Internal duplications in α-helical membrane protein topologies are common but the nonduplicated forms are rare

Abstract

Many α-helical membrane proteins contain internal symmetries, indicating that they might have evolved through a gene duplication and fusion event. Here, we have characterized internal duplications among membrane proteins of known structure and in three complete genomes. We found that the majority of large transmembrane (TM) proteins contain an internal duplication. The duplications found showed a large variability both in the number of TM-segments included and in their orientation. Surprisingly, an approximately equal number of antiparallel duplications and parallel duplications were found. However, of all 11 superfamilies with an internal duplication, only for one, the AcrB Multidrug Efflux Pump, the duplicated unit could be found in its nonduplicated form. An evolutionary analysis of the AcrB homologs indicates that several independent fusions have occurred, including the fusion of the SecD and SecF proteins into the 12-TM-protein SecDF in Brucella and Staphylococcus aureus. In one additional case, the Vitamin B12 transporter-like ABC transporters, the protein had undergone an additional fusion to form protein with 20 TM-helices in several bacterial genomes. Finally, homologs to all human membrane proteins were used to detect the presence of duplicated and nonduplicated proteins. This confirmed that only in rare cases can homologs with different duplication status be found, although internal symmetry is frequent among these proteins. One possible explanation is that it is frequent that duplication and fusion events happen simultaneously and that there is almost always a strong selective advantage for the fused form.

Figure 1

Visualisation of internal duplications detected by Structal and SHRIMP, one example for each OPM superfamily. For all proteins, the duplicated regions are extracted and thereafter one of them is rotated and aligned to the other using Structal. In 2nwlA and 2axtB, a duplication has most likely occured that includes a lateral rotation of one helix.

Figure 2

The length of the detected duplications in the three genomes and in the OPM set using predicted TM-segments. In all sets, longer chains seem to contain a higher fraction of duplications.

Figure 3

The distribution of the number of TM-segments among the homologes found after three rounds of PSI-BLAST, for one representative per OPM superfamily of all duplicated proteins found.

Figure 4

Number of genomes with certain number of hits of different lengths. Homologs with 5,6, or 7 TM-segments generally come in pairs while the longer homologs with 11,12, or 13 TM-segments can exist in a varying number of copies in different genomes.

Figure 5

Tree of homologous proteins from the 598 genomes. To simplify visualization, leaves and connecting branches to a sequence with 5, 6, or 7 predicted TM-segments are colored gray. Leaves and branches to sequences with 11, 12, or 13 predicted TM-segments that are split in half are colored red when being the N-terminal half and blue denoting the C-terminal half. Five clades can be seen; one containing all Sec-annotated proteins, one containing N-terminal parts of 11,12,13 TMH homologues and half of the six TMH homologues, and one containing C-terminal parts of long homologues and the other half of the short homologues. Bootstrap consensus tree of 2gifB and homologues taken from a database consisting of 10% randomly chosen sequences from 598 genomes.

Figure 6

Example of results from curve fitting of mono- and bi-topic Gaussian functions to homolog TM-segment distribution data. The protein exemplified is “ENSP00000284476”, a 12-TM protein from the “Patched” family of proteins involved in hedgehog signalling.

Figure 7

Distribution of the number of TM-segments in homologs to human proteins that implicates a gene duplication and merge in the past. The included query proteins (y-axis) are from Homo sapiens and have six or more predicted transmembrane segments. The heat-map shows the number of homologs to the query sequence in 650 genomes and their predicted number of TM-segments. “Q” denotes the number of TM-segments of the query sequence, “D” denotes the duplication size suggested by our SHRIMP-based duplication detection method. If no duplication was detected no “D” is present. At the right side of each row, the Pfam domains present in each protein is listed. All 48 proteins are clustered according to the similarity of the fitted curves. The proteins can roughly be divided into 13 groups, labeled A to M.