Modification of nuclear lamin proteins by a mevalonic acid derivative occurs in reticulocyte lysates and requires the cysteine residue of the C-terminal CXXM motif.

The C-terminus of nuclear lamins (CXXM) resembles a C-terminal motif (the CAAX box) of fungal mating factors and ras-related proteins. The CAAX box is subject to different types of post-translational modifications, including proteolytic processing, isoprenylation and carboxyl methylation. By peptide mapping we show that both chicken lamins A and B2 are processed proteolytically in vivo. However, whereas the entire CXXM motif is cleaved from lamin A, at most three C-terminal amino acids are removed from lamin B2. Following translation of cDNA-derived RNAs in reticulocyte lysates, lamin proteins specifically incorporate a derivative of [14C]mevalonic acid (MV), i.e. the precursor of a putative isoprenoid modification. Remarkably, no MV is incorporated into lamin B2 translated from a mutant cDNA encoding alanine instead of cysteine in the C-terminal CXXM motif. These results implicate this particular cysteine residue as the target for modification of lamin proteins by an isoprenoid MV derivative, and they indicate that isoprenylation is amenable to studies in cell-free systems. Moreover, our observations suggest that C-terminal processing of newly synthesized nuclear lamins is a multi-step process highly reminiscent of the pathway elaborated recently for ras-related proteins.     

Identification of the carboxyl-terminal membrane-anchoring region of HPC-1/syntaxin 1A with the substituted-cysteine-accessibility method and monoclonal antibodies

HPC-1/syntaxin 1A is a member of the syntaxin family, and functions at the plasma membrane during membrane fusion as the target-soluble N-ethylmaleimide-sensitive factor-attachment protein receptor (t-SNARE). We identified the membrane-anchoring region of HPC-1/syntaxin 1A, and examined its role in anchoring of a protein to the plasma membrane. A series of mutants was created from a cysteine-less mutant of HPC-1/syntaxin 1A by substitution of each residue at the C-terminus with cysteine. The accessibility of the thiol-groups in each mutant was analyzed in vivo. The cysteine (C145) within the N-terminal cytosolic segment was labeled, but not that at C271 or C272, or any of those introduced at the C-terminus. The addition of additional residues to the C-terminal tail of HPC-1/syntaxin 1A allowed labeling by thiol-specific reagents. A monoclonal antibody directed against the C-terminal tail peptide did not react with the protein located at the plasma membrane. In addition, subcellular fractionation and immunocytochemical analyses with various transmembrane mutants showed that the C-terminal tail comprising eight amino acids is essential for anchoring of HPC-1/syntaxin 1A to the plasma membrane. These results indicate that the C-terminal membrane-anchoring region, which comprises 23 amino acids, does not traverse the lipid-bilayer and that the C-terminal tail is essential for anchoring of HPC-1/syntaxin 1A to the plasma membrane.     

Role of paired basic residues of protein C-termini in phospholipid binding

It is a well known phenomenon that the occurrence of several distinct amino acids at the C-terminus of proteins is non-random. We have analysed all Saccharomyces cerevisiae proteins predicted by computer databases and found lysine to be the most frequent residue both at the last (-1) and at the penultimate amino acid (-2) positions. To test the hypothesis that C-terminal basic residues efficiently bind to phospholipids we randomly expressed GST-fusion proteins from a yeast genomic library. Fifty-four different peptide fragments were found to bind phospholipids and 40% of them contained lysine/arginine residues at the (-1) or (-2) positions. One peptide showed high sequence similarity with the yeast protein Sip18p. Mutational analysis revealed that both C-terminal lysine residues of Sip18p are essential for phospholipid-binding in vitro. We assume that basic amino acid residues at the (-1) and (-2) positions in C-termini are suitable to attach the C-terminus of a given protein to membrane components such as phospholipids, thereby stabilizing the spatial structure of the protein or contributing to its subcellular localization. This mechanism could be an additional explanation for the C-terminal amino acid bias observed in proteins of several species.     

RINT-1, a novel Rad50-interacting protein, participates in radiation-induced G(2)/M checkpoint control

Rad50, an structural maintenance of chromosomes (SMC) protein family member, participates in a variety of cellular processes, including DNA double-strand break repair, cell cycle checkpoint activation, telomere maintenance, and meiosis. Disruption of Rad50 in mice leads to lethality during early embryogenesis, indicating its essential function in normal proliferating cells. In addition to its ability to form a complex with the DNA double-strand break repair proteins Mre11 and NBS1, Rad50 may interact with other cellular proteins to execute its full range of biological activities. A novel 87-kDa protein named RINT-1 was identified using the C-terminal region of human Rad50 as the bait in a yeast two-hybrid screen. Human RINT-1 shares sequence homology with a novel protein identified in Drosophila melanogaster, including a coiled-coil domain within its N-terminal 150 amino acids, a conserved central domain of about 350 amino acids, and a C-terminal region of 90 amino acids exhibiting 35--38% identity. The conserved central and C-terminal regions of RINT-1 are required for its interaction with Rad50. While Rad50 and RINT-1 are both expressed throughout the cell cycle, RINT-1 specifically binds to Rad50 only during late S and G(2)/M phases, suggesting that RINT-1 may be involved in cell cycle regulation. Consistent with this possibility, MCF-7 cells expressing an N-terminally truncated RINT-1 protein displayed a defective radiation-induced G(2)/M checkpoint. These results suggest that RINT-1 may play a role in the regulation of cell cycle control after DNA damage.     

Protein palmitoylation and subcellular trafficking

Protein S-palmitoylation, the covalent lipid modification of the side chain of Cys residues with the 16-carbon fatty acid palmitate, is the most common acylation of proteins in eukaryotic cells. This post-translational modification provides an important mechanism for regulating protein subcellular localization, stability, trafficking, translocation to lipid rafts, aggregation, interaction with effectors and other aspects of protein function. In addition, N-terminal myristoylation and C-terminal prenylation, two well-studied post-translational modifications, frequently precede protein S-palmitoylation at a nearby spot of the polypeptide chain. Whereas N-myristoylation and prenylation are considered essentially irreversible attachments, S-palmitoylation is a tightly regulated, reversible modification. In addition, the unique reversibility of protein palmitoylation also allows proteins to rapidly shuttle between intracellular membrane compartments in a process controlled, in some cases, by the DHHC family of palmitoyl transferases. Recent cotransfection experiments using the DHHC family of protein palmitoyl transferases as well as RNA interference results have revealed that these enzymes, frequently localized to the Golgi apparatus, tightly control subcellular trafficking of acylated proteins. In this article we will give an overview of how protein palmitoylation regulates protein trafficking and subcellular localization.     

Interaction with PDZK1 is required for expression of organic anion transporting protein 1A1 on the hepatocyte surface

Although many organic anion transport protein (Oatp) family members have PDZ consensus binding sites at their C termini, the functional significance is unknown. In the present study, we utilized rat Oatp1a1 (NM_017111) as a prototypical member of this family to examine the mechanism governing its subcellular trafficking. A peptide corresponding to the C-terminal 16 amino acids of rat Oatp1a1 was used to affinity-isolate interacting proteins from rat liver cytosol. Protein mass fingerprinting identified PDZK1 as the major interacting protein. This was confirmed by immunoprecipitation of an Oatp1a1-PDZK1 complex from cotransfected 293T cells as well as from native rat liver membrane extracts. Oatp1a1 bound predominantly to the first and third PDZ binding domains of PDZK1, whereas the high density lipoprotein receptor, scavenger receptor B type I binds to the first domain. Although it is possible that PDZK1 forms a complex with these two integral membrane proteins, this did not occur, suggesting that as yet undescribed factors lead to selectivity in the interaction of these protein ligands with PDZK1. Oatp1a1 protein expression was near normal in PDZK1 knock-out mouse liver. However, it was located predominantly in intracellular structures, in contrast to its normal basolateral plasma membrane distribution. Plasma disappearance of the Oatp1a1 ligand [35S]sulfobromophthalein was correspondingly delayed in knock-out mice. These studies show a critical role for oligomerization of Oatp1a1 with PDZK1 for its proper subcellular localization and function. Because its ability to transport substances into the cell requires surface expression, this must be considered in any assessment of physiologic function.     

The effector domain of Rab6, plus a highly hydrophobic C terminus, is required for Golgi apparatus localization.

C-terminal lipid modifications are essential for the interaction of Ras-related proteins with membranes. While all Ras proteins are farnesylated and some palmitoylated, the majority of other Ras-related proteins are geranylgeranylated. One such protein, Rab6, is associated with the Golgi apparatus and has a C-terminal CXC motif that is geranylgeranylated on both cysteines. We show here that farnesylation alone cannot substitute for geranylgeranylation in targeting Rab6 to the Golgi apparatus and that whereas Ras proteins that are farnesylated and palmitoylated are targeted to the plasma membrane, mutant Rab proteins that are both farnesylated and palmitoylated associate with the Golgi apparatus. Using chimeric Ras-Rab proteins, we find that there are sequences in the N-terminal 71 amino acids of Rab6 which are required for Golgi complex localization and show that these sequences comprise or include the effector domain. The C-terminal hypervariable domain is not essential for the Golgi complex targeting of Rab6 but is required to prevent prenylated and palmitoylated Rab6 from localizing to the plasma membrane. Functional analysis of these mutant Rab6 proteins in Saccharomyces cerevisiae shows that wild-type Rab6 and C-terminal mutant Rab6 proteins which localize to the Golgi apparatus in mammalian cells can complement the temperature-sensitive phenotype of ypt6 null mutants. Interestingly, therefore, the C-terminal hypervariable domain of Rab6 is not required for this protein to function in S. cerevisiae.     

PDZ-binding motifs are unable to ensure correct polarized protein distribution in the absence of additional localization signals

The C-terminal PDZ-binding motifs are required for polarized apical/basolateral localization of many membrane proteins. To determine the specificity of the PDZ-binding motifs in establishing cellular distribution, we utilized a 111-amino acid region from the C-terminus of cystic fibrosis transmembrane conductance regulator (CFTR) that is able to direct apical localization of fused reporter proteins. Substitution of the C-terminal PDZ-binding motif of CFTR with corresponding motifs necessary for basolateral localization of other membrane proteins did not lead to the redistribution of the fusion protein to the basolateral membrane. Instead, some fusion proteins remained localized to the apical membrane, whereas others showed no specific distribution. The specificity of the PDZ-based interactions was substantially increased when specific amino acids located upstream of the classical PDZ-binding motifs were included. However, even the presence of a longer C-terminal motif from a basolateral protein could not ensure basolateral distribution of the fusion protein. Our results indicate that the C-terminal PDZ-binding motifs are not the primary signals for polarized protein distribution, although they are required for targeting and/or stabilization of protein at the given location.     

PC1/3, PC2 and PC5/6A are targeted to dense core secretory granules by a common mechanism

There are seven members of the proprotein convertase (PC) family of secreted serine proteases that cleave their substrates at basic amino acids, thereby activating a variety of hormones, growth factors, and viruses. PC1/3, PC2 and PC5/6A are the only members of the PC family that are targeted to dense core secretory granules, where they carry out the processing of proteins that are secreted from the cell in a regulated manner. Previous studies have identified alpha-helices in the C-termini of the PC1/3 and PC2 proteases that are required for this subcellular targeting. In the current study, we demonstrate that a predicted alpha-helix in the C-terminus of PC5/6A is also critical for the ability of this domain to target a heterologous protein to the regulated secretory pathway of mouse endocrine AtT-20 cells. Analysis of the subcellular distribution of fusion proteins containing the C-terminal domains of PC1/3, PC2 and PC5/6A confirmed that all three domains have the capacity to redirect a constitutively secreted protein to the granule-containing cytoplasmic extensions. Analysis of the predicted structures formed by these three granule-sorting helices shows a correlation between their granule-sorting efficiency and the clustering of hydrophobic amino acids in their granule-targeting helices.     

Harvey murine sarcoma virus p21 ras protein: biological and biochemical significance of the cysteine nearest the carboxy terminus.

Previous studies of premature chain termination mutants and in frame deletion mutants of the p21 ras transforming protein encoded by the transforming gene of Harvey murine sarcoma virus (Ha-MuSV) have suggested that the C terminus is required for cellular transformation, lipid binding, and membrane localization. We have now further characterized the post-translational processing of these mutants and have also studied two C-terminal v-rasH point mutants: one encodes serine in place of cysteine-186, the other threonine for valine-187. The Thr-187 mutant was transformation-competent, and its p21 protein was processed normally, as was the p21 encoded by a transformation-competent deletion mutant from which amino acids 166-175 had been deleted. The Ser-186 mutant was defective for transformation. The p21s encoded by the Ser-186 mutant and by the previously described transformation-defective mutants did not undergo the posttranslational processing common to biologically active ras proteins: their electrophoretic migration rate did not change, they remained in the cytosol, and they failed to bind lipid. Since the cell-encoded ras proteins also contain this cysteine, we conclude that this amino acid residue is required for all ras proteins.     

Polyisoprenylation of the CAAX motif--an in vitro protein synthesis study

A number of proteins, including most nuclear lamins, certain fungal mating pheromones, G-protein gamma-subunits and ras proteins, contain a C-terminal cysteine-aliphatic-aliphatic-undefined amino acid (CAAX) motif which is thought to be a roughly defined consensus sequence capable of directing a series of posttranslational events, beginning with the addition of a polyisoprene moiety to the cysteine. So far such a motif has been found in every protein known to have this type of modification. We have utilized the rabbit reticulocyte lysate translation system, which is capable of carrying out the polyisoprene modification in vitro, to investigate features of the C-terminal motif which affect its suitability as a substrate. We demonstrate that a cysteine is only isoprenylated when situated at position -4 from the C-terminus. We further show that the presence of a glycine at position -3 or a terminal aromatic residue, features typical of some G-protein alpha subunits, cause a reduction and abolition respectively of isoprenylation.     

Predicting the subcellular localization of human proteins using machine learning and exploratory data analysis

Identifying the subcellular localization of proteins is particularly helpful in the functional annotation of gene products. In this study, we use Machine Learning and Exploratory Data Analysis （EDA） techniques to examine and characterize amino acid sequences of human proteins localized in nine cellular compartments. A dataset of 3,749 protein sequences representing human proteins was extracted from the SWISS-PROT database. Feature vectors were created to capture specific amino acid sequence characteristics. Relative to a Support Vector Machine, a Multi-layer Perceptron, and a Naive Bayes classifier, the C4.5 Decision Tree algorithm was the most consistent performer across all nine compartments in reliably predicting the subcellular localization of proteins based on their amino acid sequences （average Precision=0.88; average Sensitivity=0.86）. Furthermore, EDA graphics characterized essential features of proteins in each compartment. As examples, proteins localized to the plasma membrane had higher proportions of hydrophobic amino acids; cytoplasmic proteins had higher proportions of neutral amino acids; and mitochondrial proteins had higher proportions of neutral amino acids and lower proportions of polar amino acids. These data showed that the C4.5 classifier and EDA tools can be effective for characterizing and predicting the subcellular localization of human proteins based on their amino acid sequences.     

Membrane topology and essential amino acid residues of Phs1, a 3-hydroxyacyl-CoA dehydratase involved in very long-chain fatty acid elongation

Yeast Phs1 is the 3-hydroxyacyl-CoA dehydratase that catalyzes the third reaction of the four-step cycle in the elongation of very long-chain fatty acids (VLCFAs). In yeast, the hydrophobic backbone of sphingolipids, ceramide, consists of a long-chain base and an amide-linked C26 VLCFA. Therefore, defects in VLCFA synthesis would be expected to greatly affect sphingolipid synthesis. In fact, in this study we found that reduced Phs1 levels result in significant impairment of the conversion of ceramide to inositol phosphorylceramide. Phs1 proteins are conserved among eukaryotes, constituting a novel protein family. Phs1 family members exhibit no sequence similarity to other dehydratase families, so their active site sequence and catalytic mechanism have been completely unknown. Here, by mutating 22 residues conserved among Phs1 family members, we identified six amino acid residues important in Phs1 function, two of which (Tyr-149 and Glu-156) are indispensable. We also examined the membrane topology of Phs1 using an N-glycosylation reporter assay. Our results suggest that Phs1 is a membrane-spanning protein that traverses the membrane six times and has an N terminus and C terminus facing the cytosol. The important amino acids are concentrated in or near two of the six proposed transmembrane regions. Thus, we also propose a catalytic mechanism for Phs1 that is not unlike mechanisms used by other hydratases active in lipid synthesis.     

Global analysis of protein palmitoylation in yeast

Protein palmitoylation is a reversible lipid modification that regulates membrane tethering for key proteins in cell signaling, cancer, neuronal transmission, and membrane trafficking. Palmitoylation has proven to be a difficult study: Specifying consensuses for predicting palmitoylation remain unavailable, and first-example palmitoylation enzymes--i.e., protein acyltransferases (PATs)--were identified only recently. Here, we use a new proteomic methodology that purifies and identifies palmitoylated proteins to characterize the palmitoyl proteome of the yeast Saccharomyces cerevisiae. Thirty-five new palmitoyl proteins are identified, including many SNARE proteins and amino acid permeases as well as many other participants in cellular signaling and membrane trafficking. Analysis of mutant yeast strains defective for members of the DHHC protein family, a putative PAT family, allows a matching of substrate palmitoyl proteins to modifying PATs and reveals the DHHC family to be a family of diverse PAT specificities responsible for most of the palmitoylation within the cell.     

Characterization of the C-terminal ER membrane anchor of PTP1B

The tyrosine phosphatase PTP1B is an important regulator of cell function. In living cells PTP1B activity is restricted to the vicinity of the endoplasmic reticulum (ER) by post-translational C-terminal attachment of PTP1B to the ER membrane network. In our study we investigated the membrane anchor of PTP1B by use of EGFP fusion proteins. We demonstrate that the membrane anchor of PTP1B cannot be narrowed down to a unique amino acid sequence with a defined start and stop point but rather is moveable within several amino acids. Removal of up to seven amino acids from the C-terminus, as well as exchange of single amino acids in the putative transmembrane sequence did not influence subcellular localization of PTP1B. With the method of bimolecular fluorescence complementation we could demonstrate dimerization of PTP1B in vivo. Homodimerization was, in contrast to other tail-anchored proteins, not dependent on the membrane anchor. Our data demonstrate that the C-terminal membrane anchor of PTP1B is formed by a combination of a single stretch transmembrane domain (TMD) followed by a tail. TMD and tail length are variable and there are no sequence-specific features. Our data for PTP1B are consistent with a concept that explains the ER membrane anchor of tail-anchored proteins as a physicochemical structure.     

Properties of Rab5 N-terminal domain dictate prenylation of C-terminal cysteines.

Rab5 is a Ras-related GTP-binding protein that is post-translationally modified by prenylation. We report here that an N-terminal domain contained within the first 22 amino acids of Rab5 is critical for efficient geranylgeranylation of the protein's C-terminal cysteines. This domain is immediately upstream from the "phosphate binding loop" common to all GTP-binding proteins and contains a highly conserved sequence recognized among members of the Rab family, referred to here as the YXYLFK motif. A truncation mutant that lacks this domain (Rab5(23-215) fails to become prenylated. However, a chimeric peptide with the conserved motif replacing cognate Rab5 sequence (MAYDYLFKRab5(23-215) does become post-translationally modified, demonstrating that the presence of this simple six amino acid N-terminal element enables prenylation at Rab5's C-terminus. H-Ras/Rab5 chimeras that include the conserved YXYLFK motif at the N-terminus do not become prenylated, indicating that, while this element may be necessary for prenylation of Rab proteins, it alone is not sufficient to confer properties to a heterologous protein to enable substrate recognition by the Rab geranylgeranyl transferase. Deletion analysis and studies of point mutants further reveal that the lysine residue of the YXYLFK motif is an absolute requirement to enable geranylgeranylation of Rab proteins. Functional studies support the idea that this domain is not required for guanine nucleotide binding since prenylation-defective mutants still bind GDP and are protected from protease digestion in the presence of GTP gamma S. We conclude that the mechanism of Rab geranylgeranylation involves key elements of the protein's tertiary structure including a conserved N-terminal amino acid motif (YXYLFK) that incorporates a critical lysine residue.     

Membrane topology of the murine fatty acid transport protein 1

The murine fatty acid transport protein (FATP1) was identified in an expression cloning screen for proteins that facilitate transport of fatty acids across the plasma membranes of mammalian cells. Hydropathy analysis of this protein suggests a model in which FATP1 has multiple membrane-spanning domains. To test this model, we inserted a hemagglutinin epitope tag at the amino terminus or a FLAG tag at the carboxyl terminus of the FATP1 cDNA and expressed these constructs in NIH 3T3 cells. Both tagged constructs produce proteins of the expected molecular masses and are functional in fatty acid import assays. Indirect immunofluorescence studies with selective permeabilization conditions and protease protection studies of sealed membrane vesicles from cells expressing epitope-tagged FATP1 were performed. These experiments show that the extreme amino terminus of tagged FATP1 is oriented toward the extracellular space, whereas the carboxyl terminus faces the cytosol. Additionally, enhanced green fluorescent protein fusion constructs containing predicted membrane-associated or soluble portions of FATP1 were expressed in Cos7 cells and analyzed by immunofluorescence and subcellular fractionation. These experiments demonstrate that amino acids 1-51, 52-100, and 101-190 contain signals for integral association with the membrane, whereas residues 258-313 and 314-475 are only peripherally membrane-associated. Amino acid residues 191-257 and 476-646 do not direct membrane association and likely face the cytosol. Taken together, these data support a model of FATP1 as a polytopic membrane protein with at least one transmembrane and multiple membrane-associated domains. This study provides the first experimental evidence for topology of a member of the family of plasma membrane fatty acid transport proteins.     

Regulation of a Drosophila melanogaster cGMP-specific phosphodiesterase by prenylation and interaction with a prenyl-binding protein

Post-translational modification by isoprenylation is a pivotal process for the correct functioning of many signalling proteins. The Drosophila melanogaster cGMP-PDE (cGMP-specific phosphodiesterase) DmPDE5/6 possesses a CaaX-box prenylation signal motif, as do several novel cGMP-PDEs from insect and echinoid species (in CaaX, C is cysteine, a is an aliphatic amino acid and X is 'any' amino acid). DmPDE5/6 is prenylated in vivo at Cys(1128) and is localized to the plasma membrane when expressed in Drosophila S2 cells. Site-directed mutagenesis of the prenylated cysteine residue (C1128S-DmPDE5/6), pharmacological inhibition of prenylation or co-expression of DmPrBP (Drosophila prenyl-binding protein)/delta each alters the subcellular localization of DmPDE5/6. Thus prenylation constitutes a critical post-translational modification of DmPDE5/6 for membrane targeting. Co-immunoprecipitation and subcellular-fractionation experiments have shown that DmPDE5/6 interacts with DmPrBP/delta in Drosophila S2 cells. Transgenic lines allow targeted expression of tagged prenylation-deficient C1128S-DmPDE5/6 in Type I (principal) cells in Drosophila Malpighian tubules, an in vivo model for DmPDE5/6 function. In contrast with wild-type DmPDE5/6, which was exclusively associated with the apical membrane, the C1128S-DmPDE5/6 mutant form was located primarily in the cytosol, although some residual association occurred at the apical membrane. Despite the profound change in intracellular localization of C1128S-DmPDE5/6, active transport of cGMP is affected in the same way as it is by DmPDE5/6. This suggests that, in addition to prenylation and interaction with DmPrBP/delta, further functional membrane-targeting signals exist within DmPDE5/6.     

Bacillus subtilis SepF Binds to the C-Terminus of FtsZ

Bacterial cell division is mediated by a multi-protein machine known as the "divisome", which assembles at the site of cell division. Formation of the divisome starts with the polymerization of the tubulin-like protein FtsZ into a ring, the Z-ring. Z-ring formation is under tight control to ensure bacteria divide at the right time and place. Several proteins bind to the Z-ring to mediate its membrane association and persistence throughout the division process. A conserved stretch of amino acids at the C-terminus of FtsZ appears to be involved in many interactions with other proteins. Here, we describe a novel pull-down assay to look for binding partners of the FtsZ C-terminus, using a HaloTag affinity tag fused to the C-terminal 69 amino acids of B. subtilis FtsZ. Using lysates of Escherichia coli overexpressing several B. subtilis cell division proteins as prey we show that the FtsZ C-terminus specifically pulls down SepF, but not EzrA or MinC, and that the interaction depends on a conserved 16 amino acid stretch at the extreme C-terminus. In a reverse pull-down SepF binds to full-length FtsZ but not to a FtsZΔC16 truncate or FtsZ with a mutation of a conserved proline in the C-terminus. We show that the FtsZ C-terminus is required for the formation of tubules from FtsZ polymers by SepF rings. An alanine-scan of the conserved 16 amino acid stretch shows that many mutations affect SepF binding. Combined with the observation that SepF also interacts with the C-terminus of E. coli FtsZ, which is not an in vivo binding partner, we propose that the secondary and tertiary structure of the FtsZ C-terminus, rather than specific amino acids, are recognized by SepF.