The Missing Link(er): A Return to Symmetry in Antigen Receptor Signaling?

The mounting of an effective adaptive immune response requires important biochemical events that are initiated in response to engagement of the antigen receptors of B and T lymphocytes. Understandably then, there have been intensive efforts to discover the molecular events that constitute antigen receptor signaling. As our understanding has grown, one of the striking features of antigen receptor signaling has been the broad parallels in signaling pathways utilized by the T cell antigen receptor (TCR) and B cell antigen receptor (BCR), despite the very different natures of these two receptors.

Ultimately, engagement of either receptor leads to the activation of the Ras–Erk (extracellular-regulated kinase) pathway, as well as phospholipase C–γ (PLC–γ ), and the subsequent rise of intracellular Ca²⁺ {[Ca²⁺ ]_i }—events that are critical for the subsequent activation of B and T lymphocytes. The biochemical processes leading to these distal events are also quite similar in the two cell types (1, 2) . Both the TCR and the BCR begin intracellular signaling by initiating the phosphorylation of specific tyrosine residues within a vital signaling motif known as the ITAM (immunoreceptor tyrosine-based activation motif). Phosphorylation of ITAMs is catalyzed by protein tyrosine kinases (PTKs) of the Src family—Lck and Fyn in the case of the TCR, and Lyn, Fyn, and Blk in the case of the BCR. Upon phosphorylation, the ITAMs serve as high affinity binding sites for a second family of PTKs, the Syk–ZAP family, where Syk is recruited to the BCR, and ZAP-70 [TCRζ (zeta) chain–associated protein of 70-kDa] to the TCR. This recruitment leads to the activation of these kinases. Notably, this general pathway is also shared by the high affinity IgG and IgE receptors (Fcγ RI and Fcε RI, respectively) present on myeloid cells (3) .

Just how the activation of ZAP-70 leads to activation of the distal signaling pathways in T cells was substantially advanced by the discovery of two proteins that serve as substrates of ZAP-70: SLP-76 [Src homology 2 (SH2) domain-containing leukocyte protein of 76-kDa] and LAT (linker for activation of T cells) (4–6) . Both of these proteins serve as linker/adapter proteins that act together to translate the signaling language specific to the TCR into the general signaling language that is common to many cell surface receptors (7) . Absence of either SLP-76 or LAT leads to severely compromised signaling through the TCR and developmental arrest in thymocytes. Biochemically, SLP-76 acts as a cytosolic linker/adapter protein that coordinates interactions between Vav, Nck, and p21-activated protein kinase (PAK) (8) , as well as between PLC–γ 1 and LAT (9, 10) , and has been implicated in Erk activation, Ca²⁺ mobilization, and cytoskeletal restructuring. LAT, a type three transmembrane protein, also acts as a linker/adapter protein, and binds directly to numerous proteins including: Gads (Grb2-related adaptor downstream of Shc), Grap (Grb2-related adapter protein), Grb2, PLC–γ 1, 3BP2 (an Abl SH3-domain-interacting protein), Shb (an SH2 domain-containing adapter protein), and Itk [interleukin-2- (IL-2)-inducible T-cell kinase]. LAT also indirectly associates with phosphatidylinositol 3-kinase (PI3K), mammalian son of sevenless (mSOS), c-Cbl, Vav and SLP-76 (11) . LAT is an essential participant in TCR-stimulated activation of the Ras–Erk pathway as well as PLC–γ 1–mediated increases in [Ca²⁺ ]_i . Importantly, the ability of LAT to support signaling to these pathways requires its specific targeting to biophysically distinct lateral domains in the plasma membrane termed lipid rafts (12, 13) .

However, it is with SLP-76 and LAT that the symmetry between the TCR and BCR signaling pathways began to break down. Despite the critical role that LAT clearly plays in TCR signaling, LAT itself is not expressed in B cells, and extensive efforts failed to identify any LAT homologs in B cells. On the other hand B cells do express what appears to be a homolog of SLP-76, termed BLNK [B cell linker protein, alternatively named SLP-65 or BASH (B cell adapter containing SH2 domain)]. BLNK shares 30% identity with SLP-76 and is a substrate for Syk (14–16) . Because BLNK can bind Grb2 and PLC–γ as well as Vav-1—the first two being associated with LAT binding in T cells, and the latter with SLP-76 binding—it was widely hypothesized that BLNK, by itself, could fulfill the function in B cells that requires both SLP-76 and LAT in T cells.

This hypothesis was never entirely satisfactory, however, because BLNK, a cytosolic protein, has no intrinsic capacity to partition into the lipid rafts. This was problematic for two reasons. First of all, the ability of LAT to support signaling was so critically dependent upon lipid raft localization; therefore, it seemed unlikely that a protein that did not share this capacity could mimic LAT biochemically. Secondly, it had already become apparent that BCR signaling was at least as dependent upon the appropriate partitioning of signaling proteins between the bulk membrane and lipid rafts as was TCR signaling (17) , implying that a LAT-like protein might be important in BCR signaling. With the recent independent cloning of a new B cell–expressed LAT-like signaling molecule by the Zhang and Horejsí laboratories (18, 19), it seems that BLNK was miscast after all.

The Horejsí laboratory was the first to report the cloning of the new protein (20) . Reasoning that a non–T cell LAT homolog would selectively partition to the lipid rafts, this group took a classical biophysical approach, and purified an unidentified, highly acidic (like LAT), 30-kDa phosphoprotein from the lipid rafts of myeloid cells. Mass spectrometric analyses of tryptic fragments of the protein indicated that it is encoded by a previously cloned but uncharacterized gene termed WBSCR5 , present on human chromosome 7 (7q11.23). The WBSCR5 cDNA encodes a polypeptide of 243 amino acids with a predicted molecular weight of 26,550 Da. Because WBSCR5 mRNA is expressed in B cells, NK cells, mast cells, and monocytes, the new protein was christened NTAL (non–T cell activation linker).

At the same time, Zhang and colleagues took a completely different approach, using bioinformatics, to identify candidates for the B cell equivalent of LAT (19) . They screened public DNA sequence databases for cDNA that met each of their three criteria. The cDNA should encode for a protein that 1) has four or more YXN (where X is any amino acid), Grb2 SH2 domain-binding motifs (LAT has five); 2) possesses a transmembrane domain; and 3) has a palmitoylation site. Only one cDNA met all of these criteria, the one encoded by the WBSCR5 gene. Finding expression in B cell-enriched, but not T cell–enriched splenocytes, they gave the protein the name LAB (linker for activation of B cells).

So is NTAL/LAB the missing LAT-like linker in B cell antigen receptor signaling? In terms of its primary sequence, it does share a similar domain structure with LAT, but NTAL/LAB shares very little overall sequence identity. As in LAT, the sequence for NTAL/LAB predicts an N-terminal transmembrane domain, with a short extracellular projection and an extended intracellular tail. The cytoplasmic juxtamembrane region of human NTAL/LAB also has the sequence ₂₄ LCVRC₂₈ , which is very similar to the ₂₅ LCVHC₂₉ sequence in human LAT that serves as the site of palmitoylation. In addition there are eight tyrosine residues that are conserved in the cytoplasmic tail of human and murine NTAL/LAB, the same number of cytoplasmic tyrosines that are conserved between human, murine, rat and bovine LAT (11) . Five of these tyrosines are within Grb2-binding YXN motifs, as is the case with LAT (Table 1⇓).

What about the ability of NTAL/LAB to serve as a lipid raft–targeted linker/adapter protein? NTAL/LAB is palmitoylated and is found primarily in the lipid raft fraction of the plasma membrane, and is rapidly tyrosine phosphorylated following BCR or Fc receptor (FcR) engagement, forming molecular complexes with Grb2, Sos1, and Gab1 in B cells, and Grb2, Sos1, Gab1, and c-Cbl in myeloid cells. Stable expression of NTAL/LAB in LAT-deficient JCaM2.5 Jurkat T cells reconstitutes Erk activation and Ca²⁺ mobilization in these cells, but only weakly. Similarly, retroviral transduction of NTAL/LAB into LAT-negative bone marrow cells partially rescues the ability of these cells to develop into normal, mature, single-positive (e.g., CD4⁺ or CD8⁺ ) thymocytes upon adoptive transfer into irradiated host mice. However, splenocytes from these mice could not increase their production of IL-2 in response to TCR stimulation, unlike LAT-rescued mice.

Conspicuous in their absence as NTAL/LAB-associating proteins are BLNK and PLC–γ . Despite intensive efforts to demonstrate a receptor-stimulated association between NTAL/LAB and BLNK or PLC–γ , none was found by either group. This contrasts sharply with LAT, which can be easily immunoprecipitated with SLP-76 and PLC–γ This might suggest that NTAL/LAB is not required for BCR-stimulated activation of PLC–γ . However, the Zhang lab observed that small interfering–RNA (siRNA)-mediated reduction of NTAL/LAB protein expression levels in A20 B cells results in marked inhibition of BCR-stimulated Ca²⁺ flux, supporting the argument that NTAL/LAB does indeed participate in PLC–γ activation.

So is NTAL/LAB the B cell homolog of LAT? Well, yes and no. The remarkable conservation of the exon–intron organization of the two genes and overall conservation of domain structure strongly suggests that these two proteins arose from divergent evolution from a duplicated ancestral gene. LAT is almost exclusively expressed in T cells, whereas NTAL/LAB is more widely expressed, including B cells but excluding T cells. Furthermore NTAL/LAB shares many of the biochemical and cell biological hallmarks of LAT, and is able to partially rescue TCR signaling in LAT-negative T cells; however, it is important to stress that NTAL/LAB permits only a partial rescue. Similarly, BLNK deficiency cannot be compensated by SLP-76 alone, rather requiring SLP-76 and LAT together (21). From the available data it is clearly an oversimplification to consider NTAL/LAB to be the functional equivalent of LAT. A more reasonable hypothesis is that LAT and SLP-76 together and NTAL/LAB and BLNK together act in tandem as functionally equivalent units for transducing agonist receptor specific signaling events to the general downstream signaling pathways. Under this scheme LAT and NTAL/LAB would have divergently evolved from a duplicated common ancestral gene. At the same time, LAT and SLP-76 (and NTAL/LAB and BLNK) would have coevolved to couple the TCR (and the BCR) to the same set of distal signaling pathways. Whether the divergence in primary sequence and in protein binding partners between LAT and NTAL/LAB will provide entrée for the selective pharmacologic manipulation of the pathways downstream of these proteins and thereby provide an additional means of independently manipulating T and B cell activation status awaits testing.

• © American Society for Pharmacology and Experimental Theraputics 2003

Ronald L. Wange, PhD , is Head of the T Lymphocyte Signaling Unit, in the Laboratory of Cellular and Molecular Biology at the National Institute on Aging. When not studying the signaling pathways that regulate T lymphocyte activation he can often be found at the local pub torturing friends and innocent bystanders with truly terrible puns.