Role of the second intradiscal loop of peripherin/rds in homo and hetero associations

Abstract

Peripherin/rds (P/rds) is a disk rim protein that assembles into homo and hetero complexes with its nonglycosylated homologue, Rom-1, to maintain the integrity of the photoreceptor outer segment. Mutations in the rds gene have been identified in a variety of human retinal degenerative diseases. More than 70% of these mutations are located in the second intradiscal (D2) loop, highlighting the functional importance of this region. This study examines the involvement of different regions of the D2 loop in protein associations using a GST pull-down assay and a heterologous coexpression system. The pull-down assay suggests an association of the N-terminal portion (Phe(120)-Phe(187)) of the D2 loop with Rom-1 as well as with other P/rds molecules. Through peptide competition experiments, the region between Cys(165) and Asn(182) of the D2 loop has been identified as the domain for these associations. In a COS-1 cell heterologous expression system, coexpression of the D2 loop along with the intact P/rds and Rom-1 hindered the association of the two full-length proteins. In contrast to the homo association of P/rds molecules, it seems that the hetero association of P/rds with Rom-1 has a more stringent structural requirement. This work defines the crucial domain of the D2 loop, which mediates homo and hetero associations, specifically the regions that lay between Cys(165) and Asn(182). Elucidation of the molecular mechanisms behind the protein-protein associations of P/rds and its partners may reveal the pathogenic defects arising from the most common mutations in this gene.

Figure 1

Interactions between P/rds and Rom-1. Reciprocal co-IP was performed using protein extracts from mouse retinas and from COS-1 cells cotransfected with cDNAs encoding the two proteins. The anti-P/rds C-terminus and anti-Rom-1 C-terminus antibodies were used in the IP, and the products were detected by Western blotting with the respective antibodies.

Figure 2

Construction and expression of GST-P/rds fusion proteins. (A) Diagram showing the structure of GST-P/rds-D2, GST-P/rds-D2-N, GST-P/rds-D2-C, and GST-P/rds-CT. (B) Coo-massie blue gel staining showing expression of the fusion proteins in E. coli BL21 cells. (C) Western blot analysis of the fusion proteins using anti-GST and anti-P/rds D2 loop and C-terminus antibodies.

Figure 3

P/rds D2 loop interaction with Rom-1. (A) Interaction of Rom-1 to GST-P/rds-D2 and GST-P/rds-D2-N detected in the pull-down assay. The GST fusion proteins attached to the glutathione resin were incubated with mouse retinal extracts (50 μg of protein) at 4 °C for 2 h. Interactions of Rom-1 and presence of the fusion proteins were verified by Western blotting with the respective antibodies. (B) Absence of Rom-1 in the pull-down assay using GST-P/rds-D2-N incubated with retinal extracts prepared from rom-1^−/− mice. (C). Rom-1 association to GST-P/rds-D2 in a concentration-dependent pattern.

Figure 4

P/rds D2 loop interaction with other P/rds molecules. (A) GST pull-down assay showing binding of P/rds to GST-P/rds-D2 and GST-P/rds-D2-N. The D2 loop fusion proteins were incubated with mouse retinal extract, and binding of P/rds was detected with the anti-P/rds C-terminus antibody. Presence of the fusion proteins in these assays was verified with anti-GST and anti-P/rds D2 loop antibodies. (B) C terminus of P/rds is not involved in associations with other P/rds molecules. GST-P/rds-CT was incubated with mouse retinal extract and P/rds binding was detected with an anti-P/rds D2 loop specific antibody. Lane 5 shows the P/rds signal detected in 5 μg of retinal extract, which is only 10% of the total amount of retinal extract used in the pull-down assay. (C) Concentration-dependent binding of P/rds to GST-P/rds-D2-N. (D) GST pull-down assay showing binding of P/rds to GST-P/rds-D2 and GST-P/rds-D2-N using rom-1^−/− retinal extracts.

Figure 5

Identification of the domain in the P/rds D2 loop required for the protein associations. (A) Peptides used in the competitive binding assay. (B) Peptide competition assay identifying the domain for Rom-1 and P/rds association. GST-P/rds-D2 was incubated with retinal extracts (50 μg of protein) in the presence and absence of the competition peptides (0.3 mM) for 2 h at 4 °C. A peptide (P4) with nonrelated sequence was included as a negative control.

Figure 6

Involvement of the P/rds D2 loop in protein associations in COS-1 cells. (A) Diagram of the minigene encoding the D2 loop sequence with a translation initiation site and stop codon. (B) Coexpression of the D2 loop with intact P/rds and Rom-1 in COS-1 cells detected by Western blot analysis with the respective antibodies. (C) Nonreducing gel analysis showing formation of disulfide-linked homodimers of the D2 loop expressed in COS-1 cells. (D) Effects of coexpression of the D2 loop on the protein interactions in a co-IP assay. Protein extracts prepared from cells expressing the D2 loop, and the intact proteins were subjected to IP analysis with anti-P/rds C-terminus antibody. The immunoprecipitants were analyzed with the respective antibodies.

Figure 7

Sequence alignment of the N-terminal portion (Phe¹²⁰–Phe¹⁸⁷) of the P/rds D2 loop. Sequences of the N-terminal portion of the D2 loop from all known P/rds homologues are aligned. The region between Cys¹⁶⁵ and Asn¹⁸² is identified as the domain for P/rds and Rom-1 binding and is marked on the top with a solid line. The residues in which mutations cause human retinal degenerations are bold and enlarged. An asterisk refers to identical residues, and a colon refers to residues with high similarity in all P/rds homologues.