanzavecchia A. Broadly neutralizing antiviral antibodies. Annu Rev Immunol. 2013;31:705-42. doi: 10.1146/annurev-immunol-032712-095916. Epub 2013 Jan 16. Review. PubMed PMID: 23330954.

Greenspan N. Boundary-breaking evolution via B lymphocyte clonal selection in response to HIV-1. September 28, 2013

Boundary-Breaking Evolution via B Lymphocyte Clonal Selection in Response to HIV-1

Klein F, Diskin R, Scheid JF, Gaebler C, Mouquet H, Georgiev IS, Pancera M, Zhou T, Incesu RB, Fu BZ, Gnanapragasam PN, Oliveira TY, Seaman MS, Kwong PD, Bjorkman PJ, Nussenzweig MC. Somatic mutations of the immunoglobulin framework are generally required for broad and potent HIV-1 neutralization. Cell. 2013 Mar  28;153(1):126-38. doi: 10.1016/j.cell.2013.03.018. PubMed PMID: 23540694.

Tan J, Pieper K, Piccoli L, Abdi A, Foglierini M, Geiger R, Maria Tully C,
Jarrossay D, Maina Ndungu F, Wambua J, Bejon P, Silacci Fregni C,
Fernandez-Rodriguez B, Barbieri S, Bianchi S, Marsh K, Thathy V, Corti D,
Sallusto F, Bull P, Lanzavecchia A. A LAIR1 insertion generates broadly reactive
antibodies against malaria variant antigens. Nature. 2015 Dec 23. doi:
10.1038/nature16450. [Epub ahead of print] PubMed PMID: 26700814.

Godfrey-Smith, Peter. Philosophy of Biology. Princeton University Press, Princeton, NJ, 2013.