Biomedical subjects

B Kemper

Publications and source records attributed to B Kemper.

At least 73 records · Page 4Linked to original sources

PubMed1995

In vitro processing of heteroduplex loops and mismatches by endonuclease VII.

Endonuclease VII is a Holliday-structure resolving enzyme of phage T4 which cleaves at junctions of branched DNAs and at mispairings. In extension of these findings we report the following: i) Endonuclease VII can discriminate between a large heteroduplex loop and a TT mismatch arranged in tandem, 6 nt distant from each other, in the same heteroduplex molecule. The enzyme cleaves two nucleotides 3' from the base of the loop or the TT mismatch. ii) Similar to its reactions with mismatches cleavage of heteroduplex loops by endonucleave VII can also initiate correction of perfect double-strandedness by T4 DNA polymerase and T4 DNA-ligase in vitro. Loops of 8 nt and 20 nt were repaired efficiently. iii) For the first time endonuclease VII cleavage sites were also mapped in single-stranded DNA if it was part of the 20-nt loop. This suggests that looping of single-stranded DNA can induce formation of secondary structures, which are recognizable by endonuclease VII.

Bacteriophage T4↗

PubMed1995

Substitution at residue 473 confers progesterone 21-hydroxylase activity to cytochrome P450 2C2.

The carboxyl-terminal 28 amino acids of rabbit cytochrome P450 2C2 are markedly different from those of other rabbit cytochrome P450 2C family members and, substitution of the equivalent amino acids of other cytochrome P450s can confer novel steroid hydroxylase activity to P450 2C2 while the normal lauric acid hydroxylase activity is retained. To determine the basis for the novel steroid hydroxylase activity, amino acids of cytochrome P450 2C1 were substituted for those of cytochrome P450 2C2 and the mutants were expressed in COS-1 cells. There are 13 differences between the sequences of cytochrome P450 2C2 and P450 2C1 in this region, including five nonconservative exchanges of charged and uncharged amino acids. However, only substitution of valine for Ser-473 increased steroid hydroxylase activity to the maximum level expected in a modified cytochrome P450 2C2, which contained additional substitutions in the 368-388 region to maximize progesterone hydroxylase activity. Introduction of this single substitution into cytochrome P450 2C2 resulted in 21-progesterone hydroxylase activity similar to that resulting from substitution of all 28 carboxyl-terminal cytochrome P450 2C1 amino acids. None of the substitutions, with one exception, substantially affected either lauric acid hydroxylase activity or the amount of immunologically reactive cytochrome P450 that was expressed. A glycine substitution for Val-477 reduced activity of both lauric acid hydroxylase and progesterone hydroxylase and altered the regioselectivity of the hydroxylation for both. Homology modeling of cytochrome P450 2C2, based on the cytochrome bacterial P450cam sequence, indicated that the side chains of residue 473 and the other five residues previously shown to affect substrate specificity face the substrate pocket. For four of the six residues, smaller and more hydrophobic residues increased progesterone relative to lauric acid hydroxylation.

Amino Acid Sequence↗

PubMed1994

Inverse relationship of cotranslational translocation with the hydrophobic moment of the bovine preproparathyroid hormone signal sequence.

To investigate the dependence of transmembrane translocation on the hydrophobic moment of the hydrophobic core of the preproparathyroid hormone signal sequence, amino acids were switched to maximize or minimize the hydrophobic moment without changing the length or overall hydrophobicity of the core. As assayed in an in vitro translation system with microsomal membranes, the efficiency of translocation of these mutants was inversely related to the hydrophobic moment, indicating the hydrophobic moment or a related property may contribute to translocational activity.

Amino Acid Sequence↗

PubMed1994

Differential effects of mutations in substrate recognition site 1 of cytochrome P450 2C2 on lauric acid and progesterone hydroxylation.

Mutations at amino acid positions 107-120, which are part of a predicted substrate recognition site [Gotoh, O. (1992) J. Biol. Chem. 267, 83-90], were analyzed in C2MstC1, a chimera of P450 2C2 and P450 2C1. This hybrid protein has a new activity for progesterone C21-hydroxylation in addition to the lauric acid (omega-1)hydroxylase activity present in both parent proteins. Various substitutions for highly conserved glycines at positions 111 and 117 and tryptophan at position 120 strongly decreased the lauric acid hydroxylase activity of P450 2C2 and C2MstC1 and the progesterone hydroxylase activity of C2MstC1. Activities of mutant proteins with substitutions at 107, 108, and 112-115 were also strongly reduced. Modest or no decreases in activity were observed for substitutions at 109, 110, 116, 118, and 119. Lauric acid hydroxylase activity decreased more in most C2MstC1 mutants than in those of P450 2C2, particularly at positions 107 and 108. A substitution of phenylalanine for valine-112 reduced progesterone hydroxylation by 30-fold while only moderately reducing lauric acid hydroxylase by 40%. This differential effect on two dissimilar substrates demonstrates the importance of residue 112 for substrate interactions. The results are consistent with a model in which residues 107-110 align with the B'-helix of the bacterial proteins P450cam and P450BM-3. This helix is followed by a substrate-contacting loop from 111 to 116, and residues 117-120 align with the C-helices of the bacterial proteins In this alignment, Trp-120 is positioned behind the heme such that it could participate in electron transfer from reductase.(ABSTRACT TRUNCATED AT 250 WORDS)

Amino Acid Sequence↗

PubMed1994

Cleavage of double-crossover molecules by T4 endonuclease VII.

DNA double-crossover molecules containing two Holliday junctions have been prepared and treated with endonuclease VII, the resolvase from bacteriophage T4. One molecule contains antiparallel double-helical domains, and the other molecule contains parallel domains. The parallel double-crossover model system has been made tractable by closing the free ends of the molecule, to convert it to a catenane. The products resulting from the two substrates differ substantially. The molecule containing antiparallel helical domains is cleaved three nucleotides 3' to the crossover points, in a fashion similar to single Holliday junction analogs. The molecule containing parallel helical domains is cleaved, but the major points of scission are five nucleotides 5' to a branch point on the crossover strands and six nucleotides 3' to the same branch point on the non-crossover strands. The major sites of scission reflect features of molecular symmetry in each case, suggesting that the resolvase recognizes structural features. The cleavage results suggest that the antiparallel structure is the natural substrate, if the Holliday junction is unconstrained within the cell. It is straightforward to reconcile antiparallel Holliday junctions with the conventional parallel paradigm of recombination. Nevertheless, the cleavage of the parallel molecule shows that a parallel substrate could also be cleaved symmetrically by endonuclease VII (but with different products) if the molecule were constrained to assume that conformation within the cell.

Base Sequence↗

PubMed1994

A transcriptional regulatory element common to a large family of hepatic cytochrome P450 genes is a functional binding site of the orphan receptor HNF-4.

Hepatic cytochrome P450 (CYP) genes, including members of CYP1 to CYP4 families, comprise the majority of the CYP gene superfamily. Previous study has demonstrated that HepG2-specific transcriptional activation of two CYP2C genes was dependent on a common element that bound a HepG2 nuclear protein designated HPF-1 (Venepally, P., Chen, D., and Kemper, B. (1992) J. Biol. Chem. 267, 17333-17338). This cis-acting element is highly homologous to the hepatocyte nuclear factor 4 (HNF-4) binding motif and is present in the promoters of more than 20 other CYP2 genes. To investigate the relationship between HPF-1 and HNF-4, we have compared their tissue distribution, DNA binding, and immunochemical characteristics, as well as transcriptional activity of their recognition elements. DNase I footprint analyses and gel-shift assays indicated that HPF-1, like HNF-4, was present in liver and kidney, but not brain and spleen. Both factors bound to either the HPF-1 site in the CYP2C2 promoter or an HNF-4 site in the human apolipoprotein CIII promoter. These complexes could be "supershifted" by an antiserum specific for HNF-4. When the sequence of the HPF-1 site in the CYP2C2 promoter was changed to that of the apolipoprotein CIII HNF-4 site, comparable transcriptional activities were obtained with the wild-type promoter and the HNF-4 mutant in transfected HepG2 cells. Cotransfection of HNF-4 with CYP2C2 promoter-luciferase constructs in COS-1 cells indicated that HNF-4 could trans-activate the promoters containing the HPF-1 site. These results indicate that the HPF-1 motif is a functional HNF-4-binding site, and the common immunological properties indicate that HPF-1 and HNF-4 are closely related and possibly identical. HNF-4, therefore, may act as a common regulator for the liver-specific expression of many CYP2 genes.

Animals↗

PubMed1994

Endonuclease VII of phage T4 nicks N-2-acetylaminofluorene-induced DNA structures in vitro.

We have tested in vitro the activity of T4 endonuclease VII on three different double-stranded oligonucleotides bearing a single N-2-acetylaminofluorene (AAF) adduct covalently bound to each of the three guanine residues located within the NarI site (G1G2CG3CC), a strong frameshift mutation hot spot in E. coli. With the oligonucleotides modified at G2 and G3 a specific cleavage pattern with T4 endonuclease VII was observed in the complementary strand while no cleavage was found in the adduct-bearing strand. On the other hand, when G1 was modified, only a very faint cleavage band was observed (< 1%). These differences in nicking among the three AAF-modified DNA substrates are discussed in terms of the polymorphic nature in adduct-induced DNA structures as previously shown. This "non-physiological" activity of a DNA resolvase is discussed in terms of a potential role for such enzymes in the induction of frameshift mutations.

2-Acetylaminofluorene↗

PubMed1994

Differential protein binding and transcriptional activities of HNF-4 elements in three closely related CYP2C genes.

A functional binding site for a liver-enriched transcription factor, hepatocyte nuclear factor-4 (HNF-4), has previously been identified around -100 in the CYP2C2 promoter and proposed to be a common regulatory motif for the hepatic expression of many CYP2 genes. The transcriptional activity of the proximal promoters of three closely related cytochrome P450 genes (CYP2C1, CYP2C2, and CYP2C3) have been compared in HepG2 cells and correlated with the relative binding affinities of the HNF-4 motifs in the genes for proteins in liver, kidney, and HepG2 extracts that react with antisera to HNF-4. Gel-shift assays suggested that these highly similar motifs bound HNF-4 with significantly different affinities. The relative binding affinities to the protein in liver extracts were estimated by competitive gel-shift binding, and the binding affinity of CYP2C2 was two-fold and eight-fold greater than that of CYP2C1 and CYP2C3, respectively. These affinity differences correlated well with the transcriptional activities of either the minimal hepatic promoters (117 to 135 bp of 5'-flanking region) of the genes in HepG2 cells or minimal CYP2C2 promoters containing the HNF-4 motif from each gene. Identification of the CYP2C3 site was somewhat indirect; the HNF-4 element in the CYP2C3 promoter could be converted to a motif with binding affinity and hepatic cell-specific transcriptional activity similar to that of CYP2C2 by a single nucleotide substitution. The activities of the minimal promoters did not correlate well with hepatic expression of each gene in vivo, as estimated previously by the concentration of mRNA in livers of untreated animals.(ABSTRACT TRUNCATED AT 250 WORDS)

Animals↗

PubMed1993

Preference for aromatic substitutions at tryptophan-120, which is highly conserved and a potential mediator of electron transfer in cytochrome P450 2C2.

A proposal that tryptophan-120 of P450 2C2 might mediate electron transfer has been tested by substitution of a series of aliphatic and aromatic residues at this position. Activity in transfected COS1 cells expressing enzymes substituted with the aromatic residues, tyrosine, phenylalanine, and histidine, was 25% to 60% of wild type. Activities in enzymes containing serine, arginine, or aliphatic amino acid substitutions were less than 10%, except for alanine (25%). Alanine is the only naturally occurring substitution for tryptophan at this position in mammalian cytochromes P450 and is present in the electron pathway proposed on the basis of the crystal structure of cytochrome c and cytochrome c peroxidase. The preference for aromatic residues or the small aliphatic amino acid, alanine, at position 120 is consistent with a role for this tryptophan in electron transfer.

Amino Acid Sequence↗

PubMed1993

Hydrophobic side chain requirements for lauric acid and progesterone hydroxylation at amino acid 113 in cytochrome P450 2C2, a potential determinant of substrate specificity.

To determine the requirements for hydrophobic amino acids at position 113 in cytochrome P450 2C2, a series of hydrophobic and uncharged polar amino acids was substituted for isoleucine in P450 2C2 and in C2MstC1, a chimera of P450 2C2 and P450 2C1. Lauric acid hydroxylase activity was determined in COS1 cells transfected with P450 2C2 mutants and both lauric acid and progesterone hydroxylase activities were determined for C2MstC1 variants. In P450 2C2, 40 to 120% of the wild type (omega-1) lauric acid hydroxylase activity was retained in all hydrophobic mutants, but activity was reduced to near background by substitutions of the polar amino acids, tyrosine and cysteine. Likewise, in C2MstC1 mutants, hydrophobic substitutions were 20 to 50% as active as wild type for lauric acid hydroxylation, and polar amino acids again resulted in strong reductions of activity. In contrast, a different pattern of activity for progesterone C21-hydroxylase was observed for C2MstC1 mutants. A valine substitution had a modest effect on activity but substitutions of leucine and alanine reduced progesterone C21-hydroxylase activity 5- to 7-fold, respectively, and the large hydrophobic amino acid, phenylalanine, reduced activity about 30-fold. No changes in the regiospecificity of progesterone hydroxylation were observed for any of the mutants. Similar steady-state levels of immunoprecipitated, radiolabeled protein were observed for each mutant except for the glycine substitution which resulted in degradation of the protein. The different patterns of the effects of the mutations on progesterone and lauric acid hydroxylase activity provide additional support for the critical role of residue 113 in substrate recognition. The low activities in mutant proteins with hydrophilic amino acid substitutions indicate that the hydrophobic nature of this residue is important. The hydrophobic requirements are more stringent for a larger, more rigid steroid substrate than for a saturated fatty acid with a flexible hydrocarbon tail.

Amino Acid Sequence↗

PubMed1993

Cassette mutagenesis of a potential substrate recognition region of cytochrome P450 2C2.

Cassette mutagenesis was used to analyze the effects of mutations at amino acid positions 107-120 in cytochrome P450 2C2 which are part of a predicted substrate recognition site, SRS-1 (Gotoh, O. (1992) J. Biol. Chem. 267, 83-90), and terminate with a highly conserved tryptophan. All the mutant enzymes were expressed in COS1 cells at approximately the same level as wild type as determined by immunoprecipitation of radiolabeled products. Substitutions of amino acids from 107 to 110 and from 116 to 119 resulted in mutant enzymes that retained substantial lauric acid hydroxylase activity suggesting that these residues are not critical determinants of substrate specificity. In contrast, amino acids 112-115 were characterized by strong decreases in activity in at least one mutation tested. A substitution of phenylalanine for valine 112 reduced activity about 3-fold. At position 113, substitution of leucine for isoleucine had little effect, but activity was reduced 15-fold by substitution with cysteine. A conservative mutation at position 114 reduced activity 5-fold, a nonconservative mutation at position 115 resulted in a 30-fold reduction of lauric acid hydroxylation. Substitution of glutamic acid for glycine at position 109 had little effect on activity, while more conservative substitutions of valine for glycines at positions 111 and 117 resulted in about 50- and 7-fold reductions of activity, respectively. Substitution of leucine for tryptophan 120 resulted in a 7-fold reduction of activity indicating that tryptophan at this position contributes to, but is not essential for, enzymatic activity in spite of its presence in nearly all eukaryotic cytochromes P450. The results are consistent with a model in which amino acids 112-114, and possibly 115, align with a loop of bacterial P450cam sequence containing substrate contacting amino acids, tyrosine 96 and phenylalanine 98. Glycines at either end of the loop in cytochrome P450 2C2 appear to be critical for activity.

Amino Acid Sequence↗

PubMed1993

The amino-terminal 29 amino acids of cytochrome P450 2C1 are sufficient for retention in the endoplasmic reticulum.

Cytochromes P450 are inserted into and anchored to the endoplasmic reticulum (ER) membrane by a hydrophobic signal sequence at the NH2 terminus. To determine whether the NH2-terminal sequence might also have an ER retention function, the NH2-terminal 29 amino acids of cytochrome P450 2C1, with and without an additional 29 amino acids containing an N-glycosylation site, were fused either to a soluble cytoplasmic protein, Escherichia coli beta-galactosidase, or to a secreted protein, E. coli alkaline phosphatase, and the hybrid proteins were expressed in COS1 cells. Subcellular fractionation indicated that both the beta-galactosidase and alkaline phosphatase hybrid proteins cosedimented with marker enzymes for ER membranes, and localization by immunofluorescent staining was consistent with an ER location. Hybrid proteins with the NH2-terminal glycosylation site were glycosylated in COS1 cells, and the carbohydrate moiety was sensitive to endoglycosidase H digestion, providing further evidence that the proteins were retained in the ER. In vitro studies of membrane insertion of the alkaline phosphatase hybrid indicated that fusion to alkaline phosphatase hybrid indicated that fusion to alkaline phosphatase did not alter the topological properties of the cytochrome P450 NH2-terminal sequence. In addition, alkaline phosphatase fused to the extracellular and transmembrane domains of epidermal growth factor receptor was transported to the plasma membrane in COS1 cells, which establishes that alkaline phosphatase as a cytoplasmic domain does not prevent transport from the ER. These observations indicate that the large cytoplasmic domain of cytochrome P450 is not required for retention in the ER and suggest that a specific sequence or structure within the NH2-terminal 29 amino acids functions as an ER retention signal.

Alkaline Phosphatase↗

PubMed1993

Endonuclease VII of phage T4 triggers mismatch correction in vitro.

The reactivity of endonuclease VII (gp49 of phage T4) with DNA-loops of eight, four, or one nucleotide, or any of 12 possible base mismatches was tested in vitro. Endonuclease VII introduces double-strand breaks by nick and counter-nick within six nucleotides 3' from the mispairings. High relative cleavage efficiencies at mismatches in heteroduplexes correlate with their decreased thermal stability and vice versa. A delay between nick and counter-nick was sufficient to allow T4 DNA-polymerase and T4 DNA-ligase to correct a C/C-mismatch in vitro, thereby saving the DNA from double-strand breakage. Very short repair tracks of three to four nucleotides mapped between the mismatch and one of the formerly induced nicks, which were subsequently sealed by DNA ligase.

Bacteriophage T4↗

PubMed1993

A single amino acid substitution confers progesterone 6 beta-hydroxylase activity to rabbit cytochrome P450 2C3.

A cDNA encoding a naturally occurring variant of cytochrome P450 (P450) 2C3 that catalyzes the 6 beta- and 16 alpha-hydroxylation of progesterone exhibits six differences of nucleotide sequence leading to five amino acid substitutions from that encoding 2C3, a progesterone 16 alpha-hydroxylase that does not catalyze 6 beta-hydroxylation. Analysis of chimeric and mutant enzymes indicates that a Ser/Thr difference at position 364 underlies the difference between the two enzymes in 6 beta-hydroxylase activity as well as sensitivity to the inhibitor, 16 alpha-methylprogesterone. In addition, an Ile/Met difference at position 178 influences the apparent Km for progesterone. The two mutations, S364T and 1178M, together convert 2C3 to a form that exhibits kinetic properties which are similar to the 2C3v enzyme, and the reciprocal mutations in 2C3v convert it to an enzyme that resembles 2C3. Interestingly, position 364 of 2C3 maps to a substrate-contacting domain suggested by models for mammalian P450 enzymes based on the structure of P450cam. Ile178 is highly conserved among mammalian microsomal P450s with the exception of CYP4A and CYP19 enzymes which exhibit a Met at this alignment position.

Amino Acids↗

PubMed1993

An N-terminal glycosylation signal on cytochrome P450 is restricted to the endoplasmic reticulum in a luminal orientation.

The mechanism of retention of cytochrome P450 in the endoplasmic reticulum is unknown, and the membrane topology of the N-terminal region remains controversial. To address these problems, a sequence of 29 amino acids encoding an internal N-glycosylation site of rabbit cytochrome P450 2C2 was attached to the N terminus of cytochrome P450 2C1. This protein is glycosylated at a single site in a cell-free translation system containing microsomal membranes, as indicated by gel mobility and sensitivity to endoglycosidase H. When expressed in COS1 cells, an immunoreactive species with the same gel mobility as the in vitro synthesized glycosylated product was detected. Treatment with endoglycosidase H changed its mobility to that of unglycosylated hybrid cytochrome P450 2C1. These results indicate that in intact cells, as in the cell-free system, the N terminus of cytochrome P450 is luminally oriented which is not consistent with a hairpin loop conformation. Sensitivity of the glycosylated protein to endoglycosidase H suggests that the protein does not reach the Golgi compartments. When transfected cells were incubated at low temperatures to inhibit retrograde transport from the intermediate pre-Golgi compartment into the endoplasmic reticulum, localization of cytochrome P450 was not changed, as assayed by subcellular fractionation and immunofluorescent staining. These observations suggest that cytochrome P450 is restricted to the endoplasmic reticulum membrane by a mechanism different from recycling through the intermediate compartment, which is a pathway utilized by soluble endoplasmic reticulum proteins.

Amino Acid Sequence↗

PubMed1993

Purification and characterization of recombinant-expressed cytochrome P450 2C3 from Escherichia coli: 2C3 encodes the 6 beta-hydroxylase deficient form of P450 3b.

Rabbit cytochrome P450 2C3 was expressed from its cDNA in Escherichia coli as a chimeric enzyme in which a portion of the N-terminal membrane anchor sequence of 2C3 was replaced with a modified sequence derived from P450 17 alpha. The nucleotide sequence encoding the N-terminus of P450 17 alpha was modified previously to achieve a high level of expression of P450 17 alpha in E. coli by altering the first eight codons of P450 17 alpha to reflect second codon preferences for high expression and to minimize the potential for the formation of a stable secondary structure of the corresponding RNA transcript. The modified P450 2C3 was expressed at > 400 nmol/liter of culture. P450 2C3 was isolated to apparent electrophoretic homogeneity and a specific content > 14 nmol P450/mg protein. When reconstituted with P450 reductase and dilauroyl-L-alpha-lecithin, the purified E. coli-expressed P450 2C3 catalyzed 16 alpha, but not 6 beta-hydroxylation of progesterone. Expression of unmodified 2C3 from its cDNA in COS-1 cells confirmed the absence of detectable 6 beta-hydroxylase activity. In addition, the enzyme expressed in E. coli is activated by the allosteric effector 5 beta-pregnane-3 beta,20 alpha-diol, with a resultant Vmax = 10 min-1 and Km = 20 microM and is not inhibited by 16 alpha-methylprogesterone. These results indicate that the 2C3 cDNA encodes an enzymatic form characteristic of IIIvo/J and B/J inbred rabbits rather than a second enzymatic form expressed in most outbred and some inbred strains that catalyzes both high efficiency 16 alpha- and 6 beta-hydroxylation of progesterone. Our results have identified the enzyme variant encoded by the 2C3 cDNA and have demonstrated the utility of E. coli for the expression of recombinant P450 enzymes.

Amino Acid Sequence↗

PubMed1993

Deletion of a conserved tetrapeptide, PPGP, in P450 2C2 results in loss of enzymatic activity without a change in its cellular location.

The sequence proline-proline-glycine-proline is highly conserved in cytochrome P450 families 1 and 2, and similar proline rich sequences are found in other cytochromes P450. Since this sequence immediately follows the NH2-terminal hydrophobic membrane insertion signal, it potentially could function as a signal either for retention of cytochrome P450 in the endoplasmic reticulum or for its correct orientation in the membrane. To test this possibility, DNA sequence coding for this tetrapeptide was deleted from cytochrome P450 2C2 cDNA. Translation of the mutated mRNA in a reticulocyte cell-free system containing canine microsomal membranes resulted in the insertion of the protein into the membrane with a topology indistinguishable from that of normal cytochrome P450 2C2. The mutated protein was expressed in COS1 cells and its distribution, assayed by immunofluorescence, was similar to that of cytochrome P450 2C2. Furthermore, if a short peptide containing a potential glycosylation site was fused to the N-terminus of the mutant protein, the new hybrid protein was glycosylated in COS1 cells and the carbohydrate moiety remained sensitive to cleavage by endoglycosidase H. These results indicate that the protein was inserted and retained in the endoplasmic reticulum membrane. Pulse-chase studies showed that the mutated protein was degraded about four times as fast as cytochrome P450 2C2. In contrast to cytochrome P450 2C2, no (omega-1) hydroxylase activity was detected in COS1 cells expressing the mutated protein at similar steady-state levels as the wild-type protein. These results indicate that, although the conserved PPGP tetrapeptide is not required for cellular localization of cytochrome P450 in the endoplasmic reticulum membrane, its deletion decreases the stability of the protein and abolishes enzymatic activity.

Amino Acid Sequence↗

PubMed1993

Function of gene 49 of bacteriophage T4 III. Isolation of Holliday structures from very fast-sedimenting DNA.

Branched DNA molecules were identified in ClaI digests of cytosine containing very fast-sedimenting DNA (VFS-DNACYT) which was isolated from Escherichia coli infected with the multiple mutant 49-GT7 of phage T4. In about 10% of randomly picked ClaI fragments branches with three arms (Y-structures) as well as four arms (Holliday structures) were seen in the electron microscope. Branched structures were absent from ClaI digests after treatment with purified endonuclease VII (gp49) in vitro.

Bacteriophage T4↗