| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
FJ
EXPRESS SUMMARY ARTICLE The Full-length version of this article is also available, published online February 9, 2005 as doi:10.1096/fj.04-2947fje. |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,1
* Division of Nephrology, Mt. Sinai School of Medicine, New York, New York, USA; and
Division of Nephrology, Department of Internal Medicine, Juntendo University School of Medicine, Tokyo, Japan
1Correspondence: Division of Nephrology, Mount Sinai School of Medicine, Annenberg Bldg., Room 23-38 One Gustave L. Levy Pl., New York, NY 10029, USA. E-mail: kayo.hayashi{at}mssm.edu
SPECIFIC AIMS
Homophilic binding of mouse sidekick-1 and sidekick-2 proteins, orthologous transmembrane proteins of the immunoglobulin superfamily, has been shown to be important in targeting and stabilizing the developing neuronal synapse and in the pathogenesis of HIV-associated nephropathy. The aims of the current work were to: 1) characterize the homophilic binding properties of sidekick molecules in cells; and 2) define the critical domains responsible for targeting and stabilizing these homophilic interactions.
PRINCIPAL FINDINGS
1. Sidekicks mediate homophilic adhesion
To determine whether homophilic binding by sidekicks mediates cell adhesion, we cloned murine sidekicks 1 and 2. Sidekick proteins were then overexpressed in HEK 293T cells and cell aggregation assays were performed. Full-length sidekick-1 was subcloned into a pIRES vector so that it also expressed green fluorescent protein in a bicistronic message. The expression construct for sidekick-2 transfection expressed DsRed in a similar fashion. In this way, cells expressing sidekick-1 and sidekick-2 could be easily distinguishing by fluorescent microscopy. Two days after transfection, cells were trypsinized and single-cell suspensions were mixed on 24 well plates. Cells transfected with control vectors showed no aggregation after one hour of incubation on a rotating platform. In contrast, cells transfected with full-length sidekick-1 demonstrated significant aggregation in a time-dependent manner. Cells transfected with full-length sidekick-2 also showed significant aggregation. These aggregates formed in the absence of Ca2+, and the addition of Ca2+ did not enhance aggregation, suggesting that sidekicks are Ca-independent adhesion molecules. When sidekick-1 and sidekick-2 transfected cells were mixed together, cells expressing sidekick-1 aggregated only with other cells expressing sidekick-1, and cells expressing sidekick-2 likewise found their homophilic partner. Even when mixed together, the cells aggregated separately. These data suggest that homophilic adhesion between each sidekick is much stronger than any potential heterophilic interaction.
2. Short variant of sidekick-1 showed decreased aggregation
Previously, we cloned a shorter transcript of mouse sidekick-1 which has four immunoglobulin domains instead of six. The 5' noncoding region and the N terminus of this variant is encoded by a different exon from that of full-length sidekick-1, suggesting that it is an alternatively spliced form. Cells transfected with the shorter splice variant of sidekick-1 showed significantly decreased aggregation. To examine whether the short splice variant was able to bind to full-length sidekick-1, sidekick-1 full-length cDNA was subcloned into pIRES-DsRed2 vector. Mixing of full-length (DsRed) and shorter transcript (EGFP)-transfected cells demonstrated aggregation only between full-length transfected cells but not between full-length and shorter transcript transfected cells.
3. The first and second Ig domains are important in homophilic adhesion
To determine which Ig domain of sidekick-1 is required for the trans-dimer formation, we generated a deletion mutant lacking the second Ig domain of sidekick-1(
Ig2) (Fig. 1
A). Expression of sidekick-1 proteins was detected in the membrane fraction of transfected HEK 293T cells by Western blot analysis (Fig. 1B
). As shown in Fig. 1C, d
, the cells transfected with (Ig2) showed decreased adhesion similar to the short splice variant (Fig. 1C, c
). These results suggested that the first and second Ig domains are required for homophilic adhesion.
|
4. Mapping of the homophilic binding site to the second Ig domain
We hypothesized that sidekick-1 trans-dimer formation was mediated by an interaction between the first Ig domain of one sidekick molecule with the second Ig domain of its homophilic partner. To test this theory, we generated eight peptides corresponding to each ß-sheet of the second Ig domain as predicted from Chou-Fasman method. We generated C-terminal His-tagged recombinant proteins containing the first Ig domain alone or all six (1–6) Ig domains of sidekick-1. Equal amounts of the 8 synthesized peptides were adsorbed on a nitrocellulose membrane and recombinant proteins of the first Ig domain and the 1–6 Ig domains with 6xHis tag were used as probes to detect protein–protein interaction. Both the first Ig domain and the 1–6 Ig domains interacted with only P6 peptide. The proteins did not react with the other seven peptides or the BSA negative control. Adhesion studies performed with cells expressing a mutated version of sidekick-1 where the QLVILA domain is deleted failed to demonstrate significant adhesion. From these results, the predicted ß-sheet strand (QLVILA) functions as an important binding site and is necessary for homophilic adhesion. Ig domains are characterized by a conserved disulfide bond that links the two ß sheets. This is not essential for the structure of the domain, however, and many Ig superfamily members lacking this disulfide bond have now been reported. Among the 6 Ig domains of sidekick-1, only the second Ig domain is predicted to lack a disulfide bond. This unique structure of the second Ig domain may be important in conferring its binding properties.
5. Sidekick targeting is mediated by the first and second Ig domains
To further confirm that the first two Ig domains are sufficient to mediate homophilic adhesion, we generated a chimeric protein in which the first two Ig domains of sidekick-2 are replaced with the corresponding two from sidekick-1. This chimeric sidekick still formed significant aggregates when transfected in 293 T cells (Fig. 2
A). When cells transfected with chimeric sidekick were mixed with sidekick-1 transfected cells, these cells formed mixed aggregates consisting of both DsRed and EGFP positive cells (Fig. 2B
). This suggests that the first two Ig domains of sidekick-1 are sufficient to confer homophilic adhesion and targeting. Since Sidekick-1 and -2 have 57% homology at the amino acid level and are predicted to consist of a similar domain organization, we hypothesized that sidekick-2 may also form trans-dimers via its first two Ig domains. To test this hypothesis, we generated the reverse of the previously described chimera, in which we replaced the first two Ig domains of sidekick-1 with the corresponding two Ig domains of sidekick-2. This chimera also forms prominent aggregates in HEK 293 T cells after 30 min of rotation (Fig. 2C
). This reverse chimera containing only the first two Ig domains of sidekick-2 is able to bind to sidekick-2 transfected cells (Fig. 2D
). The binding surface of two cells is clearly shown by yellow as indicated in Fig. 2E
. These data show that homophilic properties of each sidekick protein are conferred by the combination of the first and second Ig domain. These results clearly show that the first two Ig domains of sidekick-2 are sufficient to mediate binding specificity of two molecules in trans-location.
|
CONCLUSIONS AND SIGNIFICANCE
In the present study, we have demonstrated in a cell-based system that sidekick-1 and sidekick-2 mediate homophilic adhesion such that each sidekick strongly prefers to bind to its own kind. Our findings support the in vitro data obtained by Yamagata et. al, which showed sidekick-mediated homophilic adhesion by using extracellular domains attached to beads.
The shorter splice variant with four Ig domains showed significantly decreased aggregate formation when transfected into cells (Fig. 3
A). When mixed with full-length sidekick-1 transfected cells, the short variant expressing cells did not bind to cells expressing the full-length version, suggesting that the first two Ig domains are important in binding between two molecules in translocation. When the second Ig domain of sidekick-1 is deleted, the formation of cell aggregates is also significantly decreased. This suggested that homophilic adhesion involved the first Ig domain binding to the second Ig domain of its partner (Fig. 3B
). By generating chimeric sidekick proteins, we have confirmed this model and have shown that the homophilic adhesion and targeting of each sidekick is conferred via the first and second Ig domains.
|
The recombinant first Ig domain-His protein only interacted with the peptide, which has the sequence ENQLVILATT (P6 peptide) in the second Ig domain of sidekick-1. The amino acid sequence of this region is also conserved in human and chicken, suggesting that the adhesion site may be conserved between species. In contrast, mouse sidekick-2 has two amino acids that differ in this region. These subtle amino acid differences may explain the specificity of homologous binding by sidekick molecules. The amino acid sequence corresponding to the P6 sequence in sidekick-1 may also be the binding site of sidekick-2, although further studies are required to demonstrate this.
The function of the short splice variant of sidekick-1 remains unknown. Since both the full-length and short splice variants are abundantly expressed yet clearly differ dramatically in their ability to mediate homophilic adhesion, it is important to better clarify their functional significance. We have shown that sidekicks are expressed in the ureteric bud and collecting ducts, which are the principle derivatives of the ureteric bud. This expression pattern correlates with several other genes known to have important roles in branching morphogenesis. Sidekick proteins may have a regulatory function in mediating the cell–cell interactions required for migration during branching tubulogenesis. In developing neurons, on the other hand, Yamagata et. al have shown that sidekicks localize specifically to the synapse and function to target axons to grow to specific synapses. It is likely that sidekicks function to link and stabilize the interaction between presynaptic and postsynaptic cells. Further analysis of the regulation of sidekick proteins will lead to a better understanding of their function.
FOOTNOTES
To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.04-2947fje;
This article has been cited by other articles:
|
|
 |
|
  L. Kaufman, G. Yang, K. Hayashi, J. R. Ashby, L. Huang, M. J. Ross, M. E. Klotman, and P. E. Klotman The homophilic adhesion molecule sidekick-1 contributes to augmented podocyte aggregation in HIV-associated nephropathy FASEB J, May 1, 2007; 21(7): 1367 - 1375. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
