A novel phosphodiesterase from Aspergillus niger and its application to the study of membrane-derived oligosaccharides and other glycerol-containing biopolymers.

A novel phosphodiesterase has been found in commercially available extracts of Aspergillus niger and has been partially purified by fractionation with acetone and chromatography on carboxymethylcellulose. The enzyme attacks glycerophosphodiester bonds with the liberation of free glycerol only. The synthetic substrate glucose 6-phospho-sn-1'(3')-glycerol is hydrolyzed with production of equivalent amounts of free glycerol and glucose 6-phosphate. Similarly, the enzymic hydrolysis of sn-glycero-3-phosphocholine liberates glycerol and phosphocholine. The hydrophilic head groups of membrane phospholipids of Escherichia coli are continuously transferred to a closely related family of oligosaccharides ("membrane-derived oligosaccharides") containing glucose as the sole sugar (van Golde, L. M. G., Schulman, H., and Kennedy, E. P. (1973) Proc. Natl. Acad. Sci. U. S. A. 70, 1368--1372). Oligosaccharide A-2 contains sn-1-glycerophosphate residues (derived from phosphatidylglycerol) in phosphodiester linkage. Treatment of this oligosaccharide with the phosphodiesterase led to the liberation of nearly all of the glycerol as free glycerol. Subsequent partial acid hydrolysis of the enzyme-treated oligosaccharide led to the recovery of glucose 6-phosphate in almost quantitative yield. The sn-1-glycerophosphate residues are therefore linked to position 6 of glucose units of the oligosaccharide. The activity of the enzyme is not restricted to glycerophosphodiesterases since it will hydrolyze phosphodiesters containing other polyols such as the synthetically prepared glucose 6-phospho-DL-1'(2'-hydroxy-3'-ethoxy)propane.     

The biosynthesis of membrane-derived oligosaccharides. A membrane-bound phosphoglycerol transferase

Membrane-derived oligosaccharides, found in the Escherichia coli periplasmic space (Schulman, H., and Kennedy, E. P. (1979) J. Bacteriol. 137, 686-688), are composed of 8-10 units of glucose, the sole sugar, in beta 1 leads to 2 and beta 1 leads to 6 linkages (Schneider, J. E., Reinhold, V., Rumley, M. K., and Kennedy, E. P. (1979) J. Biol. Chem. 254, 10135-10138). Oligosaccharides in this family are variously substituted with succinyl ester residues, as well as with sn-1-phosphoglycerol and phosphoethanolamine, both derived from membrane phospholipids. These negatively charged oligosaccharides may function in cellular osmoregulation since their synthesis is under osmotic control (Kennedy, E. P. (1982) Proc. Natl. Acad. Sci. U.S.A. 79, 1092-1095). We now report initial characterization of an enzyme catalyzing transfer of phosphoglycerol residues from phosphatidylglycerol to membrane-derived oligosaccharides or to synthetic beta-glucoside acceptors. The products are sn-1,2-diglyceride and beta-glucoside-6-phosphoglycerol. Localized in the inner membrane, the transferase has a requirement for divalent cations, of which manganese is most effective, and a pH optimum of 8.9 in vitro.     

Cyclic glucans produced by Agrobacterium tumefaciens are substituted with sn-1-phosphoglycerol residues

In a previous study (Miller, K.J., Kennedy, E.P. and Reinhold, V.N. (1986) Science 231, 48-51) it was reported that the biosynthesis of periplasmic cyclic beta-1,2-glucans by Agrobacterium tumefaciens is strictly osmoregulated in a pattern closely similar to that found for the membrane-derived oligosaccharides of Escherichia coli (Kennedy, E.P. (1982) Proc. Natl. Acad. Sci. USA 79, 1092-1095). In addition to the well-characterized neutral cyclic glucan, the periplasmic glucans were found to contain an anionic component not previously reported. Biosynthesis of the anionic component is osmotically regulated in a manner indistinguishable from that of the neutral cyclic beta-1,2-glucan. We now find that the anionic component consists of cyclic beta-1,2-glucans substituted with one or more sn-1-phosphoglycerol residues. The presence of sn-1-phosphoglycerol residues represents an additional, striking similarity to the membrane-derived oligosaccharides of E. coli.     

Regulation of the synthesis of membrane-derived oligosaccharides in Escherichia coli. Assay of phosphoglycerol transferase I in vivo

Membrane-derived oligosaccharides are periplasmic constituents of Escherchia coli and other Gram-negative bacteria. Oligosaccharides in this family may be variously substituted with O-succinyl ester residues, and with sn-1-phosphoglycerol and phosphoethanolamine residues derived from membrane phospholipids. Membrane-derived oligosaccharides appear to be important in osmoregulation, because their synthesis is under strict control (Kennedy, E.P. (1982) Proc. Natl. Acad. Sci. U.S.A. 79, 1092-1095). Maximum rate of synthesis is at very low osmolarity of the medium. Phosphoglycerol residues are transferred from phosphatidylglycerol to membrane-derived oligosaccharides, or to certain beta-glucoside acceptors, in a reaction catalyzed by phosphoglycerol transferase I, an enzyme of the inner membrane (Jackson, B. J., and Kennedy, E.P. (1983) J. Biol. Chem. 258, 2394-2398). We now report that this enzyme catalyzes the transfer of phosphoglycerol residues to arbutin (p-hydroxyphenyl-beta-D-glucoside) added to the medium with Km similar to that observed with the cell-free enzyme. The active site of the enzyme must therefore be on the periplasmic face of the inner membrane. We assayed phosphoglycerol transferase I in vivo and found that it is present and completely active even in cells growing in medium of very high osmolarity, in which the synthesis of membrane-derived oligosaccharides is severely reduced. We conclude that osmotic regulation must occur at the stage of the synthesis of oligosaccharide chains. A study of the kinetics of transfer of phosphoglycerol residues to membrane-derived oligosaccharides in vivo revealed that synthesis of the polyglucose chains must stop abruptly upon transfer of cells from medium of low to high osmolarity, inconsistent with a model postulating simple dilution of some rate-limiting enzyme during growth at the higher osmolarity.     

Appearance of monoglyceride and triglyceride in the cell envelope of Escherichia coli mutants defective in diglyceride kinase

Diglyceride kinase mutants of Escherichia coli contain about 50- to 100-fold more 1,2-diglyceride than wild type cells. We now report that monoglyceride and triglyceride also accumulate in these strains. In mutant RZ60 (dgk-6) these compounds represent about 1 and 0.2%, respectively, of the total lipid fraction, while diglyceride represents 5-8% under most conditions. Monoglyceride accumulates predominantly in the outer membrane, while triglyceride builds up together with diglyceride in the cytoplasmic membrane. Under typical growth conditions about two-thirds of the diglyceride in E. coli arises in conjunction with synthesis of the membrane-derived oligosaccharides (Raetz, C.R.H., and Newman, K.F. (1979) J. Bacteriol. 137, 860-868). Inhibition of membrane-derived oligosaccharides (MDO) synthesis also curtails the accumulation of monoglyceride and triglyceride. However, there appears to be at least one other MDO-independent source of diglyceride and related metabolites. Since MDO synthesis is suppressed by high osmolarity (Kennedy, E.P. (1982) Proc. Natl. Acad. Sci. U.S. A. 79, 1092-1095), we have examined the effects of osmolarity on diglyceride accumulation in RZ60 (dgk-6). As expected, if MDO synthesis and diglyceride formation are coupled, the diglyceride level in RZ60 is higher at low osmolarity, while at high osmolarity the level of diglyceride is reduced to that observed in double mutants defective both in MDO synthesis and diglyceride kinase. Since dgk mutants do not grow at very low osmolarity, we have isolated several spontaneous phenotypic revertants that do. One class regains diglyceride kinase and has low diglyceride levels under all conditions. The other class remains defective in diglyceride kinase but tolerates higher diglyceride levels which amount to 13% of the total lipid during maximal induction of MDO synthesis at low osmolarity.     

Identification of UDP-glucose as an intermediate in the biosynthesis of the membrane-derived oligosaccharides of Escherichia coli.

The membrane-derived oligosaccharides of Escherichia coli constitute a closely related family of oligosaccharides containing approximately 9 glucose units variously substituted with sn-glycero-1-phosphate and phosphoethanolamine residues derived from the head groups of membrane phospholipids, and also with succinate in O-ester linkage (Kennedy, E.P., Rumley, M.K., Schulman, H., and van Golder, L.M.G. (1976) J. Biol. Chem. 251, 4208-4213). Studies with mutant strains defective in the synthesis of various nucleoside diphosphate sugars have now revealed that UDP-glucose is an essential intermediate in the biosynthesis of these oligosaccharides. Mutants unable to synthesize UDP-glucose do not contain significant amounts of the membrane-derived oligosaccharides. In contrast, a strain unable to synthesize ADP-glucose, the glucosyl donor for glycogen synthesis in E. coli, contained normal amounts of the membrane-derived oligosaccharides, although with a somewhat different pattern of distribution of the various subspecies. In confirmation of these genetic studies, pulse-label isotope tracer studies have been carried out with glucose of high specific activity, under conditions in which UDP-glucose comprises a large fraction of the total radioactivity in the low molecular weight pool. Subsequent "chase" experiments clearly revealed the conversion of UDP-glucose to the higher molecular weight membrane-derived oligosaccharides.     

Osmotic regulation of biosynthesis of membrane-derived oligosaccharides in Escherichia coli.

The osmotic regulation of the biosynthesis of membrane-derived oligosaccharides (MDO) in strains UB1005 and DC2 of Escherichia coli K-12 was examined; this regulation was previously reported by Clark (J. Bacteriol. 161:1049-1053, 1985) to be different from that observed by Kennedy for other strains of E. coli (Proc. Natl. Acad. Sci. USA 79:1092-1095, 1982). Osmotic regulation of the synthesis of MDO in UB1005 and DC2 is in fact indistinguishable from that previously reported for other strains of E. coli, with maximum production of MDO occurring in the medium of lowest osmolarity. The report of Clark to the contrary was apparently based on the inadequate methods for the measurement of MDO employed in that study. MDO are localized in the periplasm of wild-type E. coli cells. However, strain DC2, selected for hypersensitivity to a range of antibiotics, released most of its MDO into the medium, apparently as a result of greater outer membrane permeability.     

Chromosomal amplification of the Escherichia coli lipB region confers high-level resistance to selenolipoic acid

One of the mutants (slr7 mutant) of a wild-type Escherichia coli strain resistant to selenolipoic acid reported previously (K. E. Reed, T. W. Morris, and J. E. Cronan, Jr., Proc. Natl. Acad. Sci. USA 91:3720-3724, 1994) unexpectedly grew on minimal medium following transductional introduction of a lipA null mutation. We report that the slr7 strain carries a duplication of the lip chromosomal region that causes the phenotype of the mutant strain.     

Biosynthesis of membrane-derived oligosaccharides. Membrane-bound glucosyltransferase system from Escherichia coli requires polyprenyl phosphate.

The periplasmic glucans of Gram-negative bacteria, including the membrane-derived oligosaccharides (MDO) of Escherichia coli and the cyclic glucans of the Rhizobiaceae, are now recognized to be a family of closely related substances with important functions in osmotic adaptation and cell signaling. The synthesis of the beta-1,2-glucan backbone of MDO is catalyzed by a membrane-bound glucosyltransferase system previously shown to require UDP-glucose and (surprisingly) acyl carrier protein (Therisod, H., Weissborn, A. C., and Kennedy, E. P. (1986) Proc. Natl. Acad. Sci. U. S. A. 83, 7236-7240). In the present study, no glucan intermediates bound to acyl carrier protein or to UDP could detected. The enzyme system, however, was found to be strongly inhibited by bacitracin and by amphomycin. Because the two antibiotics function by forming specific complexes with polyprenyl phosphates, their inhibitory effect suggests a prenol requirement for MDO biosynthesis. Furthermore, the activity of the glucosyltransferase was greatly stimulated by the addition of polyprenyl phosphates such as decaprenyl-P and dihydroheptaprenyl-P, but not by farnesyl-P. The same membrane preparations carry out the synthesis of polyprenyl-P-glucose, which is also stimulated by added polyprenyl-P, including farnesyl-P, the most active of those tested. Pulse chase experiments, however, indicate that the endogenous pool of polyprenyl-P-glucose cannot be an obligate intermediate in the MDO glucosyltransferase system.     

RGS4 binds to membranes through an amphipathic alpha -helix

RGS4, a mammalian GTPase-activating protein for G protein alpha subunits, requires its N-terminal 33 amino acids for plasma membrane localization and biological activity (Srinivasa, S. P., Bernstein, L. S., Blumer, K. J., and Linder, M. E. (1998) Proc. Natl. Acad. Sci. U. S. A. 95, 5584-5589). In this study, we tested the hypothesis that the N-terminal domain mediates membrane binding by forming an amphipathic alpha-helix. RGS4 bound to liposomes containing anionic phospholipids in a manner dependent on the first 33 amino acids. Circular dichroism spectroscopy of a peptide corresponding to amino acids 1-31 of RGS4 revealed that the peptide adopted an alpha-helical conformation in the presence of anionic phospholipids. Point mutations that either neutralized positive charges on the hydrophilic face or substituted polar residues on the hydrophobic face of the model helix disrupted plasma membrane targeting and biological activity of RGS4 expressed in yeast. Recombinant mutant proteins were active as GTPase-activating proteins in solution but exhibited diminished binding to anionic liposomes. Peptides corresponding to mutants with the most pronounced phenotypes were also defective in forming an alpha-helix as measured by circular dichroism spectroscopy. These results support a model for direct interaction of RGS4 with membranes through hydrophobic and electrostatic interactions of an N-terminal alpha-helix.     

The Concave Face of Decorin Mediates Reversible Dimerization and Collagen Binding

Decorin, the prototypical small leucine-rich proteoglycan, binds to collagen and thereby regulates collagen assembly into fibrils. The crystal structure of the decorin core protein revealed a tight dimer formed by the association of two monomers via their concave faces (Scott, P. G., McEwan, P. A., Dodd, C. M., Bergmann, E. M., Bishop, P. N., and Bella, J. (2004) Proc. Natl. Acad. Sci. U.S.A. 101, 15633–15638). Whether decorin binds collagen as a dimer has been controversial. Using analytical ultracentrifugation, we determined a dissociation constant of 1.37 ± 0.30 μm for the mouse decorin dimer. Dimerization could be abolished by engineering glycosylation sites into the dimer interface; other interface mutants remained dimeric. The monomeric mutants were as stable as wild-type decorin in thermal unfolding experiments. Mutations on the concave face of decorin abolished collagen binding regardless of whether the mutant proteins retained the ability to dimerize or not. We conclude that the concave face of decorin mediates collagen binding and that the dimer therefore must dissociate to bind collagen.     

Procedure for the identification of Escherichia coli mutants affected in components containing glycerol derived from phospholipid turnover: Isolation of mutants lacking glycerol in membrane-derived oligosaccharides (MDO)

Abstract A mutant screening procedure is described which allows the identification of mutants carrying lesions in lipoprotein, membrane-derived oligosac-charides (MDO), and other compounds of the E. coli cell envelope containing glycerol derived from phospholipid metabolism. Two mutants lacking glycerol in MDO and one mutant devoid of lipoprotein demonstrate the usefulness of the procedure.     

Differential contributions of NADPH-cytochrome P450 oxidoreductase FAD binding site residues to flavin binding and catalysis

Transfer of reducing equivalents from NADPH to the cytochromes P450 is mediated by NADPH-cytochrome P450 oxidoreductase, which contains stoichiometric amounts of tightly bound FMN and FAD. Hydrogen bonding and van der Waals interactions between FAD and amino acid residues in the FAD binding site of the reductase serve to regulate both flavin binding and reactivity. The precise orientation of key residues (Arg(454), Tyr(456), Cys(472), Gly(488), Thr(491), and Trp(677)) has been defined by x-ray crystallography (Wang, M., Roberts, D. L., Paschke, R., Shea, T. M., Masters, B. S., Kim, J.-J. P. (1997) Proc. Natl. Acad. Sci. U. S. A. 94, 8411-8416). The current study examines the relative contributions of these residues to FAD binding and catalysis by site-directed mutagenesis and kinetic analysis. Mutation of either Tyr(456), which makes van der Waals contact with the FAD isoalloxazine ring and also hydrogen-bonds to the ribityl 4'-hydroxyl, or Arg(454), which bonds to the FAD pyrophosphate, decreases the affinity for FAD 8000- and 25,000-fold, respectively, with corresponding decreases in cytochrome c reductase activity. In contrast, substitution of Thr(491), which also interacts with the pyrophosphate grouping, had a relatively modest effect on both FAD binding (100-fold decrease) and catalytic activity (2-fold decrease), while the G488L mutant exhibited, respectively, 800- and 50-fold decreases in FAD binding and catalytic activity. Enzymic activity of each of these mutants could be restored by addition of FAD. Kinetic properties and the FMN content of these mutants were not affected by these substitutions, with the exception of a 3-fold increase in Y456S K(m)(cyt )(c) and a 70% decrease in R454E FMN content, suggesting that the FMN- and FAD-binding domains are largely, but not completely, independent. Even though Trp(677) is stacked against the re-face of FAD, suggesting an important role in FAD binding, deletion of both Trp(677) and the carboxyl-terminal Ser(678) decreased catalytic activity 50-fold without affecting FAD content.     

Isolation and amino-terminal sequences of subunits from the photosynthetic reaction center of Rhodopseudomonas capsulata

The amino-terminal sequences have been determined by Edman degradation for the reaction center polypeptides from a carotenoidless mutant of Rhodopseudomonas capsulata. Individual polypeptides were isolated by preparative electrophoresis and electroelution. By comparison with the sequences deduced from the DNA (Youvan, D.C., Alberti, M., Begush, H., Bylina, E.J. and Hearst, J.E. (1984) Proc. Natl. Acad. Sci. USA 81, 189–192) we conclude that the M and L subunits are processed so as to remove the amino-terminal methionine, whereas the H subunit is not processed at the amino-terminus after translation. None of the subunits is synthesized with a significant amino-terminal extension peptide.     

Active-site mutants of beta-lactamase: use of an inactive double mutant to study requirements for catalysis

We have studied the catalytic activity and some other properties of mutants of Escherichia coli plasmid-encoded RTEM beta-lactamase (EC 3.5.2.6) with all combinations of serine and threonine residues at the active-site positions 70 and 71. (All natural beta-lactamases have conserved serine-70 and threonine-71.) From the inactive double mutant Ser-70----Thr, Thr-71----Ser [Dalbadie-McFarland, G., Cohen, L. W., Riggs, A. D., Morin, C., Itakura, K., & Richards, J. H. (1982) Proc. Natl. Acad. Sci. U.S.A. 79, 6409-6413], an active revertant, Thr-71----Ser (i.e., residue 70 in the double mutant had changed from threonine to the serine conserved at position 70 in the wild-type enzyme), was isolated by an approach that allows identification of active revertants in the absence of a background of wild-type enzyme. This mutant (Thr-71----Ser) has about 15% of the catalytic activity of wild-type beta-lactamase. The other possible mutant involving serine and threonine residues at positions 70 and 71 (Ser-70----Thr) shows no catalytic activity. The primary nucleophiles of a serine or a cysteine residue [Sigal, I. S., Harwood, B. G., & Arentzen, R. (1982) Proc. Natl. Acad. Sci. U.S.A. 79, 7157-7160] at position 70 thus seem essential for enzymatic activity. Compared to wild-type enzyme, all three mutants show significantly reduced resistance to proteolysis; for the active revertant (Thr-71----Ser), we have also observed reduced thermal stability and reduced resistance to denaturation by urea.     

Intracellular transport and sorting of mutant human proinsulins that fail to form hexamers

Human proinsulin and insulin oligomerize to form dimers and hexamers. It has been suggested that the ability of prohormones to self associate and form aggregates may be responsible for the sorting process at the trans-Golgi. To examine whether insulin oligomerization is required for proper sorting into regulated storage granules, we have constructed point mutations in human insulin B chain that have been previously shown to prevent formation of insulin hexamers (Brange, J., U. Ribel, J. F. Hansen, G. Dodson, M. T. Hansen, S. Havelund, S. G. Melberg, F. Norris, K. Norris, L. Snel, A. R. Sorensen, and H. O. Voight. 1988. Nature [Lond.]. 333:679-682). One mutant (B10His----Asp) allows formation of dimers but not hexamers and the other (B9Ser----Asp) prevents formation of both dimers and hexamers. The mutants were transfected into the mouse pituitary AtT-20 cells, and their ability to be sorted into regulated secretory granules was compared to wild-type insulin. We found that while B10His----Asp is sorted somewhat less efficiently than wild-type insulin as reported previously (Carroll, R. J., R. E. Hammer, S. J. Chan, H. H. Swift, A. H. Rubenstein, and D. F. Steiner. 1988. Proc. Natl. Acad. Sci. USA. 85:8943-8947; Gross, D. J., P. A. Halban, C. R. Kahn, G. C. Weir, and L. Villa-Kumaroff. 1989. Proc. Natl. Acad. Sci. USA. 86:4107-4111). B9Ser----Asp is targeted to granules as efficiently as wild-type insulin. These results indicate that self association of proinsulin into hexamers is not required for its targeting to the regulated secretory pathway.     

Epstein-Barr virus DNA XII. A variable region of the Epstein-Barr virus genome is included in the P3HR-1 deletion.

The P3HR-1 subclone of Jijoye differs from Jijoye and from other Epstein-Barr virus (EBV)-infected cell lines in that the virus produced by P3HR-1 cultures lacks the ability to growth-transform normal B lymphocytes (Heston et al., Nature (London) 295:160-163, 1982; Miller et al., J. Virol. 18:1071-1080, 1976; Miller et al., Proc. Natl. Acad. Sci. U.S.A. 71:4006-4010, 1974; Ragona et al., Virology 101:553-557, 1980). The P3HR-1 virus was known to be deleted for a region which encodes RNA in latently infected, growth-transformed cells (Bornkamm et al., J. Virol. 35:603-618, 1980; Heller et al., J. Virol. 38:632-648, 1981; King et al., J. Virol. 36:506-518, 1980; Raab-Traub et al., J. Virol. 27:388-398, 1978; van Santen et al., Proc. Natl. Acad. Sci. U.S.A. 78:1930-1934, 1980). This deletion is now more precisely defined. The P3HR-1 genome contains less than 170 base pairs (and possibly none) of the 3,300-base pair U2 region of EBV DNA and is also lacking IR2 (a 123-base pair repeat which is the right boundary of U2). A surprising finding is that EBV isolates vary in part of the U2 region. Two transforming EB viruses, AG876 and Jijoye, are deleted for part of the U2 region including most or all of a fragment, HinfI-c, which encodes part of one of the three more abundant cytoplasmic polyadenylated RNAs of growth-transformed cells (King et al., J. Virol. 36:506-518, 1980; King et al., J. Virol. 38:649-660, 1981; van Santen et al., Proc. Natl. Acad. Sci. U.S.A. 78:1930-1934).     

Intramolecular cross-linking of myosin subfragment 1 with bimane

We previously showed that the fluorescent inter-thiol cross-linker dibromobimane (DBB) [Kosower, N. S., Kosower, E. M., Newton, G. L., & Ranney, H. M. (1979) Proc. Natl. Acad. Sci. U.S.A. 76, 3382-3386] cross-links two [50 and 20 kilodaltons (kDa)] of the three major fragments of myosin subfragment 1 (S-1); on intact S-1, DBB quenches tryptophans and inhibits all ATPases [Mornet, D., Ue, K., & Morales, M. F. (1985) Proc. Natl. Acad. Sci. U.S.A. 82, 1658-1662]. Here we characterize the modification chemically: DBB cross-links Cys-522 (50 kDa) with Cys-707 (20 kDa), thereby sealing a large preexisting heavy-chain loop containing important functionalities. Cross-linking rate is insensitive to nucleotides, but apparently sterically, either monobromobimane or DBB reduces Ca2+-ATPase to low, nonzero levels.     

TraJ protein of plasmid RP4 binds to a 19-base pair invert sequence repetition within the transfer origin

Transfer of plasmid RP4 during bacterial conjugation requires the plasmid-encoded TraJ protein, which binds to the transfer origin (Fürste, J. P., Pansegrau, W., Ziegelin, G., Kr?ger, M., and Lanka, E. (1989) Proc. Natl. Acad. Sci. U.S.A. 86, 1771-1775). As indicated by traJ mutants, the TraJ protein is a constituent of the relaxosome, the initiation complex of transfer DNA replication. The traJ gene maps adjacent to the transfer origin (oriT). The structural gene consists of a 372-base pair sequence encoding a polypeptide of 122 amino acids (13,282 Da). TraJ was purified from an Escherichia coli strain overproducing the protein. DNA footprinting experiments involving DNase I demonstrated that the purified protein binds to the right arm of a 19-base pair inverted repeat within oriT. Hydroxyl radical footprints of the DNA-protein complex revealed that TraJ protein is bound to only one side of the DNA helix.