Role of the Kaposi's sarcoma-associated herpesvirus K15 SH3 binding site in inflammatory signaling and B-cell activation

Abstract

The Kaposi's sarcoma-associated herpesvirus (KSHV) contains several open reading frames (ORFs) that encode proteins capable of initiating and modulating cellular signaling pathways. Among them is ORF K15, encoding a 12-transmembrane-spanning protein with a cytoplasmic C-terminal domain. Through conserved binding motifs, such as Src homology 2 (SH2) and SH3 binding sites, K15 interacts with cellular proteins, activates the NF-kappaB, MEK/Erk, and Jun N-terminal protein kinase (JNK) pathways, and induces the expression of several inflammatory and angiogenic genes. In this study, we investigated the role of an SH3 domain binding site centered on a PPLP motif in K15. We screened libraries of cellular SH3 domains to identify signaling molecules interacting with the KSHV PPLP motif. We found its affinities for two Src kinase family members, Lyn and Hck, to exceed those of other viral proteins. While the SH2 binding motif YEEV is essential for the inflammatory response induced by KSHV K15, recruitment of Lyn and Hck to the K15 PPLP motif seems to be dispensable for this inflammatory response. However, the PPLP motif is essential for the decrease in B-cell receptor-mediated signaling induced by K15, as measured by calcium mobilization assays.

FIG. 1.

The SH3 domains of Hck and Lyn bind to the putative SH3 binding sites of K15-P and K15-M. (a) Overlapping peptides corresponding to the cytoplasmic region of K15-P or K15-M and control peptides were immobilized via their C termini on cellulose membranes and were incubated with recombinant GST alone (left panels) or GST Hck SH3 or GST Lyn SH3 protein (right panels). Spots A1 to E19 (K15-P) and G1 to K5 (K15-M) represent peptides of 20 amino acids, with 1 amino acid slide per peptide to cover the entire cytoplasmic regions of the two K15 proteins. Spots M1 to 21 are peptides derived from proteins known to bind to Hck and Lyn as indicated in the list. Spots M22 to N9 (K15-P) and N10 to 27 (K15-M) represent truncation peptides of the K15 proteins to map the core binding regions within the K15 SH3-B sites (printed in boldface in panel b). Spots N28 to P27 (K15-P) and P28 to R30 represent an alanine scan through the identified K15 motifs (printed in boldface in panel b). (b) Amino acid sequences of the entire cytoplasmic regions of K15-P and K15-M. The identified SH3-B site within the cytoplasmic region of each K15 protein is printed in boldface. (c) Relative binding of individual SH3 domain-displaying phages to GST-K15 (P- and M-type)-coated wells. Wells were coated with recombinantly expressed fusion proteins consisting of GST fused to the cytoplasmic tail of K15-P or K15-M. Individual SH3 domain-displaying phages were incubated with GST fusion protein-coated wells, and the strength of binding was estimated based on the increase (1- to 10,000-fold) of phages bound compared with that in plates coated with plain GST. The table indicates semiquantitative binding strengths of the individual SH3-displaying phages (left vertical column) to GST-K15-P or GST-K15-M fusion proteins (top horizontal column). Binding strengths: +++, strong (>1,000-fold enrichment); ++, good (>100-fold enrichment); +, moderate (>10-fold enrichment); (+)/−, weak/none (<10-fold enrichment).

FIG. 2.

Measurement of binding strengths among K15 proteins and the Src kinases Hck and Lyn by isothermal titration calorimetry (ITC). (a) ITC experiments were performed as described in Materials and Methods. Amidated peptides consisting of 20 amino acids of K15-P and K15-M (printed in boldface in Fig. 1b) were used to determine binding affinities to Src kinases Hck and Lyn. The 20-amino-acid peptide of the HIV Nef mutant was included as a positive control. (b) Binding strengths (K_D values), as calculated with the software ORIGIN (version 5.0), and standard deviations (SD) of the K15-P, -M, or HIV Nef peptides with the SH3 domains of the Src kinases Lyn and Hck are indicated.

FIG. 3.

Binding of Src kinases to K15 mutants. HEK 293-T cells were transiently transfected with expression vectors for myc-tagged PTKs Lyn, Src, and Hck. Cell lysates were incubated overnight at 4°C with the GST-K15-P and GST-K15-M exon 8 (ex8) wild-type (WT) proteins, and their respective mutants (YEEV, PPLP, YEEV PPLP, and YASI), immobilized to glutathione-Sepharose beads. Precipitated Lyn, Src, and Hck proteins were analyzed by immunoblotting with an anti-c-myc antibody. Purified GST or GST-K15 fusion proteins separated on an SDS-PAGE gel were visualized by Coomassie blue staining to show that equal amounts of fusion protein were used for GST pulldown experiments (bottom). α, anti.

FIG. 4.

Contribution of conserved K15 SH2 and SH3 binding motifs to inflammatory signaling. (a) HeLa cells were transfected with expression vectors for the K15-P or K15-M wild-type (WT) protein or its mutant YEEV, PPLP, or YASI versions. Extracted RNA was subjected to low-density microarray analysis using a Human Inflammation Array (Ocimum Biosolutions). Ratios of relative gene expression were calculated by dividing signal intensity values originating from K15-P (top), K15-M (bottom), or mutant protein samples by values obtained for the empty vector control. The graphs show ratio values for genes that were induced at least 1.5-fold by both the K15-P WT and the K15-M WT proteins. (b) HeLa cells were transfected with expression vectors for the K15-P or K15-M WT or its mutant YEEV, PPLP, or YASI versions, as described for low-density microarrays, and IL-8 levels (pg/ml) in cleared supernatants were measured by ELISA. (c) Cleared lysates from transfected HeLa cells were analyzed with an anti-K15 antibody for equal K15-P expression, an anti-Flag antibody for equal K15-M expression, and an anti-actin antibody as a control.

FIG. 5.

Effect of K15 expression on intracellular free calcium mobilization induced by anti-human IgM antibody. BJAB cells were transiently cotransfected with K15 expression plasmids and an mRFP plasmid in a 7:1 ratio to monitor successful transfections. The EBV LMP2A expression plasmid was included as a positive control for inhibition of calcium mobilization. Cells were loaded with 3 μM Indo-1 dye and stimulated with 10 μg/ml of goat anti-human IgM antibody. K15- or LMP2A-expressing cells were gated for high levels of mRFP expression, and calcium mobilization was monitored by changes in the ratio of violet to blue (405 to 485 nm) fluorescence of cells (median value, y axis) over time (x axis). Baseline calcium levels were established for 2 min prior to addition of goat anti-human IgM antibody. All measurements were performed at room temperature.