Drosophila TRP channels require a protein with a distinctive motif encoded by the inaF locus

Abstract

In both vertebrates and invertebrates, ion channels of the TRP superfamily are known to be influenced by a variety of accessory factors, but the list of interacting proteins is acknowledged to be incomplete. Although previous work showed that Drosophila TRP function is disrupted by mutations in the inaF locus, the mechanism of this effect has remained obscure. Here we show that a previously overlooked small protein, INAF-B, is encoded by the locus and fulfills its critical role in retinal physiology. The 81-aa INAF-B gene product is an integral membrane protein that colocalizes to rhabdomeres along with TRP channels. Immunoprecipitation experiments demonstrate that the two proteins participate in a complex, and blotting experiments show that neither protein survives in the absence of the other. Both proteins are normally part of a large supramolecular assembly, the signalplex, but their interaction persists even in the absence of the scaffold for this structure. The inaF locus encodes three other proteins, each of which has diverged from INAF-B except for a 32-aa block of residues that encompasses a transmembrane domain. This conserved sequence defines an inaF motif, representatives of which are found in proteins from organisms as diverse as nematodes, fish, and humans. Given the role of INAF-B, these proteins are good candidates for interacting partners of other members of the TRP superfamily.

Fig. 1.

The inaF transcript and its coding potential. (A) Diagram of genomic DNA from the region of the inaF message. Two exons are shown that are spliced to make an abundant eye-enriched 3.1-kb transcript. The second exon contains an ORF whose first methionine and stop codon are, respectively, indicated by a caret and asterisk; this 241-aa polypeptide has been proposed to be the CG2457 gene product (5, 6). However, the first exon contains an 81-aa ORF that we show herein to be necessary and sufficient for inaF function. The hatched bar above the exon/intron diagram shows the extent of the P106x deletion (5) that inactivates the gene. (B) Sequence of the 81-aa protein. The overlined region marks a predicted membrane-spanning domain that runs from Leu-42 to Ile-63. Below the sequence are given the amino acid changes for three substitution mutants (m1–m3), a frameshift mutant (m4), and the additional amino acids that are appended to the C terminus to generate a tag (HA).

Fig. 2.

Transgenic rescue of the inaF phenotype. (A) Total extracts of head proteins from various strains were probed with anti-TRP. All strains (except for a) carried the X-linked inaF^P106x deletion allele. Strains labeled c–g also carried a 13-kb segment of genomic DNA from the inaF region on an autosome. In strain c, this transgene carried the wild-type sequence; in strain d, the wild-type sequence was kept intact, but an HA tag was appended to the C terminus of the 81-aa ORF; in strains e and f, this ORF was modified by introduction of a triple substitution and a frameshift, respectively; in strain g, the 241-aa ORF on the transgene was modified by introduction of a frameshift. Strains labeled h–j carried, as indicated, a transgene bearing the eye-specific GMR-GAL4 driver (23), a UAS transgene bearing the 3.1-kb inaF cDNA, or both transgenes. Construction of the inaF transgenes is described in Materials and Methods; the sequence alteration of the transgene in strains d–f is given in Fig. 1B. (B) Electroretinograms recorded as described (24) from adults of the strains whose genotype is given in A. The bar above each trace shows the 20-s period during which these dark-adapted flies were exposed to white light (88,000 lux). Each trace was normalized to the potential difference that developed 0.25–0.50 s after the onset of light pulse. In traces a–g and h–j, the absolute value of this potential ranged from 16 to 22 mV and from 12 to 14 mV, respectively. In unrescued or poorly rescued flies (b, e, f, h, and i), by the end of the light pulse the photoreceptor potential decayed to the baseline or close to it. Conversely, in wild-type and well rescued flies (a, c, d, g, and j), the photoreceptor potential is still substantial at the end of the light pulse.

Fig. 3.

Association of HA-tagged INAF-PB with TRP channels. (A) Immunolocalization. A representative confocal section of retina, costained with anti-TRP and anti-HA, from adult flies bearing the tagged version of inaF-PB, is shown. To eliminate interference by eye pigment, the transgenic line also carried bw and st mutations. Within the ommatidium shown, signal from INAF-B is detected in rhabdomeres (and not their surrounding cell bodies) and colocalizes with signal from TRP. (B–F) Immunoprecipitation. The top line of the key for each panel shows the antibody used to precipitate dodecyl-β-maltoside extracts of fly heads: anti-HA (H), anti-TRP (T), anti-INAD (D), or a nonspecific control antibody (N). The second line of the key indicates the presence (+) or absence (−) of an HA tag on a transgenic copy of INAF-B. Where appropriate, the key also indicates the wild-type (+) or mutant (−) status of the locus whose genotype varies in the samples of the panel. Immunoprecipitates were electrophoresed and then probed with the antibody indicated at the left of each panel. The lanes with IP status marked (−) contain a sample (≈3%) of material used as input for the corresponding immunoprecipitations.

Fig. 4.

The inaF motif. (A) Diagram of the Drosophila genome from the inaF region showing additional exons that flank the exon for INAF-B. Shaded boxes delineate ORFs in the exons; the initial methionine and stop codon for each of them are marked with a caret and an asterisk, respectively. Open boxes enclosed by solid lines indicate UTRs deduced from the sequence of cDNAs. These cDNAs reveal that exons A–D are each spliced to the acceptor site of the rightmost exon; as a result, the methionine in this exon that was marked with a caret in Fig. 1A does not appear to function as the start of an ORF (it is an internal methionine of the inaF-D gene) and is now left unmarked. Open boxes enclosed by dashed lines indicate 5′ UTRs whose extent is uncertain. As in Fig. 1A, the extent of the P106x deletion is diagrammed with a hatched bar. (B) Alignment of a segment of the four ORFs diagramed in A. Residues are colored according to the Clustal (www.ebi.ac.uk/clustalw) convention. Asterisks and semicolons below the last sequence mark positions of identity or conservative substitution, respectively. (C) Alignment of the inaF motif found in INAF-D with similar motifs found in from ORFs from the indicated organisms. Residues are colored as in B, as are positions of conservation.