- Volume 151, Issue 5, 2005

The dif operon is essential for fruiting body formation, fibril (exopolysaccharide) production and social motility of Myxococcus xanthus. The dif locus contains a gene cluster homologous to chemotaxis genes such as mcp (difA), cheW (difC), cheY (difD), cheA (difE) and cheC (difF), as well as an unknown ORF called difB. This study used yeast two-hybrid analysis to investigate possible interactions between Dif proteins, and determined that DifA, C, D and E interact in a similar fashion to chemotaxis proteins of Escherichia coli and Bacillus subtilis. It also showed that DifF interacted with DifD, and that the novel protein DifB did not interact with Dif proteins. Furthermore, DifA–F proteins were used to determine other possible protein–protein interactions in the M. xanthus genomic library. The authors not only confirmed the specific interactions among known Dif proteins, but also discovered two novel interactions between DifE and Nla19, and DifB and YidC, providing some new information about the Dif signalling pathway. Based on these findings, a model for the Dif signalling pathway is proposed.

Creating and maintaining cell polarity are complex processes that are not fully understood. Fungal hyphal tip growth is a highly polarized and dynamic process involving both F-actin and microtubules (MTs), but the behaviour and roles of the latter are unclear. To address this issue, MT dynamics and subunit distribution were analysed in a strain of Aspergillus nidulans expressing GFP–α-tubulin. Apical MTs are the most dynamic, the bulk of which move tipwards from multiple subapical spindle pole bodies, the only clear region of microtubule nucleation detected. MTs populate the apex predominantly by elongation at rates about three times faster than tip extension. This polymerization was facilitated by the tipward migration of MT subunits, which generated a tip-high gradient. Subapical regions of apical cells showed variable tubulin subunit distributions, without tipward flow, while subapical cells showed even tubulin subunit distribution and low MT dynamics. Short MTs, of a similar size to those reported in axons, also occasionally slid into the apex. During mitosis in apical cells, MT populations at the tip varied. Cells with less distance between the tip and the first nucleus were more likely to loose normal MT populations and dynamics. Reduced MTs in the tip, during mitosis or after exposure to the MT inhibitor carbendazim (MBC), generally correlated with reduced, but continuing growth and near-normal tip morphology. In contrast, the actin-disrupting agent latrunculin B reduced growth rates much more severely and dramatically distorted tip morphology. These results suggest substantial independence between MTs and hyphal tip growth and a more essential role for F-actin. Among MT-dependent processes possibly contributing to tip growth is the transportation of vesicles. However, preliminary ultrastructural data indicated a lack of direct MT–organelle interactions. It is suggested that the population of dynamic apical MTs enhance migration of the ‘cytomatrix’, thus ensuring that organelles and proteins maintain proximity to the constantly elongating tip.

Pseudomonas aeruginosa is an opportunistic pathogen which causes a variety of diseases, including respiratory tract infections in patients suffering from cystic fibrosis. Therapeutic treatment of P. aeruginosa infections is still very difficult because the bacteria exhibit high intrinsic resistance against a variety of different antibiotics and, in addition, form stable biofilms, e.g. in the human lung. Several virulence factors are produced by P. aeruginosa, among them the two lectins LecA and LecB, which exert different cytotoxic effects on respiratory epithelial cells and presumably facilitate bacterial adhesion to the airway mucosa. Here, the physiology has been studied of the lectin LecB, which binds specifically to l-fucose. A LecB-deficient P. aeruginosa mutant was shown to be impaired in biofilm formation when compared with the wild-type strain, suggesting an important role for LecB in this process. This result prompted an investigation of the subcellular localization of LecB by cell fractionation and subsequent immunoblotting. The results show that LecB is abundantly present in the bacterial outer-membrane fraction. It is further demonstrated that LecB could be released specifically by treatment of the outer-membrane fraction with p-nitrophenyl α-l-fucose, whereas treatment with d-galactose had no effect. In contrast, a LecB protein carrying the mutation D104A, which results in a defective sugar-binding site, was no longer detectable in the membrane fraction, suggesting that LecB binds to specific carbohydrate ligands located at the bacterial cell surface. Staining of biofilm cells using fluorescently labelled LecB confirmed the presence of these ligands.

Quorum sensing (QS) communication systems are thought to afford bacteria with a mechanism to strategically cause disease. One example is Pseudomonas aeruginosa, which infects immunocompromised individuals such as cystic fibrosis patients. The authors have previously documented that blockage of the QS systems not only attenuates Ps. aeruginosa but also renders biofilms highly susceptible to treatment with conventional antibiotics. Filamentous fungi produce a battery of secondary metabolites, some of which are already in clinical use as antimicrobial drugs. Fungi coexist with bacteria but lack active immune systems, so instead rely on chemical defence mechanisms. It was speculated that some of these secondary metabolites could interfere with bacterial QS communication. During a screening of 100 extracts from 50 Penicillium species, 33 were found to produce QS inhibitory (QSI) compounds. In two cases, patulin and penicillic acid were identified as being biologically active QSI compounds. Their effect on QS-controlled gene expression in Ps. aeruginosa was verified by DNA microarray transcriptomics. Similar to previously investigated QSI compounds, patulin was found to enhance biofilm susceptibility to tobramycin treatment. Ps. aeruginosa has developed QS-dependent mechanisms that block development of the oxidative burst in PMN neutrophils. Accordingly, when the bacteria were treated with either patulin or penicillic acid, the neutrophils became activated. In a mouse pulmonary infection model, Ps. aeruginosa was more rapidly cleared from the mice that were treated with patulin compared with the placebo group.

Pseudomonas sp. strain PCL1171 undergoes reversible colony phase variation between opaque phase I and translucent phase II colonies, which is dependent on spontaneous mutations in the regulatory genes gacA and gacS. Mutation of the mutS gene and constitutive expression of rpoS increases the frequency at which gac mutants appear 1000- and 10-fold, respectively. Experiments were designed to study the relationship between gacS, rpoS and mutS. These studies showed that (i) a functional gac system is required for the expression of rpoS, (ii) RpoS suppresses the expression of mutS and therefore increases the frequency of gac mutants, and (iii) upon mutation of rpoS and gacS, the expression of mutS is increased. Mutation of gacS abolishes suppression of mutS expression in stationary growth, suggesting that additional gac-dependent factors are involved in this suppression. In conclusion, inefficient mutation repair via MutS, of which the expression is influenced by gacA/S itself and by rpoS in combination with other factors, contributes to the high frequency of mutations accumulating in gacA/S. The role of RpoS in the growth advantage of a gac mutant was analysed, and mutation of rpoS only reduced the length of the lag phase, but did not affect the growth rate, suggesting a role for both RpoS and a reduction of metabolic load in the growth advantage of a gac mutant.

NADP-dependent glutamate dehydrogenase (NADP-GDH) mediates fungal ammonium assimilation through reductive synthesis of glutamate from 2-oxoglutarate. By virtue of its position at the interface of carbon and nitrogen metabolism, biosynthetic NADP-GDH is a potential candidate for metabolic control. In order to facilitate characterization, a new and effective dye-affinity method was devised to purify NADP-GDH from two aspergilli, Aspergillus niger and Aspergillus nidulans. The A. niger NADP-GDH was characterized at length and its kinetic interaction constants with glutamate (K m 34·7 mM) and ammonium (K m 1·05 mM; K i 0·4 mM) were consistent with an anabolic role. Isophthalate, 2-methyleneglutarate and 2,4-pyridinedicarboxylate were significant inhibitors, with respective K i values of 6·9, 9·2 and 202·0 μM. The A. niger enzyme showed allosteric properties and a sigmoid response (n H=2·5) towards 2-oxoglutarate saturation. The co-operative behaviour was a feature common to NADP-GDH from Aspergillus awamori, A. nidulans and Aspergillus oryzae. NADP-GDH may therefore be a crucial determinant in adjusting 2-oxoglutarate flux between the tricarboxylic acid cycle and glutamate biosynthesis in aspergilli.

The gene encoding a putative high-potential iron–sulfur protein (HiPIP) from the strictly acidophilic and chemolithoautotrophic Acidithiobacillus ferrooxidans ATCC 33020 has been cloned and sequenced. This potential HiPIP was overproduced in the periplasm of the neutrophile and heterotroph Escherichia coli. As shown by optical and EPR spectra and by electrochemical studies, the recombinant protein has all the biochemical properties of a HiPIP, indicating that the iron–sulfur cluster was correctly inserted. Translocation of this protein in the periplasm of E. coli was not detected in a ΔtatC mutant, indicating that it is dependent on the Tat system. The genetic organization of the iro locus in strains ATCC 23270 and ATCC 33020 is different from that found in strains Fe-1 and BRGM. Indeed, in A. ferrooxidans ATCC 33020 and ATCC 23270 (the type strain), iro was not located downstream from purA but was instead downstream from petC2, encoding cytochrome c 1 from the second A. ferrooxidans cytochrome bc 1 complex. These findings underline the genotypic heterogeneity within the A. ferrooxidans species. The results suggest that Iro transfers electrons from a cytochrome bc 1 complex to a terminal oxidase, as proposed for the HiPIP in photosynthetic bacteria.

Azo dyes are a predominant class of colourants used in tattooing, cosmetics, foods and consumer products. A gene encoding NADPH-flavin azoreductase (Azo1) from the skin bacterium Staphylococcus aureus ATCC 25923 was identified and overexpressed in Escherichia coli. RT-PCR results demonstrated that the azo1 gene was constitutively expressed at the mRNA level in S. aureus. Azo1 was found to be a tetramer with a native molecular mass of 85 kDa containing four non-covalently bound FMN. Azo1 requires NADPH, but not NADH, as an electron donor for its activity. The enzyme was resolved to dimeric apoprotein by removing the flavin prosthetic groups using hydrophobic-interaction chromatography. The dimeric apoprotein was reconstituted on-column and in free stage with FMN, resulting in the formation of a fully functional native-like tetrameric enzyme. The enzyme cleaved the model azo dye 2-[4-(dimethylamino)phenylazo]benzoic acid (Methyl Red) into N,N-dimethyl-p-phenylenediamine and 2-aminobenzoic acid. The apparent K m values for NADPH and Methyl Red substrates were 0·074 and 0·057 mM, respectively. The apparent V max was 0·4 μM min−1 (mg protein)−1. Azo1 was also able to metabolize Orange II, Amaranth, Ponceau BS and Ponceau S azo dyes. Azo1 represents the first azoreductase to be identified and characterized from human skin microflora.

Direct estimation of the molecular mass of single molecular species of trehalose 6-monomycolate (TMM), a ubiquitous cell-wall component of mycobacteria, was performed by matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectrometry. When less than 1 μg TMM was analysed by MALDI-TOF mass spectrometry, quasimolecular ions [M+Na]+ of each molecular species were demonstrated and the numbers of carbons and double bonds (or cyclopropane rings) were determined. Since the introduction of oxygen atoms such as carbonyl, methoxy and ester groups yielded the appropriate shift of mass ions, the major subclasses of mycolic acid (α, methoxy, keto and wax ester) were identified without resorting to hydrolytic procedures. The results showed a marked difference in the molecular species composition of TMM among mycobacterial species. Unexpectedly, differing from other mycoloyl glycolipids, TMM from Mycobacterium tuberculosis showed a distinctive mass pattern, with abundant odd-carbon-numbered monocyclopropanoic (or monoenoic) α-mycolates besides dicyclopropanoic mycolate, ranging from C75 to C85, odd- and even-carbon-numbered methoxymycolates ranging from C83 to C94 and even- and odd-carbon-numbered ketomycolates ranging from C83 to C90. In contrast, TMM from Mycobacterium bovis (wild strain and BCG substrains) possessed even-carbon-numbered dicyclopropanoic α-mycolates. BCG Connaught strain lacked methoxymycolates almost completely. These results were confirmed by MALDI-TOF mass analysis of mycolic acid methyl esters liberated by alkaline hydrolysis and methylation of the original TMM. Wax ester-mycoloyl TMM molecular species were demonstrated for the first time as an intact form in the Mycobacterium avium–intracellulare group, M. phlei and M. flavescens. The M. avium–intracellulare group possessed predominantly C85 and C87 wax ester-mycoloyl TMM, while M. phlei and the rapid growers tested contained C80, C81, C82 and C83 wax ester-mycoloyl TMM. This technique has marked advantages in the rapid analysis of not only intact glycolipid TMM, but also the mycolic acid composition of each mycobacterial species, since it does not require any degradation process.

Staphylococcus epidermidis is a ubiquitous human skin commensal that has emerged as a major cause of foreign-body infections. Eleven genes encoding putative cell-wall-anchored proteins were identified by computer analysis of the publicly available S. epidermidis unfinished genomic sequence. Four genes encode previously described proteins (Aap, Bhp, SdrF and SdrG), while the remaining seven have not been characterized. Analysis of primary sequences of the Staphylococcus epidermidis surface (Ses) proteins indicates that they have a structural organization similar to the previously described cell-wall-anchored proteins from S. aureus and other Gram-positive cocci. However, not all of the Ses proteins are direct homologues of the S. aureus proteins. Secondary and tertiary structure predictions suggest that most of the Ses proteins are composed of several contiguous subdomains, and that the majority of these predicted subdomains are folded into β-rich structures. PCR analysis indicates that certain genes may be found more frequently in disease isolates compared to strains isolated from healthy skin. Patients recovering from S. epidermidis infections had higher antibody titres against some Ses proteins, implying that these proteins are expressed during human infection. Western blot analyses of early-logarithmic and late-stationary in vitro cultures suggest that different regulatory mechanisms control the expression of the Ses proteins.

The phosphofructokinase from the non-conventional yeast Yarrowia lipolytica (YlPfk) was purified to homogeneity, and its encoding gene isolated. YlPfk is an octamer of 869 kDa composed of a single type of subunit, and shows atypical kinetic characteristics. It did not exhibit cooperative kinetics for fructose 6-phosphate (Hill coefficient, h 1·1; S 0·5 52 μM), it was inhibited moderately by MgATP (K i 3·5 mM), and it was strongly inhibited by phosphoenolpyruvate (K i 61 μM). Fructose 2,6-bisphosphate did not activate the enzyme, and AMP and ADP were also without effect. The gene YlPFK1 has no introns, and encodes a putative protein of 953 aa, with a molecular mass consistent with the subunit size found after purification. Disruption of the gene abolished growth in glucose and Pfk activity, while reintroduction of the gene restored both properties. This indicates that Y. lipolytica has only one gene encoding Pfk, and supports the finding that the enzyme consists of identical subunits. Glucose did not interfere with growth of the Ylpfk1 disruptant in permissive carbon sources. The unusual kinetic characteristics of YlPfk, and the intracellular concentrations of glycolytic intermediates during growth in glucose, suggest that YlPfk may play an important role in the regulation of glucose metabolism in Y. lipolytica, different from the role played by the enzyme in Saccharomyces cerevisiae.

The 650 kDa large toxin complex (L-TC) produced by Clostridium botulinum serotype D strain 4947 (D-4947) has a subunit structure composed of unnicked components, i.e. neurotoxin (NT), non-toxic non-haemagglutinin (NTNHA) and three haemagglutinin subcomponents (HA-70, HA-33 and HA-17). In this study, subunit interactions were investigated through the susceptibilities of the toxin components to limited trypsin proteolysis. Additionally, complex forms were reconstituted in vitro by various combinations of individual components. Trypsin treatment of intact D-4947 L-TC led to the formation of mature L-TC with nicks at specific sites of each component, which is usually observed in other strains of serotype D. NT, NTNHA and HA-17 were cleaved at their specific sites in either the single or complex forms, but HA-33 showed no sign of proteolysis. Unlike the other components, HA-70 was digested into random fragments as a single form, but it was cleaved into two fragments in the complex form. Based on the relative position of exposed or hidden regions of the individual components in the complex derived from their tryptic susceptibilities, an assembly model is proposed for the arrangement of individual subunits in the botulinum L-TC.

Bacillus anthracis, the aetiological agent of anthrax, synthesizes two surface-layer (S-layer) proteins. S-layers are two-dimensional crystalline arrays that completely cover bacteria. In rich medium, the B. anthracis S-layer consists of Sap during the exponential growth phase. Sap is a modular protein composed of an SLH (S-layer homology)-anchoring domain followed by a putative crystallization domain (Sapc). A projection map of the two-dimensional Sap array has been established on deflated bacteria. In this work, the authors used two approaches to investigate whether Sapc is the crystallization domain. The purified Sapc polypeptide (604 aa) was sufficient to form a crystalline structure, as illustrated by electron microscopy. Consistent with this result, the entire Sapc domain promoted auto-interaction in a bacterial two-hybrid screen developed for the present study. The screen was derived from a system that takes advantage of the Bordetella pertussis cyclase subdomain structure to enable one to identify peptides that interact. A screening strategy was then employed to study Sapc subdomains that mediate interaction. A random library, derived from the Sapc domain, was constructed and screened. The selected polypeptides interacting with the complete Sapc were all larger (155 aa and above) than the mean size of the randomly cloned peptides (approx. 60 residues). This result suggests that, in contrast with observations for other interactions studied with this two-hybrid system, large fragments were required to ensure efficient interaction. It was noteworthy that only one polypeptide, which spanned aa 148–358, was able to interact with less than the complete Sapc, in fact, with itself.

Clostridium cellulovorans produces a cellulase enzyme complex called the cellulosome. When cells were grown on different carbon substrates such as Avicel, pectin, xylan, or a mixture of all three, the subunit composition of the cellulosomal subpopulations and their enzymic activities varied significantly. Fractionation of the cellulosomes (7–11 fractions) indicated that the cellulosome population was heterogeneous, although the composition of the scaffolding protein CbpA, endoglucanase EngE and cellobiohydrolase ExgS was relatively constant. One of the cellulosomal fractions with the greatest endoglucanase activity also showed the highest or second highest cellulase activity under all growth conditions tested. The cellulosomal fractions produced from cells grown on a mixture of carbon substrates showed the greatest cellulase activity and contained CbpA, EngE/EngK, ExgS/EngH and EngL. High xylanase activity in cellulose, pectin and mixed carbon-grown cells was detected with a specific cellulosomal fraction which had relatively larger amounts of XynB, XynA and unknown proteins (35–45 kDa). These results in toto indicate that the assembly of cellulosomes occurs in a non-random fashion.

A putative dimethylallyltryptophan synthase gene, fgaPT2, was identified in the genome sequence of Aspergillus fumigatus. fgaPT2 was cloned and overexpressed in Saccharomyces cerevisiae. The protein FgaPT2 was purified to near homogeneity and characterized biochemically. This enzyme was found to convert l-tryptophan to 4-dimethylallyltryptophan, a reaction known to be the first step in ergot alkaloid biosynthesis. FgaPT2 is a soluble, dimeric protein with a subunit size of 52 kDa, and contains no putative prenyl diphosphate binding site (N/D)DXXD. K m values for l-tryptophan and dimethylallyl diphosphate (DMAPP) were determined as 8 and 4 μM, respectively. Metal ions, such as Mg2+ and Ca2+, enhance the reaction velocity, but are not essential for the enzymic reaction. FgaPT2 showed a relatively strict substrate specificity for both tryptophan and DMAPP. FgaPT2 is the first enzyme in the biosynthesis of ergot alkaloids to be purified and characterized in homogeneous form after heterologous overproduction.

Daptomycin is a 13 amino acid, cyclic lipopeptide produced by a non-ribosomal peptide synthetase (NRPS) mechanism in Streptomyces roseosporus. A 128 kb region of S. roseosporus DNA was cloned and verified by heterologous expression in Streptomyces lividans to contain the daptomycin biosynthetic gene cluster (dpt). The cloned region was completely sequenced and three genes (dptA, dptBC, dptD) encoding the three subunits of an NRPS were identified. The catalytic domains in the subunits, predicted to couple five, six or two amino acids, respectively, included a novel activation domain and amino-acid-binding pocket for incorporating the unusual amino acid l-kynurenine (Kyn), three types of condensation domains and an extra epimerase domain (E-domain) in the second module. Novel genes (dptE, dptF) whose products likely work in conjunction with a unique condensation domain to acylate the first amino acid, as well as other genes (dptI, dptJ) probably involved in supply of the non-proteinogenic amino acids l-3-methylglutamic acid and Kyn, were located next to the NRPS genes. The unexpected E-domain suggested that daptomycin would have d-Asn, rather than l-Asn, as originally assigned, and this was confirmed by comparing stereospecific synthetic peptides and the natural product both chemically and microbiologically.

Microcystins (MCs) are toxic heptapeptides which are synthesized by the filamentous cyanobacterium Planktothrix and other genera via non-ribosomal peptide synthesis. MCs share the common structure cyclo(-d-ala1-l-x2-d-erythro-β-iso-aspartic acid3-l-z4-adda5-d-Glu6-N-methyl-dehydroalanine7) [Adda; (2S, 3S, 8S, 9S)-3-amino-9-methoxy-2,6,8-trimethyl-10-phenyldeca-4,6-dienoic acid], in which numerous MC variants have been reported. In general, the variation in structure is due to different amino acid residues in positions 7, 2 and 4 within the MC molecule, which are thought to be activated by the adenylation domains mcyAAd1, mcyBAd1 and mcyCAd, respectively. It was the aim of the study (i) to identify MC ecotypes that differed in the production of specific MC variants and (ii) to correlate the genetic variation within adenylation domains with the observed MC variants among 17 Planktothrix strains. Comparison of the sequences of mcyAAd1 revealed two distinctive Ad-genotypes differing in base pair composition and the insertion of an N-methyl transferase (NMT) domain. The mcyAAd1 genotype with NMT (2854 bp) correlated with N-methyl-dehydroalanine and the mcyAAd1 genotype without NMT (1692 bp) correlated with dehydrobutyrine in position 7. Within mcyBAd1, a lower genetic variation (0–4 %) and an exclusive correlation between one Ad-genotype and homotyrosine as well as another Ad-genotype and arginine in position 2 was found. The sequences of mcyCAd were found to be highly similar (0–1 % dissimilarity) and all strains contained arginine in position 4. The results on adenylation domain polymorphism do provide insights into the evolutionary origin of adenylation domains in Planktothrix and may be combined with ecological research in order to provide clues about the abundance of genetically defined MC ecotypes in nature.

Identification of microbial community members in complex environmental samples is time consuming and repetitive. Here, ribosomal sequences and hidden Markov models are used in a novel approach to rapidly assign fungi to their presumptive genera. The ITS1 and ITS2 fragments from a range of axenic, anaerobic gut fungal cultures, including several type strains, were isolated and the RNA secondary structures predicted for these sequences were used to generate a fingerprinting program. The methodology was then tested and the algorithms improved using a collection of environmentally derived sequences, providing a rapid indicator of the fungal diversity and numbers of novel sequence groups within the environmental sample from which they were derived. While the methodology was developed to assist in investigations involving the rumen ecosystem, it has potential generic application in studying diversity and population dynamics in other microbial ecosystems.

There is growing evidence that Pseudomonas aeruginosa biofilms exhibit a multicellular developmental life cycle analogous to that of the myxobacteria. In non-mucoid PAO1 biofilms cultured in glass flow cells the phenotypic differentiation of microcolonies into a motile phenotype in the interior of the microcolony and a non-motile surrounding ‘wall phenotype’ are described. After differentiation the interior cells coordinately evacuated the microcolony from local break out points and spread over the wall of the flow cell, suggesting that the specialized microcolonies were analogous to crude fruiting bodies. A microcolony diameter of approximately 80 μm was required for differentiation, suggesting that regulation was related to cell density and mass transfer conditions. This phenomenon was termed ‘seeding dispersal’ to differentiate it from ‘erosion’ which is the passive removal of single cells by fluid shear. Using the flow cell culturing method, in which reproducible seeding phenotype in PAO1 wild-type was demonstrated, the effects of quorum sensing (QS) and rhamnolipid production (factors previously identified as important in determining biofilm structure) on seeding dispersal using knockout mutants isogenic with PAO1 was investigated. Rhamnolipid (rhlA) was not required for seeding dispersal but las/rhl QS (PAO1-JP2) was, in our system. To assess the clinical relevance of these data, mucoid P. aeruginosa cystic fibrosis isolate FRD1 was also investigated and was seeding-dispersal-negative.

Enterococcus faecalis divIVA (divIVA Ef) is an essential gene implicated in cell division and chromosome segregation. This gene was disrupted by insertional inactivation creating E. faecalis JHSR1, which was viable only when a wild-type copy of divIVA Ef was expressed in trans, confirming the essentiality of the gene. The absence of DivIVAEf in E. faecalis JHSR1 inhibited proper cell division, which resulted in abnormal cell clusters possessing enlarged cells of altered shape instead of the characteristic diplococcal morphology of enterococci. The lower viability of the divIVA Ef mutant is caused by improper nucleoid segregation and impaired septation within the numerous cells generated in each cluster. Overexpression of DivIVAEf in Escherichia coli KJB24 resulted in enlarged cells with disrupted cell division, suggesting that this round E. coli mutant strain could be used as an indicator for functionality of DivIVAEf. A Bacillus subtilis divIVA mutant was not complemented by DivIVAEf, indicating that this protein does not recognize DivIVA-specific target sites in B. subtilis, or that it does not interact with other proteins of the cell division machinery of this micro-organism. DivIVAEf also failed to complement a Streptococcus pneumoniae divIVA mutant, supporting the phylogenetic distance between Enterococcus and Streptococcus. Our results indicate that DivIVA is a species-specific multifunctional protein implicated in cell division and chromosome segregation in E. faecalis.