Hydrodynamic and thermodynamic analysis of PEGylated human serum albumin.

Covalent labeling of therapeutic drugs and proteins with polyethylene glycol (PEGylation) is an important modification for improving stability, solubility, and half-life. PEGylation alters protein solution behavior through its impact on thermodynamic nonideality by increasing the excluded volume, and on hydrodynamic nonideality by increasing the frictional drag. To understand PEGylation's impact, we investigated the thermodynamic and hydrodynamic properties of a model system consisting of PEGylated human serum albumin derivatives using analytical ultracentrifugation (AUC) and dynamic light scattering (DLS). We constructed PEGylated human serum albumin derivatives of single, linear 5K, 10K, 20K, and 40K PEG chains and a single branched-chain PEG of 40K (2 × 20K). Sedimentation velocity (SV) experiments were analyzed using SEDANAL direct boundary fitting to extract ideal sedimentation coefficients so, hydrodynamic nonideality ks, and thermodynamic nonideality 2BM1SV terms. These quantities allow the determination of the Stokes radius Rs, the frictional ratio f/fo, and the swollen or entrained volume Vs/v, which measure size, shape, and solvent interaction. We performed sedimentation equilibrium experiments to obtain independent measurements of thermodynamic nonideality 2BM1SE. From DLS measurements, we determined the interaction parameter, kD, the concentration dependence of the apparent diffusion coefficient, D, and from extrapolation of D to c = 0 a second estimate of Rs. Rs values derived from SV and DLS measurements and ensemble model calculations (see complementary study) are then used to show that ks + kD = theoretical 2B22M1. In contrast, experimental BM1 values from SV and sedimentation equilibrium data collectively allow for similar analysis for protein-PEG conjugates and show that ks + kD = 1.02-1.07∗BM1, rather than the widely used ks + kD = 2BM1 developed for hard spheres. The random coil behavior of PEG dominates the colloidal properties of PEG-protein conjugates and exceeds the sum of a random coil and hard-sphere volume due to excess entrained water.

Evolution of Rev7 interactions in eukaryotic TLS DNA polymerase Polζ.

Translesion synthesis (TLS) DNA polymerase Polζ is crucial for the bypass replication over sites of DNA damage. The Rev7 subunit of Polζ is a HORMA (Hop1, Rev7, Mad2) protein that facilitates recruitment of Polζ to the replication fork via interactions with the catalytic subunit Rev3 and the translesion synthesis scaffold protein Rev1. Human Rev7 (hRev7) interacts with two Rev7-binding motifs (RBMs) of hRev3 by a mechanism conserved among HORMA proteins whereby the safety-belt loop of hRev7 closes on the top of the ligand. The two copies of hRev7 tethered by the two hRev3-RBMs form a symmetric head-to-head dimer through the canonical HORMA dimerization interface. Recent cryo-EM structures reveal that Saccharomyces cerevisiae Polζ (scPolζ) also includes two copies of scRev7 bound to distinct regions of scRev3. Surprisingly, the HORMA dimerization interface is not conserved in scRev7, with the two scRev7 protomers forming an asymmetric head-to-tail dimer with a much smaller interface than the hRev7 dimer. Here, we validated the two adjacent RBM motifs in scRev3, which bind scRev7 with affinities that differ by two orders of magnitude and confirmed the 2:1 stoichiometry of the scRev7:Rev3 complex in solution. However, our biophysical studies reveal that scRev7 does not form dimers in solution either on its own accord or when tethered by the two RBMs in scRev3. These findings imply that the scRev7 dimer observed in the cryo-EM structures is induced by scRev7 interactions with other Polζ subunits and that Rev7 homodimerization via the HORMA interface is a mechanism that emerged later in evolution.

Design of High-Affinity Metal-Controlled Protein Dimers.

The ability to precisely control protein complex formation has high utility in the expanding field of biomaterials. Driving protein-protein binding through metal-ligand bridging interactions is a promising method of achieving this goal. Furthermore, the capacity to precisely regulate both complex formation and dissociation enables additional control not available with constitutive protein complexes. Here we describe the design of three metal-controlled protein dimers that are completely monomeric in the absence of metal yet form high-affinity symmetric homodimers in the presence of zinc sulfate. The scaffold used for the designed dimers is the ß1 domain of streptococcal protein G. In addition to forming high-affinity dimers in the presence of metal, the complexes also dissociate upon addition of EDTA. Biophysical characterization revealed that the proteins maintain relatively high thermal stability, bind with high affinity, and are completely monodisperse in the monomeric and dimeric states. High-resolution crystal structures revealed that the dimers adopt the target structure and that the designed metal-binding histidine residues successfully bind zinc and function to drive dimer formation.

Structural Basis of Protein Kinase R Autophosphorylation.

The RNA-activated protein kinase, PKR, is a key mediator of the innate immunity response to viral infection. Viral double-stranded RNAs induce PKR dimerization and autophosphorylation. The PKR kinase domain forms a back-to-back dimer. However, intermolecular ( trans) autophosphorylation is not feasible in this arrangement. We have obtained PKR kinase structures that resolves this dilemma. The kinase protomers interact via the known back-to-back interface as well as a front-to-front interface that is formed by exchange of activation segments. Mutational analysis of the front-to-front interface support a functional role in PKR activation. Molecular dynamics simulations reveal that the activation segment is highly dynamic in the front-to-front dimer and can adopt conformations conducive to phosphoryl transfer. We propose a mechanism where back-to-back dimerization induces a conformational change that activates PKR to phosphorylate a "substrate" kinase docked in a front-to-front geometry. This mechanism may be relevant to related kinases that phosphorylate the eukaryotic initiation factor eIF2α.

Structural and Functional Studies of Bacterial Enolase, a Potential Target against Gram-Negative Pathogens.

Enolase is a glycolytic metalloenzyme involved in carbon metabolism. The advantage of targeting enolase lies in its essentiality in many biological processes such as cell wall formation and RNA turnover and as a plasminogen receptor. We initially used a DARTS assay to identify enolase as a target in Escherichia coli. The antibacterial activities of α-, ß-, and γ-substituted seven-member ring tropolones were first evaluated against four strains representing a range of Gram-negative bacteria. We observed that the chemical properties and position of the substituents on the tropolone ring play an important role in the biological activity of the investigated compounds. Both α- and ß-substituted phenyl derivatives of tropolone were the most active with minimum inhibitory concentrations in the range of 11-14 µg/mL. The potential inhibitory activity of the synthetic tropolones was further evaluated using an enolase inhibition assay, X-ray crystallography, and molecular docking simulations. The catalytic activity of enolase was effectively inhibited by both the naturally occurring ß-thujaplicin and the α- and ß-substituted phenyl derivatives of tropolones with IC50 values in range of 8-11 µM. Ligand binding parameters were assessed by isothermal titration calorimetry and differential scanning calorimetry techniques and agreed with the in vitro data. Our studies validate the antibacterial potential of tropolones with careful consideration of the position and character of chelating moieties for stronger interaction with metal ions and residues in the enolase active site.

A specific IFIH1 gain-of-function mutation causes Singleton-Merten syndrome.

Singleton-Merten syndrome (SMS) is an infrequently described autosomal-dominant disorder characterized by early and extreme aortic and valvular calcification, dental anomalies (early-onset periodontitis and root resorption), osteopenia, and acro-osteolysis. To determine the molecular etiology of this disease, we performed whole-exome sequencing and targeted Sanger sequencing. We identified a common missense mutation, c.2465G>A (p.Arg822Gln), in interferon induced with helicase C domain 1 (IFIH1, encoding melanoma differentiation-associated protein 5 [MDA5]) in four SMS subjects from two families and a simplex case. IFIH1 has been linked to a number of autoimmune disorders, including Aicardi-Goutières syndrome. Immunohistochemistry demonstrated the localization of MDA5 in all affected target tissues. In vitro functional analysis revealed that the IFIH1 c.2465G>A mutation enhanced MDA5 function in interferon beta induction. Interferon signature genes were upregulated in SMS individuals' blood and dental cells. Our data identify a gain-of-function IFIH1 mutation as causing SMS and leading to early arterial calcification and dental inflammation and resorption.

Conformational exchange at a C2H2 zinc-binding site facilitates redox sensing by the PML protein.

The promyelocytic leukemia protein, PML, plays a vital role in the cellular response to oxidative stress; however, the molecular mechanism of its action remains poorly understood. Here, we identify redox-sensitive sites of PML. A molecule of PML is cysteine-rich and contains three zinc-binding domains including RING, B-box1, and B-box2. Using in vitro assays, we have compared the sensitivity of the isolated RING and B-box1 domains and shown that B-box1 is more sensitive to oxidation. NMR studies of PML dynamics showed that one of the Zn-coordination sites within the B-box1 undergoes significant conformational exchange, revealing a hotspot for exposure of reactive cysteines. In agreement with the in vitro data, enhancement of the B-box1 Zn-coordination dynamics led to more efficient recruitment of PML into PML nuclear bodies in cells. Overall, our results suggest that the increased sensitivity of B-box1 to oxidative stress makes this domain an important redox-sensing component of PML.

Structural and functional analysis of a new subfamily of glycosyltransferases required for glycosylation of serine-rich streptococcal adhesins.

Serine-rich repeat glycoproteins (SRRPs) are a growing family of bacterial adhesins found in many streptococci and staphylococci; they play important roles in bacterial biofilm formation and pathogenesis. Glycosylation of this family of adhesins is essential for their biogenesis. A glucosyltransferase (Gtf3) catalyzes the second step of glycosylation of a SRRP (Fap1) from an oral streptococcus, Streptococcus parasanguinis. Although Gtf3 homologs are highly conserved in SRRP-containing streptococci, they share minimal homology with functionally known glycosyltransferases. We report here the 2.3 Å crystal structure of Gtf3. The structural analysis indicates that Gtf3 forms a tetramer and shares significant structural homology with glycosyltransferases from GT4, GT5, and GT20 subfamilies. Combining crystal structural analysis with site-directed mutagenesis and in vitro glycosyltransferase assays, we identified residues that are required for UDP- or UDP-glucose binding and for oligomerization of Gtf3 and determined their contribution to the enzymatic activity of Gtf3. Further in vivo studies revealed that the critical amino acid residues identified by the structural analysis are crucial for Fap1 glycosylation in S. parasanguinis in vivo. Moreover, Gtf3 homologs from other streptococci were able to rescue the gtf3 knock-out mutant of S. parasanguinis in vivo and catalyze the sugar transfer to the modified SRRP substrate in vitro, demonstrating the importance and conservation of the Gtf3 homologs in glycosylation of SRRPs. As the Gtf3 homologs only exist in SRRP-containing streptococci, we conclude that the Gtf3 homologs represent a unique subfamily of glycosyltransferases.

NADH/NAD+ binding and linked tetrameric assembly of the oncogenic transcription factors CtBP1 and CtBP2.

The activation of oncogenic C-terminal binding Protein (CtBP) transcriptional activity is coupled with NAD(H) binding and homo-oligomeric assembly, although the level of CtBP assembly and nucleotide binding affinity continues to be debated. Here, we apply biophysical techniques to address these fundamental issues for CtBP1 and CtBP2. Our ultracentrifugation results unambiguously demonstrate that CtBP assembles into tetramers in the presence of saturating NAD+ or NADH with tetramer to dimer dissociation constants about 100 nm. Isothermal titration calorimetry measurements of NAD(H) binding to CtBP show dissociation constants between 30 and 500 nm, depending on the nucleotide and paralog. Given cellular levels of NAD+ , CtBP is likely to be fully saturated with NAD under physiological concentrations suggesting that CtBP is unable to act as a sensor for NADH levels.

Structure-guided functional studies of plasmid-encoded dihydrofolate reductases reveal a common mechanism of trimethoprim resistance in Gram-negative pathogens.

Two plasmid-encoded dihydrofolate reductase (DHFR) isoforms, DfrA1 and DfrA5, that give rise to high levels of resistance in Gram-negative bacteria were structurally and biochemically characterized to reveal the mechanism of TMP resistance and to support phylogenic groupings for drug development against antibiotic resistant pathogens. Preliminary screening of novel antifolates revealed related chemotypes that showed high levels of inhibitory potency against Escherichia coli chromosomal DHFR (EcDHFR), DfrA1, and DfrA5. Kinetics and biophysical analysis, coupled with crystal structures of trimethoprim bound to EcDHFR, DfrA1 and DfrA5, and two propargyl-linked antifolates (PLA) complexed with EcDHFR, DfrA1 and DfrA5, were determined to define structural features of the substrate binding pocket and guide synthesis of pan-DHFR inhibitors.

Expression, purification, crystallization and preliminary X-ray analysis of ORF60, the small subunit (R2) of ribonucleotide reductase from Kaposi's sarcoma-associated herpesvirus (KSHV).

Ribonucleotide reductase (RNR) is responsible for converting ribonucleotides to deoxyribonucleotides, which are the building blocks of DNA. The enzyme is present in all life forms as well as in some large DNA viruses such as herpesviruses. The alpha-herpesviruses and gamma-herpesviruses encode two class Ia RNR subunits, R1 and R2, while the beta-herpesvirus subfamily only encode an inactive R1 subunit. Here, the crystallization of the R2 subunit of RNR encoded by the ORF60 gene from the oncovirus Kaposi's sarcoma-associated gamma-herpesvirus (KSHV) is reported. These are the first crystals of a viral R2 subunit; the use of in situ proteolysis with chymotrypsin and the addition of hexamine cobalt(III) chloride that were necessary to obtain crystals are described. Optimization of the crystallization conditions yielded crystals that diffracted to 2.0 A resolution. The crystals belonged to space group P2(1), with unit-cell parameters a = 63.9, b = 71.2, c = 71.8 A, alpha = 90, beta = 106.7, gamma = 90 degrees. The data set collected was 95.3% complete, with an R(merge) of 9.6%. There are two molecules in the asymmetric unit, corresponding to a solvent content of 43.4%.

The crystal structure of the protein YhaK from Escherichia coli reveals a new subclass of redox sensitive enterobacterial bicupins.

YhaK is a protein of unknown function found in low abundance in the cytosol of Escherichia coli. DNA array studies have revealed that YhaK is strongly up-regulated by nitroso-glutathione (GSNO) and also displays a 12-fold increase in expression during biofilm growth of E. coli 83972 and VR50 in human urine. We have determined the YhaK crystal structure and demonstrated that in vitro YhaK is a good marker for monitoring oxidative stresses in E. coli. The YhaK protein structure shows a bicupin fold where the two cupin domains are crosslinked with one intramolecular disulfide bond (Cys10 to Cys204). We found that the third cysteine in YhaK, Cys122, is oxidized to a sulfenic acid. Two chloride ions are found in the structure, one close to the reactive Cys122, and the other on a hydrophobic surface close to a symmetry-related molecule. There are major structural differences at the N-terminus of YhaK compared with similar structures that also display the bicupin fold (YhhW and hPirin). YhaK showed no quercetinase and peroxidase activity. However, reduced YhaK was very sensitive to reactive oxygen species (ROS). The complete, functional E. coli glutaredoxin or thioredoxin systems protected YhaK from oxidation. E. coli thioredoxin reductase and NADPH produced ROS and caused oxidation and oligomerization of reduced YhaK. Taken together, we propose that YhaK is the first of a new sub-class of bicupins that lack the canonical cupin metal-binding residues of pirins and may be involved in chloride binding and/or sensing of oxidative stress in enterobacteria.

HEPN: a common domain in bacterial drug resistance and human neurodegenerative proteins.

A novel domain - HEPN (higher eukarytoes and prokaryotes nucleotide-binding domain) - found in several bacterial species is also present in the human protein, sacsin, a chaperonin implicated in an early-onset neurodegenerative disease. The distant structural similarity suggests that this domain might be involved in nucleotide binding.

Drugging the Folate Pathway in Mycobacterium tuberculosis: The Role of Multi-targeting Agents.

The folate biosynthetic pathway offers many druggable targets that have yet to be exploited in tuberculosis therapy. Herein, we have identified a series of small molecules that interrupt Mycobacterium tuberculosis (Mtb) folate metabolism by dual targeting of dihydrofolate reductase (DHFR), a key enzyme in the folate pathway, and its functional analog, Rv2671. We have also compared the antifolate activity of these compounds with that of para-aminosalicylic acid (PAS). We found that the bioactive metabolite of PAS, in addition to previously reported activity against DHFR, inhibits flavin-dependent thymidylate synthase in Mtb, suggesting a multi-targeted mechanism of action for this drug. Finally, we have shown that antifolate treatment in Mtb decreases the production of mycolic acids, most likely due to perturbation of the activated methyl cycle. We conclude that multi-targeting of the folate pathway in Mtb is associated with highly potent anti-mycobacterial activity.

Functional and structural basis of E. coli enolase inhibition by SF2312: a mimic of the carbanion intermediate.

Many years ago, the natural secondary metabolite SF2312, produced by the actinomycete Micromonospora, was reported to display broad spectrum antibacterial properties against both Gram-positive and Gram-negative bacteria. Recent studies have revealed that SF2312, a natural phosphonic acid, functions as a potent inhibitor of human enolase. The mechanism of SF2312 inhibition of bacterial enolase and its role in bacterial growth and reproduction, however, have remained elusive. In this work, we detail a structural analysis of E. coli enolase bound to both SF2312 and its oxidized imide-form. Our studies support a model in which SF2312 acts as an analog of a high energy intermediate formed during the catalytic process. Biochemical, biophysical, computational and kinetic characterization of these compounds confirm that altering features characteristic of a putative carbanion (enolate) intermediate significantly reduces the potency of enzyme inhibition. When SF2312 is combined with fosfomycin in the presence of glucose-6 phosphate, significant synergy is observed. This suggests the two agents could be used as a potent combination, targeting distinct cellular mechanism for the treatment of bacterial infections. Together, our studies rationalize the structure-activity relationships for these phosphonates and validate enolase as a promising target for antibiotic discovery.

The structure of the RNA m5C methyltransferase YebU from Escherichia coli reveals a C-terminal RNA-recruiting PUA domain.

Nucleotide methylations are the most common type of rRNA modification in bacteria, and are introduced post-transcriptionally by a wide variety of site-specific enzymes. Three 5-methylcytidine (m(5)C) bases are found in the rRNAs of Escherichia coli and one of these, at nucleotide 1407 in 16 S rRNA, is the modification product of the methyltransferase (MTase) YebU (also called RsmF). YebU requires S-adenosyl-l-methionine (SAM) and methylates C1407 within assembled 30 S subunits, but not in naked 16 S rRNA or within tight-couple 70 S ribosomes. Here, we describe the three-dimensional structure of YebU determined by X-ray crystallography, and we present a molecular model for how YebU specifically recognizes, binds and methylates its ribosomal substrate. The YebU protein has an N-terminal SAM-binding catalytic domain with structural similarity to the equivalent domains in several other m(5)C RNA MTases including RsmB and PH1374. The C-terminal one-third of YebU contains a domain similar to that in pseudouridine synthases and archaeosine-specific transglycosylases (PUA-domain), which was not predicted by sequence alignments. Furthermore, YebU is predicted to contain extended regions of positive electrostatic potential that differ from other RNA-MTase structures, suggesting that YebU interacts with its RNA target in a different manner. Docking of YebU onto the 30 S subunit indicates that the PUA and MTase domains make several contacts with 16 S rRNA as well as with the ribosomal protein S12. The ribosomal protein interactions would explain why the assembled 30 S subunit, and not naked 16 S rRNA, is the preferred substrate for YebU.

Structural comparison of bacterial and human iron-dependent phenylalanine hydroxylases: similar fold, different stability and reaction rates.

Structure determination of bacterial homologues of human disease-related proteins provides an efficient path to understanding the three-dimensional fold of proteins that are associated with human diseases. However, the precise locations of active-site residues are often quite different between bacterial and human versions of an enzyme, creating significant differences in the biological understanding of enzyme homologs. To study this hypothesis, phenylalanine hydroxylase from a bacterial source has been structurally characterized at high resolution and comparison is made to the human analog. The enzyme phenylalanine hydroxylase (PheOH) catalyzes the hydroxylation of l-phenylalanine into l-tyrosine utilizing the cofactors (6R)-l-erythro-5,6,7,8 tetrahydrobiopterin (BH(4)) and molecular oxygen. Previously determined X-ray structures of human and rat PheOH, with a sequence identity of more than 93%, show that these two structures are practically identical. It is thus of interest to compare the structure of the divergent Chromobacterium violaceum phenylalanine hydroxylase (CvPheOH) ( approximately 24% sequence identity overall) to the related human and rat PheOH structures. We have determined crystal structures of CvPheOH to high resolution in the apo-form (no Fe-added), Fe(III)-bound form, and 7,8-dihydro-l-biopterin (7,8-BH(2)) plus Fe(III)-bound form. The bacterial enzyme displays higher activity and thermal melting temperature, and structurally, differences are observed in the N and C termini, and in a loop close to the active-site iron atom.

Mechanisms underlying responsiveness to tetrahydrobiopterin in mild phenylketonuria mutations.

A subtype of phenylalanine hydroxylase (PAH) deficiency that responds to cofactor (tetrahydrobiopterin, BH4) supplementation has been associated with phenylketonuria (PKU) mutations. The underlying molecular mechanism of this responsiveness is as yet unknown and requires a detailed in vitro expression analysis of the associated mutations. With this aim, we optimized the analysis of the kinetic and cofactor binding properties in recombinant human PAH and in seven mild PKU mutations, i.e., c.194T>C (p.I65T), c.204A>T (p.R68S), c.731C>T (p.P244L), c.782G>A (p.R261Q), c.926C>T (p.A309V), c.1162G>A (p.V388M), and c.1162G>A (p.Y414C) expressed in E. coli. For p.I65T, p.R68S, and p.R261Q, we could in addition study the equilibrium binding of BH4 to the tetrameric forms by isothermal titration calorimetry (ITC). All the mutations resulted in catalytic defects, and p.I65T, p.R68S, p.P244L, and most probably p.A309V, showed reduced binding affinity for BH4. The possible stabilizing effect of the cofactor was explored using a cell-free in vitro synthesis assay combined with pulse-chase methodology. BH4 prevents the degradation of the proteins of folding variants p.A309V, p.V388M, and p.Y414C, acting as a chemical chaperone. In addition, for wild-type PAH and all mild PKU mutants analyzed in this study, BH4 increases the PAH activity of the synthesized protein and protects from the rapid inactivation observed in vitro. Catalase and superoxide dismutase partially mimic this protection. All together, our results indicate that the response to BH4 substitution therapy by PKU mutations may have a multifactorial basis. Both effects of BH4 on PAH, i.e., the chemical chaperone effect preventing protein misfolding and the protection from inactivation, may be relevant mechanisms of the responsive phenotype.

PAHdb 2003: what a locus-specific knowledgebase can do.

PAHdb, a legacy of and resource in genetics, is a relational locus-specific database (http://www.pahdb.mcgill.ca). It records and annotates both pathogenic alleles (n = 439, putative disease-causing) and benign alleles (n = 41, putative untranslated polymorphisms) at the human phenylalanine hydroxylase locus (symbol PAH). Human alleles named by nucleotide number (systematic names) and their trivial names receive unique identifier numbers. The annotated gDNA sequence for PAH is typical for mammalian genes. An annotated gDNA sequence is numbered so that cDNA and gDNA sites are interconvertable. A site map for PAHdb leads to a large array of secondary data (attributes): source of the allele (submitter, publication, or population); polymorphic haplotype background; and effect of the allele as predicted by molecular modeling on the phenylalanine hydroxylase enzyme (EC 1.14.16.1) or by in vitro expression analysis. The majority (63%) of the putative pathogenic PAH alleles are point mutations causing missense in translation of which few have a primary effect on PAH enzyme kinetics. Most apparently have a secondary effect on its function through misfolding, aggregation, and intracellular degradation of the protein. Some point mutations create new splice sites. A subset of primary PAH mutations that are tetrahydrobiopterin-responsive is highlighted on a Curators' Page. A clinical module describes the corresponding human clinical disorders (hyperphenylalaninemia [HPA] and phenylketonuria [PKU]), their inheritance, and their treatment. PAHdb contains data on the mouse gene (Pah) and on four orthologous mutant mouse models and their use (for example, in research on oral treatment of PKU with the enzyme phenylalanine ammonia lyase [EC 4.3.1.5]).

The highly conserved domain of unknown function 1792 has a distinct glycosyltransferase fold.

More than 33,000 glycosyltransferases have been identified. Structural studies, however, have only revealed two distinct glycosyltransferase (GT) folds, GT-A and GT-B. Here we report a 1.34-Å resolution X-ray crystallographic structure of a previously uncharacterized 'domain of unknown function' 1792 (DUF1792) and show that the domain adopts a new fold and is required for glycosylation of a family of serine-rich repeat streptococcal adhesins. Biochemical studies reveal that the domain is a glucosyltransferase, and it catalyses the transfer of glucose to the branch point of the hexasaccharide O-linked to the serine-rich repeat of the bacterial adhesin, Fap1 of Streptococcus parasanguinis. DUF1792 homologues from both Gram-positive and Gram-negative bacteria also exhibit the activity. Thus, DUF1792 represents a new family of glycosyltransferases; therefore, we designate it as a GT-D glycosyltransferase fold. As the domain is highly conserved in bacteria and not found in eukaryotes, it can be explored as a new antibacterial target.