LONDON — In infancy, an encounter with the Epstein-Barr virus frequently passes unnoticed. Those unlucky enough not to meet with the virus until early adulthood can expect a dose of glandular fever, otherwise known as infectious mononucleosis. Starting like a bad case of influenza, the extreme fatigue which follows may mean a couple of months off college or work.

But for an unfortunate few — one or two males per million — infection with the Epstein-Barr virus heralds the onset of a variety of severe immunological abnormalities. These individuals may experience infectious mononucleosis so severe that it kills them, or they may develop hypogammaglobulinemia (lack of infection-fighting antibodies in the blood), or malignant lymphoma such as non-Hodgkin's lymphoma. The average age at which these problems start is 2.5 years, and 70 percent of those affected die by the age of 10.

This condition is called X-linked lymphoproliferative disease (XLP), also known as Duncan disease. Now a group of researchers led by David Bentley of the Sanger Centre, in Cambridge, U.K., have reported that they have identified the gene which, when mutated, causes XLP.

The finding will make diagnosis of affected boys in families with a history of this condition much easier. It may also lead to better understanding of how the varying immune response to infectious mononucleosis is regulated. Alison Coffey, along with Bentley and colleagues in several collaborating groups, report their results in a paper in the October issue of Nature Genetics, titled "Host response to EBV infection in X-linked lymphoproliferative disease results from mutations in an SH2-domain encoding gene."

In a simultaneous publication in Nature, Joan Sayos and colleagues from the Beth Israel Deaconess Medical Center at Harvard Medical School, in Boston, report that they, too, have found the gene for XLP. Their paper is titled "The X-linked lymphoproliferative-disease gene product SAP regulates signals induced through the co-receptor SLAM."

Robert Brooksbank, research associate at the Sanger Centre and joint corresponding author of the paper in Nature Genetics, said linkage studies had already shown that the locus of the XLP gene must be around the Xq25 region, which is on the long arm of the human X chromosome. He told BioWorld International: "Three patients had been reported with a deletion of some of the DNA from this region. These were overlapping deletions, which made it likely that the reason they had the disease was because the gene was deleted in these patients."

The team obtained DNA from one of these patients and checked whether certain genetic markers were present or absent in it. They were able to construct a contig of yeast artificial chromosomes, and eventually bacterial clones, across the critical region defined by the deletion.

Using exon trapping to identify genes present in the deleted DNA, Bentley's team found that the protein products of one gene were homologous to a family of proteins involved in immune regulation. "This gave us the first clue that we might have found a potential candidate gene," Brooksbank said, "and we subsequently used sequence derived from the exon traps to screen cDNA libraries and identify clones corresponding to the full length of the gene."

They next checked the DNA from 16 unrelated XLP patients and 50 normal male controls, to see if they could find mutations in the candidate gene. They found mutations in nine of the patients, and none in 49 of the controls. (The remaining control had a point mutation which did not change the amino acid sequence of the protein.) "The nature of the mutations we found was conclusive enough to allow us to be confident that we had identified the gene that was responsible for XLP," Brooksbank said.

Bentley's group called the gene it had found SH2D1A. Further studies showed the gene was predominantly expressed in tissues that play a role in the immune system, particularly in the thymus. Brooksbank added: "We predicted that the protein was probably going to be involved in regulating cell signaling between lymphocytes, and our next question was going to be to find out what the gene product binds to."

Conveniently, the paper by Sayos and others has the answer. This group started tackling the problem from the opposite direction. It was studying a molecule called SLAM, which stands for signaling lymphocyte-activation molecule. This is present on both B and T lymphocytes and plays a role in signaling between these two types of cells.

This interaction is key to the immune response controlling infectious mononucleosis. Epstein-Barr virus initially infects the throat but then invades the B cells, which are responsible for manufacturing antibodies, causing them to multiply rapidly. In infectious mononucleosis, the expansion of B lymphocytes is controlled by T lymphocytes, which recognize the infected cells as foreign and trigger a massive inflammatory response. In XLP, this control is lacking, so clones of both B and T cells expand.

Sayos and colleagues were looking for proteins that interact with SLAM — and identified a protein called SLAM-associated protein, or SAP, which turns out to be the product of SH2D1A. Therefore, the answer to the question posed by Brooksbank and his colleagues was SLAM. In Nature, Sayos' group writes: "Our results support a model in which SAP controls signal transduction pathways that are initiated by interactions between SLAM molecules on the interface between T and B cells."

Michel Sadelain, of the Memorial Sloan-Kettering Cancer Center, in New York, and Elliott Kieff, of Harvard Medical School and the Brigham and Women's Hospital, in Boston, comment on the findings in a "News and Views" article in the October issue of Nature Genetics, titled "Why commonplace encounters turn to fatal attraction." They write: "It is . . . plausible that mutations in SH2D1A deregulate the formation of important signaling complexes, resulting in a massive but ineffective immune response. In this model of XLP, the excessive stimulation induced by Epstein-Barr virus-infected B cells and other recruited antigen-presenting cells overwhelms T cells that fail to control their metabolic response to activation."

The Sanger Centre group is now planning to search for additional mutations in patients with XLP, and hopes to be able to link particular mutations to certain phenotypes of the disease. It is collaborating with other researchers to make a knockout mouse which does not have SH2D1A, to produce a mouse model of the disease. The scientists will also be sequencing the homologous region in the mouse and comparing the human and murine sequences.

Sadelain and Kieff predict, now that the gene for XLP is known, the disease could lend itself to gene therapy. They suggest that the normal gene for SH2D1A could be introduced into autologous hematopoietic stem cells, for example. *