Abstract
The processes of nutrient uptake and signal sensing are crucial for microbial survival and adaptation. Membrane-embedded proteins involved in these functions (transporters and receptors) are commonly regarded as unrelated in terms of sequence, structure, mechanism of action and evolutionary history. Here, we analyze the protein structural universe using recently developed artificial intelligence-based structure prediction tools, and find an unexpected link between prominent groups of microbial transporters and receptors. The so-called S-components of Energy-Coupling Factor (ECF) transporters, and the membrane domains of sensor histidine kinases of the 5TMR cluster share a structural fold. The discovery of their relatedness manifests a widespread case of prokaryotic “transceptors” (related proteins with transport or receptor function), showcases how artificial intelligence-based structure predictions reveal unchartered evolutionary connections between proteins, and provides new avenues for engineering transport and signaling functions in bacteria.
Similar content being viewed by others
Introduction
Membrane-embedded sensory kinases transduce information across the lipid bilayer, while membrane transporters translocate molecules and ions. The two classes of proteins are generally considered unrelated, having evolved from different ancestral proteins1,2. In bacteria, signal sensing is often mediated by the two-component systems consisting of a multi-domain sensor histidine kinase (SHK) that binds the signaling molecule and initiates a phosphorylation cascade, and a response regulator that orchestrates gene regulation3. In contrast, transported nutrients are recognized by unrelated proteins that undergo a series of conformational changes leading to molecular translocation across the membrane4. Although operated by distinct entities, transport and signal sensing can function synergistically, as demonstrated by the complex formed by the C4-dicarboxylate transporter DctA and the fumarate sensor DcuS in Escherichia coli5, or as shown with the structural characterization of the BceAB-S module from Bacillus subtilis6, where an ABC transporter (BceAB) interfaces an histidine kinase (BceS) to respond to antimicrobial peptides.
Over the past decades, rare cases of membrane proteins named ‘transceptors’7,8,9 have been reported indicating that this distinction between receptors and transporters may not be absolute. We here define transceptors strictly as membrane folds, thus excluding scenarios in which a dedicated soluble domain acts as a bridge between transport proteins and signal transduction systems (e.g., STAC domain alone or incorporated into CbrA from Pseudomonas putida)10,11. The identified membrane transceptors structurally resemble transporters, yet function as receptors, suggesting that the two processes might be evolutionary linked. However, based on the extremely low number of reported cases, and their exclusive presence in eukaryotic organisms (detailed list in Supplementary Table 1)12,13,14,15,16,17,18,19, transceptors appear to be oddities. The apparent rareness is further underscored because each known transceptor case constitutes only a single protein in a large transporter family that is likely to have evolved from a transporter to a receptor function.
Sequence comparisons have not been able to detect a more widespread occurrence of transceptors across the domains of life, but it is well known that protein structures are better conserved than amino acid sequences, and therefore better suited to find distant evolutionary relations20. In a recent study, a new transceptor was found when the structure of a KDEL receptor, involved in intracellular trafficking of eukaryotic proteins, was solved. The protein turned out to be a structural homolog of the sugar transporters of the SWEET family, despite the lack of appreciable sequence identity between these membrane proteins21. While structural similarity searches may be used to find more transceptor cases, the limited availability of experimentally determined structures often makes it impossible to uncover such relations. Computational structure prediction could potentially fill this gap, yet until recently its lack of accuracy has hindered successful use in protein structure determination. The emergence of AlphaFold22,23, and the resulting millions of accurately predicted protein structures, including those of integral membrane proteins24, now allow to search for proteins structurally related to transporters and receptors that were previously out of the reach of conventional structural biology.
Here, we used AlphaFold structures to search for transceptors in prokaryotes. We focused on bacterial ATP-binding cassette (ABC) transporters of type III, the so-called Energy-Coupling Factor (ECF) transporters25,26. ECF transporters consist of a three-subunit motor module (termed ECF module) that binds and hydrolyzes ATP, and an integral membrane subunit (S-component or EcfS) that binds and translocates the cognate substrate across the membrane upon association with the motor module (Fig. 1)27,28. While the ECF module is well conserved with multiple typifying motifs (i.e., Walker A and B motifs, signature motifs of the ATPase subunits, and Ala-Arg-Gly motifs of the scaffold protein EcfT)26,29, S-components with specificity for different transported substrates display highly variable sequences, which has complicated the identification of the encoding genes. Nonetheless, all structurally characterized S-components have a conserved characteristic fold with six membrane embedded helices. We used Alphafold structure predictions to identify previously undetected S-components. Surprisingly, we found that S-components are structurally similar to the membrane domains common to a large number of sensor histidine kinases (the so-called 5TMR-SHKs) and chemotactic receptors, suggesting an evolutionary link between these groups of proteins. Thus, we hypothesize that SHKs might have evolved from ancestral transporters that gained the new function of redirecting the metabolic response without translocating their ligand or vice versa. We show that the S-component fold represents the most widespread manifestation of transceptors to date, and the first case found in prokaryotes. We also conclude that the S-component fold is much more abundant and versatile than previously thought, and can specialize in multiple functions, including molecular transport and signal transduction.
Results
The fold of S-components is conserved in proteins with different function from micronutrient import
S-components from ECF transporters, specific for different substrates, all share a characteristic fold consisting of six membrane-spanning α-helices, but they do not exhibit significant sequence identity (∼15%)26, a feature that makes the identification of related proteins based on sequences non-trivial30. We used the Foldseek webserver31 to search for proteins structurally related to S-components in the databases of protein structures predicted by the AlphaFold neural network. As input for the search, we took experimentally determined structures of four S-components: RibU32 from Staphylococcus aureus TCH60, FolT33 from Lactobacillus delbrueckii ATC 11842, ThiT34 and BioY35 from Lactococcus lactis MG1363 – that mediate transport of riboflavin, folate, thiamine and biotin, respectively. In parallel, we also used the AlphaFold predicted structures of the same four proteins as input in our searches. The search in the AlphaFold model databases AFDB proteome and AFDB-SwissProt retrieved proteins with the same fold as the queries (Fig. 2a). Unsurprisingly, a substantial portion of the annotated hits are other S-components (PFAM entry: PF12822), regardless of the predictive or experimental nature of the S-component structure used as entry. In all cases, the best-scoring matches were homologous S-components with identical predicted substrate specificity to the protein query. Unexpectedly and consistently, we also found proteins functionally unrelated to S-components. The most abundant ones are annotated as sensor histidine kinases (SHKs, PFAM entry: PF07694), specifically the pyruvate sensors YpdA and BtsS, and the sporulation kinase B KinB. In the past, before accurate structure predictions were possible, these histidine kinases had been named 5TMR-LYT (5 TransMembrane Receptor of the LytS-YhcK type)36, a name that is now known to be incorrect, because the proteins contain more than five transmembrane domains (discussed in detail below)37. In addition, we found rod-shape determining proteins of the MreD family (PFAM entry: PF04093), which had already been suggested to have the S-component fold38. SHKs constituted a larger fraction of the hits (up to 25% of hits in the AFDB-proteome database, depending on the query protein) than MreD proteins, most likely because the sole function of MreD is tightly coupled to determining cell architecture in rod-shaped bacteria39,40,41,42, while signal sensing receptors have diversified to respond to different stimuli43. Finally, entries not functionally annotated represented 20 and 60% of the hits, depending on the specific search, and in most cases their AI-predicted structure suggests the function of S-components (although not annotated as such in protein databases). Both for the SHK and MreD hits, the structural predictions were of high confidence (Supplementary Table 2). To validate AlphaFold-predicted structural similarity between S-components and 5TMR-SHKs, we used well-established approaches of protein fold recognition44,45,46 using the sequence of the membrane domain of 5TMR-SHKs (Supplementary Tables 3 and 4): crystal structures of S-components were always scoring among the high-confidence hits for structural relatedness with SHKs, consistently with the similarity revealed from the AI-based tools.
The core of the shared fold, which we refer to as “S-component fold”, is composed of six alpha-helices (H1–H6) connected by 5 loops protruding alternately towards the extracellular side and the cytoplasm (Fig. 2b–d)47,48. The 5TMR domain of SHKs has an additional helix at the N-terminus (H0) linked to the six-helical core by a cytosolic loop, resulting in a total of 7 transmembrane helices, as recently demonstrated biochemically for the high-affinity pyruvate receptor BtsS37. H0 has its N-terminus on the extracellular side, and the C-terminus at the cytoplasmic side of the membrane. H6 delimits the C-terminal end of the protein in the case of S-components and MreD, but not of the SHKs, which instead extend from the membrane part towards the cytoplasm into soluble domains required for signal transduction49. Because the exact role of MreDs is unclear, and mechanistic insight in rod-shape determination is lacking, we focus primarily on the relation between S-components and SHKs, and discuss MreDs only in the context of phylogenetic distribution.
The shared fold is widespread across bacterial phyla
S-components of ECF transporters are found in some archaea, but are much more prevalent across bacteria50 (Fig. 3a). A similar distribution was observed for the SHKs with shared S-component fold, while MreDs are entirely absent in Archaea. This aligns well with a crenactin-based cytoskeleton (rather than employing MreB, MreC, and MreD proteins)40 responsible for shape determination in most rod-shaped archaea51.
Among bacteria, S-components of ECF transporters have predominant presence in Gram-positive microorganisms: more than 75% of known S-components were found in Bacillota and Actinomycetota25,26 (Fig. 3b). A very different picture emerges for the SHKs and MreDs, as they both prevail in the Pseudomonadota phylum, constituting 46% and 49% of the respective total distributions, but are also well represented on Gram-positive bacteria (41% and 37% of the total). Among Gram-positive microbes, SHKs with the S-component fold are found almost exclusively in Bacillota, while MreDs are also present in Actinomycetota. Collectively, we can conclude that the shared S-component fold is diverse not only in regard of the functional specialization but also for widespread occurrence across microorganisms with different morphologies and metabolisms.
In the absence of sequence similarity, common function or comparable taxonomic distributions, we tried to unravel the relationship between the protein groups sharing the S-component fold by taking advantage of their AI-predicted structures. The AlphaFold structures of representative proteins with the S-component fold from ECF transporters, SHKs and MreDs were used for structural multi-alignments and to build a phylogenetic tree (Supplementary Fig. 1 and Supplementary Table 5). However, the resolution of the tree based on the structural alignment was not high enough for definitive conclusions about the evolutionary origin of the fold. In future, experimental structures, improved structural predictions and AI-based structural phylogenetics may overcome this limitation52,53.
Fold specialization of the 5TMR-containing sensor histidine kinases
We reasoned that the previously proposed role of the N-terminal helix H0 of SHKs as signal peptide was potentially incorrect since its membrane topology (with the N-and C-termini on the extra- and intracellular sides, respectively) is opposite to the canonical configuration in which the cleavage site downstream of the helix is located extracellularly (Fig. 2c)54. Consistently, the predicted probability of a signal peptide sequence in this helix by the algorithm SignalP 5.0 was virtually zero (Supplementary Fig. 2). Instead, the H0 domains of SHKs contain well conserved regularly spaced leucine residues (Fig. 4a), reminiscent of leucine zipper-like motifs, a feature that in the case of sensor histidine kinases might facilitate dimerization within the bilayer, which is required for the receptor functionality55. In addition, the conservation of two consecutive polar amino acids located towards the extracellular side of the alpha helix (shaded in orange in the Fig. 4a) further supports the potential function of helix H0 in promoting dimerization56,57. AlphaFold predictions for the dimeric conformation of 5TMR-SHKs highlighted the close proximity – and probable interaction - of the conserved polar side chains (around 4 Å), corroborating the hypothesis of the formation of an intramembranous salt bridge between two protomers.
A further conserved sequence characteristic of the 5TMR domain of SHKs is a NXR motif located between helix 1 and helix 2, towards the extracellular side of the membrane36. The side chains of the asparagine and arginine residues point toward the central cavity delineated by the transmembrane helices, where substrate binding takes place in the S-components of ECF transporters32,58. Therefore, the two residues in the 5TMR domain may play a role for the binding of the external signaling molecule. Four additional conserved motifs face the same cavity, all of which located at positions that in S-components are involved in substrate binding (Fig. 4b, and Supplementary Figs. 3 and 4). The conserved residues are distributed over multiple helices of the core fold, suggesting that substrate binding leads to intramolecular conformational changes, similar to what has been shown for S-components of ECF transporters. While in ECF transporters, the binding of substrate leads to association with the ECF module, we speculate that in SHKs, conformational changes upon substrate binding may lead to a specific reorganization of the pre-existing interactions between the two receptor protomers, and the transmission of the signal from the membrane to the soluble domains.
Intracellular domains of 5TMR-SHK receptors
A limited number of combinations of intracellular domains has been reported to be linked to the 5TMR (S-component-like) membrane domain of SHKs36, including GGDEF and EAL domains (producing and degrading cyclic-di-GMP)59, PAS and GAF domains (involved in receptor dimerization and binding of cyclic nucleotides)60,61, and HK and HK-ATPase domains (synergically responsible for signal transduction in numerous histidine kinases)62. Analysis of the much larger group of 5TMR-SHKs (PFAM entry: PF07694) found by the structure prediction revealed that the soluble domains are much more diverse (Fig. 4c, and Supplementary Table 6), reaching almost 500 different functionally annotated combinations. Although the soluble PAS, GGDEF and EAL domains are often fused with the 5TMR domain, they can be combined in a modular manner. While YpdA falls into the category of 5TMR-SHKs with a GAF domain separating the membrane fold from the soluble HK and HK-ATPase domains, variations of the classic configuration with only HK and HK-ATPase also are present. They have one or more response regulator receiver (REC) domains fused to the C-terminus of the HK-ATPase of the receptor, thus forming the so-called hybrid histidine kinases63,64, which are specialized in the multistep phosphorelay within the receptor or towards other response regulator proteins.
Surprisingly, we also discovered three unprecedented domains fused to the 5TMR fold namely methyl-accepting chemotaxis protein (MCP) signaling domain (PFAM entry: PF00015), PPM-type phosphatase-like (annotated as SpoIIE in the PFAM entry: PF07228), and natural resistance-associated macrophage protein (NRAMP)-like (PFAM entry: PF01566). The association with a MCP domain strongly suggests that the substrate recognized by 5TMR-receptors can directly trigger a chemotactic response in the microorganisms equipped with such a domain architecture (Supplementary Table 7). In some cases, a PAS domain separates the 5TMR fold from the MCP domain, mostly in bacteria belonging to the order Hyphomicrobiales, but there are different cases where the membrane and the chemotactic domains are directly attached by a short coil (e.g., in certain species of Clostridium and Treponema). The PPM-type phosphatase and the NRAMP domains are prevalent in archaea rather than in bacteria. The PPM domain was already known in the context of archaeal signal transduction65,66. The fusion of the sensorial 5TMR domain with NRAMP-like proteins constitutes instead a more exceptional scenario: not only has this architecture the 5TMR fold at the C-terminus of the protein sequence, but it also represents the only protein assembly in which an integral membrane transporter is used as a fused domain. NRAMP proteins are membrane transporters for divalent metals67, potentially indicating a unique metabolic specialization bridging ion transport and signal sensing to ensure intracellular homeostasis.
The assembly with an ECF module for transport is associated with the conservation of the AXXXA domain in helix 1
S-components from ECF transporters are specialized in binding their substrate with high affinity, and enable the translocation of substrate upon the association with a tripartite ATPase ECF module27,28. The association leads to a characteristic toppling motion in which the S-component rotates relatively to the membrane plane, which brings the substrate binding site from an outward- to an inward-facing orientation, allowing substrate release into the cytoplasm. The assembly of the S-component and the ECF module into a full ECF transporter (Fig. 5a) involves an AXXXA motif in helix 1, which is relatively conserved among S-components, as evidenced by the structures of the full complexes determined to date32,68,69. This motif is responsible for the tight association of the S-component with the scaffold protein of the ECF module (EcfT). Mutation of either alanine prevented the S-component ThiT to interact with the ECF module, and cause a complete loss of transport ability70. This typifying motif in helix 1 is absent from MreDs and SHKs (Fig. 5b). Instead, amino acids with large hydrophobic (leucine, isoleucine, valine) or aromatic (phenylalanine, tryptophan) side chains mostly occur at the positions of the two alanines, suggesting their inability to interact with the motor module of ECF transporters, and consistent with the different roles of MreDs (scaffold needed for cell shape determination)42 and 5TMR-containing SHKs (signal sensing)71.
Discussion
The advancement of structure prediction and analysis tools based on artificial intelligence training (i.e. AlphaFold and Foldseek) is transforming life sciences subfields72,73. Combining these AI-based tools with bioinformatic approaches, we found that two major groups of bacterial membrane proteins share a core fold: S-components of ECF-type ABC transporters for micronutrient uptake and 5TMR-SHK receptors of two-component systems involved in signal sensing and chemotaxis. This structural relatedness between transporters and receptors manifests the first transceptor case identified in prokaryotes (receptors structurally related to transporters but not functionally), and the most abundant one to date. The evolutionary origin of the shared fold described in this work remains elusive, as we could not build conclusive phylogenetic trees from the AI-based structures currently available. The improvement of the computational methods, corroborated by experimentally determined structures and biochemical studies in vivo and in vitro52, will allow to shed further light on the evolutionary origin of the bacterial transceptor here reported.
In the absence of experimentally determined structures for 5TMR-SHKs, the structural relatedness between S-components and these receptors must await experimental confirmation. In fact, the structures predicted by AlphaFold do not always correspond with those determined by conventional techniques for structure determination52. However, the confidence of our approach is high for multiple reasons: (i) we used both AI-predicted and experimentally determined structures of S-components to search for structural homologs (Fig. 2a), (ii) the TM-scores obtained for these protein folds of approximately 200 amino acids were in most cases above 0.5 (Supplementary Table 2), (iii) the corresponding RMSD values were mostly comprised between 4 and 7 Å (Supplementary Table 2), (iv) fold recognition from the primary sequence of the membrane domain of the 5TMR-SHKs always gave crystallographic structures of S-components (Supplementary Tables 3-4), (v) certain amino acids well conserved in the pocket for substrate recognition in 5TMR-SHKs (Fig. 4b) correspond to amino acids involved in substrate binding in S-components with different specificities (Fig. S4).
The S-component fold may be suitable for both transport (in the context of the ECF module) and signal transduction function for two reasons. First, when S-components are not associated with the ECF module, the substrate binding sites are located close to the extracellular side of the membrane. This location is also suitable for receptors, as for instance highlighted by G-protein-coupled receptors in eukaryotes74. Comparison of the AlphaFold structures of 5TMR-SHKs with the crystal structures of S-components in complex with their substrates, supports a similar position of the substrate binding sites (Supplementary Fig. 4): the NXR motif, Cys114 in helix 4 and Met177 in helix 6 of SHKs are equivalent to amino acids interacting with the cognate ligand in S-components (Fig. 4b). This proposed location of the ligand binding site is consistent with mutagenesis by Qiu and coworkers37 of the pyruvate sensor BtsS.
Second, we speculate that analogously to S-components of ECF transporters (Fig. 6), substrate binding to 5TMR-SHKs may also initiate a toppling-like movement of the membrane fold. However, in 5TMR-SHKs the motion cannot be completed because the proteins do not interact with an ECF module, and it is counteracted by H0 and the cytoplasmic extensions of H6. Rather than toppling and associating with the ECF module, the conformational changes induced by the substrate recognition at the membrane domain interface may culminate into an orchestrated series of intramolecular rearrangements of the cytosolic domains, and eventually enable the downstream phosphorelay cascade. In the proposed step of ancestral fold specialization, the particular adaptation of the regions involved in the conformational changes towards the intracellular portion might have determined the specialization into either substrate translocators or signal sensing proteins. It is noteworthy that a few S-components have been found to be able of transporting a substrate across the membrane without the need for an ECF module (biotin by some BioY proteins and vitamin B12 by BtuM)30,75. These proteins might be considered candidate intermediates between ECF transporters and SHKs, although the mechanism that these solitary S-components use for transport is still unclear.
Our work may inspire engineering efforts to transform receptors in transporters and vice versa, with microbial manufacturing as main application76. Alternatively, the exclusive occurrence of S-components and sensor histidine kinases in bacteria can be exploited as common targets to design novel antimicrobial compounds inhibiting either micronutrient uptake or signal sensing and transduction in pathogenic strains77,78, or both.
Methods
Structural similarity search
The entries of RibU (UniProt accession number E5QVT2), FolT (UniProt accession number Q1G930), ThiT (UniProt accession number A2RI47) and BioY (UniProt accession number A2RMJ9) from the protein database UniProt79 (UniProt release 2023_04) were employed to obtain both experimentally determined (PDB entries: 3P5N, 5D0Y, 4POP, 4DVE) and AlphaFold-predicted structures (PDB entries: AF-E5QVT2-F1, AF-Q1G930-F1, AF-A2RI47-F1, AF-A2RMJ9-F1). The 3D maps were used as input entries for the structural similarity search on the Foldseek webserver31 (release 8-ef4e960), and the resulting top hits were extracted from the AlphaFold model databases AFDB proteome and AFDB-SwissProt (Fig. 2a). The probability cut-off value for including an entry among the top hits with identical protein fold was set to 50%. In the case of experimentally determined structures, only the hits from chain A were considered for further analyses. Details on the structural similarity results obtained with YpdA. BtsS and KinB (e.g., template modeling score and root mean square deviation value) are available on Supplementary Table 2. The AlphaFold structures showing the shared fold (Fig. 2b–d) were visualized using UCSF CHIMERAX software (version 1.6.1)80.
The amino acid sequences corresponding to the membrane domain (first 200 amino acids) of well-known 5TMR-SHKs36,37,81,82 (UniProt accession numbers: P94513, P54595, P0AA93, P0AD14, Q08430, Q2G061) were used as entries to generate sequence–structure alignments via the PSIPRED workbench46, and find crystallized structures with the same protein fold. The output of the fold prediction and protein family classification searches from the pGenTHREADER and pDomTHREADER methods44 are reported on Supplementary Tables 3 and 4.
Protein fold taxonomic distribution analysis
S-components and related protein hits from the structural similarity search were searched on the UniProt resource under the entry ‘Family and domain databases’. The PFAM database, now integrated in the InterPro webserver (release InterPro 96.0)83, returned as results the entries PF12822 (ECF transporter, substrate-specific component), PF07694 (5TMR of 5TMR-LYT) and PF04093 (rod shape-determining protein MreD), which were then consulted on the corresponding ‘Taxonomy’ section. For each entry, protein occurrence across domains of life (Fig. 3a) and bacterial phyla (Fig. 3b) were displayed in terms of number of species in percentage or absolute numbers, respectively. The distributions were represented using GraphPad Prism 9.0.0.
Bioinformatic analyses on the shared fold
The primary structures of the abovementioned well-known 5TMR-SHKs were used to query the SignalP 5.0 webserver84 with the aim to predict any signal peptide sequence at the N-terminus of the proteins (Supplementary Fig. 2). The sequence of RibU was used as a negative control, while that of GlnP85 (UniProt accession number: Q9CES5) was used as a positive control. The inquiry on SignalP 5.0 accounted for the Gram-positive/negative origin of the organism harboring the receptors.
The sequence logo of helix 0 (Fig. 4a, on the left) was generated from the multiple sequence alignment of 5TMR-SHKs, performed on ClustalOmega86, and consequently visualized via WebLogo87. The sequences used for the multialignment were collected from the InterPro resource, more precisely from the ‘Alignment’ section (https://www.ebi.ac.uk/interpro/entry/pfam/PF07694/entry_alignments/?type=seed), and were implemented with the sequences of the 5TMR-containing SHKs previously employed for the signal peptide prediction. To ensure that the alignment involved only the 5TMR domain, we excluded sequences in which the 5TMR domain is at the C-terminus (proteins with the NRAMP-1 in their domain architecture), and cut the sequences to just the first 180 amino acids.
The conserved regions of the 5TMR-containing SHKs (Fig. 4b) were highlighted from the multialignment of the same sequences mentioned above in the alignment, but involving exclusively the 5TMR domain (the amino acids corresponding to helix 0, cytosolic loop linking H0 to the shared fold, and intracellular domains were excluded). Also in this case, the visualization of the sequence logo was carried out via WebLogo. The top view of the six-helical fold localizing the conserved motifs of the localization of the conserved regions was generated by UCSF CHIMERAX software.
The dimeric conformation of 5TMR-receptors (Fig. 4a, on the right) was obtained using the ColabFold suite (version 1.5.2)88 available on the web-based gateway COSMIC289. Specifically, the primary sequence of YpdA from E. coli was employed as a template for the prediction of the homodimeric state selecting standard parameters (5 models, 3 recycles and stop at score of 80). Upon verification of the models, the predicted multimeric state was visualized via CHIMERAX.
The multiple combinations of sensor histidine kinases containing the 5TMR-domain (Fig. 4c) were retrieved from the ’Domain architectures’ entry available on InterPro (https://www.ebi.ac.uk/interpro/entry/pfam/PF07694/domain_architecture/). Representative AlphaFold structures of SHKs showcasing the variegated intracellular assembly of the receptors were downloaded and visualized with CHIMERAX.
The ECF transporter shown in Fig. 5a corresponds to the folate ECF-transporter (PDB entry: 5D3M)33 and its cartoon was generated using CHIMERAX, as well as the helix 1 of the respective S-component represented in Fig. 5b. The sequence logo of helix 1 (Fig. 5b) was obtained for each protein family (entries PF12822, PF07694, PF04093) from the multiple sequence alignment available on the InterPro resource at the correspondent ‘Alignment’ entry, selecting the sequences provided by the SEED database90.
Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
Data availability
All data, protein identifiers (UniProt accession numbers: number E5QVT2, Q1G930, A2RI47, A2RMJ9, P94513, P54595, P0AA93, P0AD14, Q08430, Q2G061, Q9CES5, Q2JYM9, Q488U0, F5RD42, C6WYA8, Q64AC4, E8TNX1, A8TXH3, A0A4R3E8R1, A2SU24) and corresponding protein structures were obtained from online resources publicly available, and described in the main document and the supplementary information. Further details on the analyses carried out in this study are available upon reasonable request to the corresponding author.
References
Saier, M. H. Tracing pathways of transport protein evolution. Mol. Microbiol. 48, 1145–1156 (2003).
Wuichet, K., Cantwell, B. J. & Zhulin, I. B. Evolution and phyletic distribution of two-component signal transduction systems. Curr. Opin. Microbiol. 13, 219–225 (2010).
Laub, M. T. & Goulian, M. Specificity in two-component signal transduction pathways. Annu. Rev. Genet. 41, 121–145 (2007).
Burkovski, A. & Krämer, R. Bacterial amino acid transport proteins: occurrence, functions, and significance for biotechnological applications. Appl. Microbiol. Biotechnol. 58, 265–274 (2002).
Witan, J. et al. Interaction of the Escherichia coli transporter DctA with the sensor kinase DcuS: Presence of functional DctA/DcuS sensor units. Mol. Microbiol. 85, 846–861 (2012).
George, N. L. & Orlando, B. J. Architecture of a complete Bce-type antimicrobial peptide resistance module. Nat. Commun. 14, 3896 (2023).
Thevelein, J. M. & Voordeckers, K. Functioning and evolutionary significance of nutrient transceptors. Mol. Biol. Evol. 26, 2407–2414 (2009).
Diallinas, G. Transceptors as a functional link of transporters and receptors. Microb. Cell 4, 69–73 (2017).
Steyfkens, F., Zhang, Z., Van Zeebroeck, G. & Thevelein, J. M. Multiple transceptors for macro- and micro-nutrients control diverse cellular properties through the PKA pathway in yeast: a paradigm for the rapidly expanding world of eukaryotic nutrient transceptors up to those in human cells. Front. Pharmacol. 9, 1–22 (2018).
Korycinski, M. et al. STAC - a new domain associated with transmembrane solute transport and two-component signal transduction systems. J. Mol. Biol. 427, 3327–3339 (2015).
Wirtz, L., Eder, M., Schipper, K., Rohrer, S. & Jung, H. Transport and kinase activities of CbrA of Pseudomonas putida KT2440. Sci. Rep. 10, 1–12 (2020).
Donaton, M. C. V. et al. The Gap1 general amino acid permease acts as an amino acid sensor for activation of protein kinase A targets in the yeast Saccharomyces cerevisiae. Mol. Microbiol. 50, 911–929 (2003).
Van Zeebroeck, G., Bonini, B. M., Versele, M. & Thevelein, J. M. Transport and signaling via the amino acid binding site of the yeast Gap1 amino acid transceptor. Nat. Chem. Biol. 5, 45–52 (2009).
Hyde, R., Cwiklinski, E. L., MacAulay, K., Taylor, P. M. & Hundal, H. S. Distinct sensor pathways in the hierarchical control of SNAT2, a putative amino acid transceptor, by amino acid availability. J. Biol. Chem. 282, 19788–19798 (2007).
Goberdhan, D. C. I., Meredith, D., Boyd, C. A. R. & Wilson, C. PAT-related amino acid transporters regulate growth via a novel mechanism that does not require bulk transport of amino acids. Development 132, 2365–2375 (2005).
Lorenz, M. C. & Heitman, J. The MEP2 ammonium permease regulates pseudohyphal differentiation in Saccharomyces cerevisiae. EMBO J. 17, 1236–1247 (1998).
Schothorst, J., Zeebroeck, G. V. & Thevelein, J. M. Identification of Ftr1 and Zrt1 as iron and zinc micronutrient transceptors for activation of the PKA pathway in saccharomyces cerevisiae. Microb. Cell 4, 74–89 (2017).
Popova, Y., Thayumanavan, P., Lonati, E., Agrochão, M. & Thevelein, J. M. Transport and signaling through the phosphate-binding site of the yeast Pho84 phosphate transceptor. Proc. Natl Acad. Sci. USA. 107, 2890–2895 (2010).
Kankipati, H. N., Rubio-Texeira, M., Castermans, D., Diallinas, G. & Thevelein, J. M. Sul1 and Sul2 sulfate transceptors signal to protein kinase A upon exit of sulfur starvation. J. Biol. Chem. 290, 10430–10446 (2015).
Illergård, K., Ardell, D. H. & Elofsson, A. Structure is three to ten times more conserved than sequence - a study of structural response in protein cores. Proteins Struct. Funct. Bioinforma. 77, 499–508 (2009).
Newstead, S. & Barr, F. Molecular basis for KDEL-mediated retrieval of escaped ER-resident proteins – SWEET talking the COPs. J. Cell Sci. 133, jcs250100 (2020).
Jumper, John et al. Highly accurate protein structure prediction with AlphaFold. Nature 596, 583–589 (2021).
Varadi, M. et al. AlphaFold protein structure database: massively expanding the structural coverage of protein-sequence space with high-accuracy models. Nucleic Acids Res 50, D439–D444 (2022).
Hegedűs, T., Geisler, M., Lukács, G. L. & Farkas, B. Ins and outs of AlphaFold2 transmembrane protein structure predictions. Cell. Mol. Life Sci. 79, 1–12 (2022).
Eitinger, T., Rodionov, D. A., Grote, M. & Schneider, E. Canonical and ECF-type ATP-binding cassette importers in prokaryotes: Diversity in modular organization and cellular functions. FEMS Microbiol. Rev. 35, 3–67 (2011).
Rempel, S., Stanek, W. K. & Slotboom, D. J. ECF-Type ATP-binding cassette transporters. Annu. Rev. Biochem. 88, 551–576 (2019).
Karpowich, N. K., Song, J. M., Cocco, N. & Wang, D. N. ATP binding drives substrate capture in an ECF transporter by a release-and-catch mechanism. Nat. Struct. Mol. Biol. 22, 565–571 (2015).
Thangaratnarajah, C. et al. Expulsion mechanism of the substrate- translocating subunit in ECF transporters. Nat. Commun. 14, 4484 (2023).
Neubauer, O. et al. Two essential arginine residues in the T components of energy-coupling factor transporters. J. Bacteriol. 191, 6482–6488 (2009).
Rempel, S., Colucci, E., de Gier, J. W., Guskov, A. & Slotboom, D. J. Cysteine-mediated decyanation of vitamin B12 by the predicted membrane transporter BtuM. Nat. Commun. 9, 3038 (2018).
van Kempen, M. et al. Fast and accurate protein structure search with Foldseek. Nat. Biotechnol. 1–4 (2023). https://doi.org/10.1038/s41587-023-01773-0.
Zhang, P., Wang, J. & Shi, Y. Structure and mechanism of the S component of a bacterial ECF transporter. Nature 468, 717–720 (2010).
Swier, L. J. Y. M., Guskov, A. & Slotboom, D. J. Structural insight in the toppling mechanism of an energy-coupling factor transporter. Nat. Commun. 7, 11072 (2016).
Swier, L. J. Y. M. et al. Structure-based design of potent small-molecule binders to the S-component of the ECF transporter for thiamine. ChemBioChem 16, 819–826 (2015).
Berntsson, R. P. A. et al. Structural divergence of paralogous S components from ECF-type ABC transporters. Proc. Natl Acad. Sci. USA. 109, 13990–13995 (2012).
Anantharaman, V. & Aravind, L. Application of comparative genomics in the identification and analysis of novel families of membrane-associated receptors in bacteria. BMC Genomics 4, 1–20 (2003).
Qiu, J., Gasperotti, A., Sisattana, N., Zacharias, M. & Jung, K. The LytS-type histidine kinase BtsS is a 7-transmembrane receptor that binds pyruvate. MBio 14, e0108923 (2023).
Rohs, P. D. A. et al. Identification of potential regulatory domains within the MreC and MreD components of the cell elongation machinery. J. Bacteriol. 203, 10–1128 (2021).
Thibessard, A. et al. Identification of streptococcus thermophilus CNRZ368 genes involved in defense against superoxide stress. Appl. Environ. Microbiol. 70, 2220–2229 (2004).
Kruse, T., Bork-Jensen, J. & Gerdes, K. The morphogenetic MreBCD proteins of Escherichia coli form an essential membrane-bound complex. Mol. Microbiol. 55, 78–89 (2005).
Leaver, M. & Errington, J. Roles for MreC and MreD proteins in helical growth of the cylindrical cell wall in Bacillus subtilis. Mol. Microbiol. 57, 1196–1209 (2005).
White, C. L., Kitich, A. & Gober, J. W. Positioning cell wall synthetic complexes by the bacterial morphogenetic proteins MreB and MreD. Mol. Microbiol. 76, 616–633 (2010).
Ishii, E. & Eguchi, Y. Diversity in sensing and signaling of bacterial sensor histidine kinases. Biomolecules 11, 1–27 (2021).
Lobley, A., Sadowski, M. I. & Jones, D. T. pGenTHREADER and pDomTHREADER: New methods for improved protein fold recognition and superfamily discrimination. Bioinformatics 25, 1761–1767 (2009).
Pierri, C. L., Parisi, G. & Porcelli, V. Computational approaches for protein function prediction: a combined strategy from multiple sequence alignment to molecular docking-based virtual screening. Biochim. Biophys. Acta Proteins Proteom. 1804, 1695–1712 (2010).
Buchan, D. W. A. & Jones, D. T. The PSIPRED protein analysis workbench: 20 years on. Nucleic Acids Res 47, W402–W407 (2019).
von Heijne, G. The distribution of positively charged residues in bacterial inner membrane proteins correlates with the trans-membrane topology. EMBO J. 5, 3021–3027 (1986).
Slotboom, D. J. Structural and mechanistic insights into prokaryotic energy-coupling factor transporters. Nat. Rev. Microbiol. 12, 79–87 (2014).
Fried, L., Behr, S. & Jung, K. Identification of a target gene and activating stimulus for the YpdA/YpdB histidine kinase/response regulator system in Escherichia coli. J. Bacteriol. 195, 807–815 (2013).
Rodionov, D. A. et al. A novel class of modular transporters for vitamins in prokaryotes. J. Bacteriol. 91, 42–51 (2009).
Ettema, T. J. G., Lindås, A. C. & Bernander, R. An actin-based cytoskeleton in archaea. Mol. Microbiol. 80, 1052–1061 (2011).
Terwilliger, T. C. et al. AlphaFold predictions are valuable hypotheses, and accelerate but do not replace experimental structure determination. Nat. Methods 1–7 (2023). https://doi.org/10.1038/s41592-023-02087-4.
Puente-Lelievre, C. et al. Tertiary-interaction characters enable fast, model-based structural phylogenetics beyond the twilight zone. Preprint at bioRxiv. https://doi.org/10.1016/S0262-4079(09)62693-2. (2023).
von Heijne, G. The signal peptide. J. Membr. Biol. 103, 195–201 (1990).
Li, E., Wimley, W. C. & Hristova, K. Transmembrane helix dimerization: beyond the search for sequence motifs. Biochim. Biophys. Acta Biomembr. 1818, 183–193 (2012).
Zhou, F. X., Merianos, H. J., Brunger, A. T. & Engelman, D. M. Polar residues drive association of polyleucine transmembrane helices. Proc. Natl Acad. Sci. USA. 98, 2250–2255 (2001).
Call, M. E., Pyrdol, J., Wiedmann, M. & Wucherpfennig, K. W. The organizing principle in the formation of the T cell receptor-CD3 complex. Cell 111, 967–979 (2002).
Karpowich, N. K., Song, J. & Wang, D. N. An aromatic cap seals the substrate binding site in an ECF-type S subunit for riboflavin. J. Mol. Biol. 428, 3118–3130 (2016).
Simm, R., Morr, M., Kader, A., Nimtz, M. & Römling, U. GGDEF and EAL domains inversely regulate cyclic di-GMP levels and transition from sessibility to motility. Mol. Microbiol. 53, 1123–1134 (2004).
Stuffle, E. C., Johnson, M. S. & Watts, K. J. PAS domains in bacterial signal transduction. Curr. Opin. Microbiol. 61, 8–15 (2021).
Martinez, S. E., Beavo, J. A. & Hol, W. G. J. GAF domains: two-billion-year-old molecular switches that bind cyclic nucleotides. Mol. Interv. 2, 317–323 (2002).
Gao, R. & Stock, A. M. Biological insights from structures of two-component proteins. Annu. Rev. Microbiol. 63, 133–154 (2009).
Gao, R., Bouillet, S. & Stock, A. M. Structural basis of response regulator function. Annu. Rev. Microbiol. 73, 175–197 (2019).
Brüderlin, M. et al. Structural features discriminating hybrid histidine kinase Rec domains from response regulator homologs. Nat. Commun. 14, 1002 (2023).
Kennelly, P. J. Archaeal protein kinases and protein phosphatases: Insights from genomics and biochemistry. Biochem. J. 370, 373–389 (2003).
Esser, D. et al. Protein phosphorylation and its role in archaeal signal transductiona. FEMS Microbiol. Rev. 40, 625–647 (2016).
Nevo, Y. & Nelson, N. The NRAMP family of metal-ion transporters. Biochim. Biophys. Acta Mol. Cell Res. 1763, 609–620 (2006).
Thangaratnarajah, C., Rheinberger, J., Paulino, C. & Slotboom, D. J. Insights into the bilayer-mediated toppling mechanism of a folate-specific ECF transporter by cryo-EM. Proc. Natl Acad. Sci. USA. 118, 1–16 (2021).
Zhang, M. et al. Structure of a pantothenate transporter and implications for ECF module sharing and energy coupling of group II ECF transporters. Proc. Natl. Acad. Sci. USA. 111, 18560–18565 (2014).
Erkens, G. B. et al. The structural basis of modularity in ECF-type ABC transporters. Nat. Struct. Mol. Biol. 18, 755–760 (2011).
Mascher, T., Helmann, J. D. & Unden, G. Stimulus perception in bacterial signal-transducing histidine kinases. Microbiol. Mol. Biol. Rev. 70, 910–938 (2006).
Xu, Y. et al. Artificial intelligence: a powerful paradigm for scientific research. Innovation 2, 100179 (2021).
Perrakis, A. & Sixma, T. K. AI revolutions in biology. EMBO Rep. 22, 1–6 (2021).
Peeters, M. C., Van Westen, G. J. P., Li, Q. & Ijzerman, A. P. Importance of the extracellular loops in G protein-coupled receptors for ligand recognition and receptor activation. Trends Pharmacol. Sci. 32, 35–42 (2011).
Finkenwirth, F., Kirsch, F. & Eitinger, T. Solitary bio Y proteins mediate biotin transport into recombinant Escherichia coli. J. Bacteriol. 195, 4105–4111 (2013).
Zhu, Y., Zhou, C., Wang, Y. & Li, C. Transporter engineering for microbial manufacturing. Biotechnol. J. 15, 1–11 (2020).
Bousis, S., Setyawati, I., Diamanti, E., Slotboom, D. J. & Hirsch, A. K. H. Energy-coupling factor transporters as novel antimicrobial targets. Adv. Ther. 2, 1–17 (2019).
Ma, P. & Phillips-Jones, M. K. Membrane sensor histidine kinases: Insights from structural, ligand and inhibitor studies of full-length proteins and signalling domains for antibiotic discovery. Molecules 26, 5110 (2021).
Consortium, T. U. UniProt: the Universal Protein Knowledgebase in 2023. Nucleic Acids Res. 51, D523–D531 (2023).
Pettersen, E. F. et al. UCSF ChimeraX: Structure visualization for researchers, educators, and developers. Protein Sci. 30, 70–82 (2021).
Fabret, C., Feher, V. A. & Hoch, J. A. Two-component signal transduction in Bacillus subtilis: How one organism sees its world. J. Bacteriol. 181, 1975–1983 (1999).
Shang, F. et al. The Staphylococcus aureus GGDEF domain-containing protein, GdpS, influences protein A gene expression in a cyclic diguanylic acid-independent manner. Infect. Immun. 77, 2849–2856 (2009).
Paysan-Lafosse, T. et al. InterPro in 2022. Nucleic Acids Res. 51, D418–D427 (2023).
Almagro Armenteros, J. J. et al. SignalP 5.0 improves signal peptide predictions using deep neural networks. Nat. Biotechnol. 37, 420–423 (2019).
Fulyani, F. et al. Functional diversity of tandem substrate-binding domains in ABC transporters from pathogenic bacteria. Structure 21, 1879–1888 (2013).
Sievers, F. et al. Fast, scalable generation of high-quality protein multiple sequence alignments using Clustal Omega. Mol. Syst. Biol. 7, 539 (2011).
Crooks, G. E., Hon, G., Chandonia, J. M. & Brenner, S. E. WebLogo: A sequence logo generator. Genome Res. 14, 1188–1190 (2004).
Mirdita, M. et al. ColabFold: making protein folding accessible to all. Nat. Methods 19, 679–682 (2022).
Cianfrocco, M. A., Wong-Barnum, M., Youn, C., Wagner, R. & Leschziner, A. COSMIC2: A science gateway for cryo-electron microscopy structure determination. Proc. Pract. Exp. Adv. Res. Comput. 1–5 (2017). https://doi.org/10.1145/3093338.3093390.
Overbeek, R. et al. The SEED and the Rapid Annotation of microbial genomes using Subsystems Technology (RAST). Nucleic Acids Res. 42, 206–214 (2014).
Acknowledgements
M.P. thanks Natalia Skinder for the precious assistance with the visuals and constant support, Mark Nijland for insightful discussions, and Prof. Dr. Ciro Leonardo Pierri for valuable remarks on computational protein structure prediction. D.J.S acknowledges funding from the Dutch Research Council: NWO TOP Grant 714.018.003.
Author information
Authors and Affiliations
Contributions
M.P. and D.J.S. conceived the study; M.P. carried out the structural and bioinformatic analyses; M.P. and D.J.S wrote the manuscript.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review
Peer review information
Communications Biology thanks Benjamin Orlando, Dmitry Rodionov and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Primary Handling Editors: Tobias Goris. A peer review file is available.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Partipilo, M., Jan Slotboom, D. The S-component fold: a link between bacterial transporters and receptors. Commun Biol 7, 610 (2024). https://doi.org/10.1038/s42003-024-06295-2
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s42003-024-06295-2
- Springer Nature Limited