|
|
|
|||||||
|
|||||||||
(Received for publication, September 16, 1994; and in revised form, December 28, 1994) From the
Analysis of the primary structure of peptide synthetases
involved in non-ribosomal synthesis of peptide antibiotics revealed a
highly conserved and ordered domain structure. These functional units,
which are about 1000 amino acids in length, are believed to be
essential for amino acid activation and thioester formation. To
delineate the minimal extension of such a domain, we have amplified and
cloned truncated fragments of the grsA gene, encoding the
1098-amino acid multifunctional gramicidin S synthetase 1, GrsA. The
overexpressed His
Non-ribosomally synthesized peptides produced by several
bacterial and fungal species belong to a diverse family of natural
products that includes antibiotics, immunosuppressants, plant and
animal toxins, and enzyme inhibitors. The manifold biological
activities are emphasized by numerous structural variations of these
peptides. The structures are linear, cyclic, or branched, and they
contain D-, hydroxy, and N-methylated amino acids as
well as other amino acid (aa) ( Sequence comparison of the deduced
primary structure of an increasing number of peptide synthetases
revealed the multidomain arrangement of these enzymes(7) ,
whose specific linkage order has been shown to define the sequence of
the incorporated amino acid
residues(4, 7, 9, 10, 11, 12, 13) .
Protein chemical studies have identified proteolytic fragments with
molecular masses of about 110-120 kDa able to activate an
individual amino
acid(10, 14, 15, 16, 17) .
These fragments seemed to correspond with the characterized activating
spots, defined by Lipmann (18) . On the other hand, sequence
comparisons revealed for a diverse number of multifunctional peptide
synthetases the presence of homologous domains of about 600 amino acids
in length(2, 7) . These regions contain six highly
conserved motifs, designated core sequences (see Fig. 1b), which are typical for a superfamily of
adenylate- and thioester-forming enzymes(7) . Previous studies,
including site-directed mutagenesis and photoaffinity labeling with ATP
analogous, have shown the involvement of the core sequences 2-5
in ATP binding and
hydrolysis(19, 20, 21, 22) . They
also suggested the association of core 6 in covalent binding of the
substrate amino acid(5, 20, 23) . These
results hint that only a conserved domain, rather than a complete
peptide synthetase, performed the specific activation of a cognate
amino acid. However, although core motifs were suggested to be
essential for this activity, it was never shown biochemically that only
the conserved domain could activate an individual amino acid.
Figure 1:
Organization of the
gramicidin S biosynthesis operon and structural architecture of GrsA
and derivatives. a, organization of the entire grs operon, comprising the genes grsT, grsA, and grsB, and the location of the gsp gene, associated
with the gramicidin S biosynthesis system(40) , are shown. The
linkage order of the domains (black boxes, adenylation module; shaded area, thioester formation module) has been proposed to
define the sequence of the incorporated amino acid residues in the
peptide product(2) . b, several highly conserved
motifs of peptide synthetases were identified within gramicidin S
synthetase 1, GrsA. The localization of putative adenylate-forming,
thioester-forming, and racemase units are shown. c, the
structural architecture and the restored core motifs of the deletion
mutants PheATS-His, PheAT-His, and PheA-His are shown. The GrsA
derivative PheATS-His is devoid of the putative racemase region, and
the constructs PheAT-His and PheA-His possess the extension of the
conserved regions of peptide synthetase and adenylate-forming enzymes,
respectively(7) . d, table summarizes the highly
conserved core motifs, their consensus sequences, and their putative
functions within ATP-dependent amino acid
activation.
Based
on sequence alignments and a limited number of biochemical studies, two
types of domains have been characterized within bacterial and fungal
peptide synthetases. Type I comprises about 1100 amino acids and bears
the functions of amino acid activation and thioester
formation(1, 2, 3) , whereas type II carries
in addition an insertion of about 430 amino acids, which may function
as a N-methyltransferase module(12, 13) . The
latter type of domain was only found in fungal peptide synthetases
involved in the synthesis of the cyclic peptides enniatin and
cyclosporin. Both synthetases catalyze the incorporation of N-methylated amino acids. Furthermore, the sequence analysis
revealed a conserved region located to the C-terminal end of type I
domains that activate D-configurated amino acids. Therefore,
this region was designated as a putative racemization
module(8) . In order to understand the structure/function
relationship of peptide synthetases and to determine the arrangement of
putative modules facilitating substrate adenylation, thioester
formation and racemization, we used GrsA as a model enzyme. The grsA gene encodes the 1098-amino acid residue gramicidin S
synthetase 1 (GrsA), which catalyzes the first step in the biosynthesis
of the cyclic decapeptide gramicidin S(4, 24) . It
activates L-phenylalanine and racemizes it to the D-isomer. Therefore, GrsA represents a model for a type I
domain that contains all integrated functions within a single
polypeptide chain with a molecular mass of 126.7 kDa. We define in this
work, by deletion studies, the minimal size of an amino acid activating
domain and determine the locations of integrated modules involved in
ATP-dependent acyladenylation, thioester formation, and epimerization.
The amplification of the grsA fragments was performed using Deep
Vent The PCR products were purified using the
QIAquick-spin PCR purification kit as described by the
manufacturer's protocol (Qiagen, Hilden, Germany). Standard
procedures were used for the digestion with restriction enzymes, the
cloning of the DNA fragments, and the preparation of the transformed
plasmid DNA(28) . We also inspected both the ATG and the
His
Amplification of the grsA fragments
encoding the multienzyme derivatives, was performed as described under
``Experimental Procedures.'' The DNA fragments obtained were
purified and cloned as described. In addition, each mutant plasmid was
confirmed by sequencing the junctions between vector and inserted grsA fragments.
Figure 2:
Expression and purification of the GrsA
derivatives. Samples of the expression and purification of the deletion
mutants were analyzed on 10% SDS-polyacrylamide slab gels (see
``Experimental Procedures''). a, Coomassie-stained
gel showing the total cell extracts of E. coli M15(pREP4/pPheATS-His) at t
The purification of the
expressed proteins was performed as described under ``Experimental
Procedures.'' Using these optimized conditions, we obtained the
recombinant proteins to near homogeneity. Fig. 2b shows
the SDS-PAGE of the purified protein pools. The gel revealed molecular
masses of 91, 70, and 62 kDa for the truncated GrsA deletion mutants
PheATS-His, PheAT-His, and PheA-His, respectively. These data
correspond to the values deduced from the DNA sequences. The protein
pools were dialyzed against assay buffer, and protein concentration was
determined.
The substrate specificity of GrsA deletions was
investigated with respect of the cognate amino acid phenylalanine, as
well as of the non-cognate amino acids that comprise the peptide
antibiotic gramicidin S and the Phe analogue 3-phenylpropionic acid.
These studies were performed using saturating concentrations for all
amino acids tested. As shown in Fig. 3, the amino acid-dependent
activation revealed for all deletion mutants high exchange activities
for D- and L-phenylalanine and almost no activation
in the case of the Phe-carboxyl acid analogue 3-phenylpropionic acid,
proline, valine, ornithine, and leucine. The highest nonspecific
exchange activities obtained for the non-cognate hydrophobic amino
acids L-Val and L-Leu were achieved for the mutant
PheAT-His. This construct also revealed a slight reduction in D-Phe-dependent exchange activity. The slightly reduced
specificity of the GrsA mutant PheAT-His was significant, although the
reason for remains unclear, because the smaller fragment, PheA-His,
restored high specificity for the cognate amino acids D- and L-phenylalanine. Nevertheless, this is the first clear
indication that the core activities involving amino acid recognition
and the ATP-dependent activation are located within the first 556 aa
residues of the multifunctional peptide synthetase GrsA.
Figure 3:
Amino acid specificity of the ATP-PP
All deletion mutants bind the substrates ATP and
phenylalanine with nearly the same affinity as the wild type protein.
The K
The results shown in Fig. 4clearly indicate that the deletions PheATS-His and
PheAT-His restored the ability for thioester formation, whereas the
deletion within GrsA mutant PheA-His caused complete loss of this
activity. We found almost wild type levels for the first two constructs
and decreased binding capacity of 13% for mutant PheA-His. This value
remains in good correspondence with data obtained for TycA mutants,
which are incapable of covalent binding of substrate amino acid. The
latter deletion, PheA-His, which lost the ability to covalently bind
phenylalanine, is devoid of the putative thioester binding site located
within core 6 (Fig. 1, b and c)(5, 20) . Therefore, the results obtained
represent additional support for the possible involvement of the
deleted region in thioester formation. In this context it is important
to notice that the 556-aa residue fragment PheA-His is still active in
amino acid recognition and ATP-dependent activation. The C-terminal 100
aa of PheAT-His including core 6 may represent the site of substrate
covalent binding.
Figure 4:
Thioester formation of GrsA and
derivatives. The wild type protein and the GrsA derivatives were
charged with [
Cleavage of the covalently bound
[
Figure 5:
Analysis of the
protein-[
The results presented define by deletion studies on a peptide
synthetase, GrsA, for the first time the minimal size of an amino acid
activating domain and determine the locations of integrated modules
involved in ATP-dependent acyladenylation, thioester formation, and
epimerization. We dissected the multifunctional peptide synthetase GrsA
and cloned, overexpressed, and purified the truncated fragments. The
successive deletion at the C terminus of GrsA resulted in the defined
mutants PheATS-His, PheAT-His, and PheA-His, possessing molecular
masses of 92.7, 76.2, and 64.3 kDa, respectively. The purified deletion
mutants were examined for their amino acid-dependent ATP-PP All three mutants restored full
ATP-PP In conclusion, all these
findings indicate an almost undisturbed structure/function relationship
of the truncated GrsA derivatives even in the case of a C-terminal
542-aa deletion from the native peptide synthetase (PheA-His). For the
first time, a 556-aa fragment was identified by biochemical studies,
which mediates amino acid recognition and ATP-dependent activation upon
a peptide synthetase. Therefore, it is designated acyladenylation
module. It comprises the first five core motifs and possesses nearly
half the size of native GrsA (Fig. 1c). To study the
formation of the GrsA phenylalanine thioester linkage, deletion mutants
were tested for amino acid incorporation. We found, in good
correspondence to earlier investigations, that the covalent binding of
the substrate amino acid to recombinant GrsA and derivatives expressed
in E. coli is very inefficient(20) . Because the
Phe-activating domain seems to be a poor substrate for the holo-acyl
carrier protein synthetase, which normally catalyzes the transfer of
the cofactor 4`-phosphopantetheine from coenzyme A to the acyl carrier
protein(39) , only about 14% of recombinant GrsA from E.
coli could be charged with [ Sequence analysis
revealed homologous domains for a diverse number of multifunctional
peptide synthetases(7) . The conserved domains of these enzymes
bear, in addition to a sole adenylation domain (see above), an
extension of about 100 aa at their C terminus. This additional region
comprises the putative 4`-phosphopantetheine binding motif
LGGHSL(5, 20) . We imitated both forms of
adenylate-forming enzymes by the constructed deletion mutants PheA-His
and PheAT-His, and we studied their ability for covalent binding of the
cognate amino acid. The results shown verify that the deletions
PheATS-His and PheAT-His almost restored the wild type level for
covalent binding of the substrate amino acid, whereas the deletion of
the putative thioester binding fragment within the GrsA derivative
PheA-His led to complete loss of thioester formation ability.
Therefore, the findings represent additional support for the role of
the deleted region as a thioester formation module. The state of this
C-terminal area as a functional unit was verified by the identification
of a second class of domain (type II) found in fungal peptide
synthetases. These type II domains carry between core 5 and 6, in
addition to type I domains, an insertion of about 430 aa residues that
may function as a N-methyltransferase. All type II domains
known, one in the enniatin synthetase and seven within the deduced
amino acid sequence of the 1,689-kDa cyclosporin synthetase, correspond
with the occurrence of N-methylated amino acids in the peptide
antibiotics(12, 13) . This observation demonstrates
that the acyladenylation module and the thioester-forming module could
be separated by an additional unit necessary for the modification (N-methylation) of the activated amino acid. Another well
known function of the native peptide synthetase GrsA is the
racemization of the substrate amino acid, which takes place on the
thioesterified phenylalanine(39) . A comparison of GrsA with
other peptide synthetases capable of racemizing the L-configurated substrate amino acid revealed four homologous
motifs at the C terminus, which are absent in the non-racemizing
domains. The deletion of this putative racemization site, which is
located within the C-terminal 291 aa residues of GrsA, was part of the
mutant PheATS-His. This GrsA derivative is able to activate
phenylalanine as acyladenylate and to form a covalently bound
thioester, but completely loses the racemization activity observed in
the wild type protein. This finding indicates the possible involvement
of the C-terminal region of GrsA in racemization of the substrate amino
acid. In conclusion, our dissection studies on the multifunctional
GrsA protein as a model for a type I domain of peptide synthetases
clearly point out for the first time the locations of modules involved
in substrate recognition, acylation, thioester formation, and
racemization. Based on biochemical data we identified the amino
acid-activating domain of a peptide synthetase as the 656-aa region,
which is conserved within a superfamily of adenylate- and
thioester-forming enzymes. The extent of those domains represent nearly
half the size of any proteolytic fragments so far described for
specific amino acid
activation(10, 14, 15, 16, 17) .
Therefore, this finding will be very helpful for the development and
realization of domain exchanging experiments in peptide synthetases.
Volume 270,
Number 11,
Issue of March 17, 1995 pp. 6163-6169
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
-tagged GrsA derivatives were
affinity-purified, and the catalytic properties of the deletion mutants
were examined by biochemical studies including ATPdependent amino acid
activation, carboxyl thioester formation, and the ability to racemize
the covalently bound phenylalanine from L- to the D-isomer. These studies revealed a core fragment (PheAT-His)
that comprises the first 656 amino acid residues of GrsA, which
restored all activities of the native protein, except racemization of
phenylalanine. A further deletion of about 100 amino acids at the
C-terminal end of the GrsA core fragment (PheAT-His), including the
putative thioester binding motif LGGHSL, produced a 556-amino acid
fragment (PheA-His) that shows a phenylalanine-dependent aminoacyl
adenylation, but almost no thioester formation. A 291-amino acid
deletion at the C terminus of the native GrsA, that contains a putative
racemization site resulted in complete loss of racemization ability
(PheATS-His). However, it retained the functions of specific amino acid
activation and thioester formation. The results presented defined
biochemically the minimum size of a peptide synthetase domain and
revealed the locations of the functional modules involved in substrate
recognition and ATP-dependent activation as well as in thioester
formation and racemization.
)constituents that can
undergo extensive modifications, including acylation and
glycosylation(1, 2, 3) . The synthesis of
these peptides is accomplished by large multienzyme complexes that
possess a multidomain structure and employ the thiotemplate
mechanism(4) . The reaction sequence involves ATP-dependent
amino acid activation, followed by transferring the acyl adenylate to
specific thiol groups and formation of a carboxyl thioester-bound
substrate. The elongation reaction is assisted by an enzyme-bound
cofactor, the
4`-phosphopantetheine(1, 2, 3, 4) .
Recent biochemical and genetic studies suggested a model for the
elongation reaction that assumes the involvement of multiple cofactors
of the 4`-phosphopantetheine type. One cofactor for each amino acid
activating domain was suggested, rather than the presence of a
centrally located sole cofactor(5, 6) . These
cofactors may facilitate the ordered shift of the carboxyl
thioester-activated amino acids between the domains that assemble the
multienzyme, resulting in the formation of a peptide chain with a
defined sequence(7, 8, 9) . Termination of
this enzyme-catalyzed peptide synthesis is induced by the release of
the thioester-bound peptide by cyclization, hydrolysis, or specific
transfer to a functional group.
PCR Amplification and DNA Manipulations
We used
the pQE60 His tag fusion vector, purchased from Qiagen
(Hilden, Germany), for cloning of the truncated grsA fragments. pQE60 is a derivative of pDS described by Bujard et
al.(25) and contains information required for replication
as described for pACYC184 (26) , the T5 promotor, a ribosome
binding site, a His tag, a small multiple cloning site, and
a transcriptional terminator(27) . The vector allowed the
expression of the genes from the authentic ATG codon, but required the
modification of the sequences around that codon to form a NcoI
restriction site. It was also necessary to create a BamHI site
at the 3` end, because the inserts had to be ligated in-frame
with the His tag included in the vector. To fit these
requirements, the insert sequences could be modified using 5`-modified
PCR primers to generate the terminal restriction sites needed. The
sequences of the oligonucleotides, which were synthesized by Dr.
Michael Krause (Mikrochemische Einheit, FB20,
Philipps-Universität Marburg, Germany), were as
follows (underlined, modified sequences; bold, restriction sites):
oligo 5`-Phe-NcoI: 5`-ATA TCC ATG GTA AAC AGT TCT AAA
AG-3`; oligo 3`-PheA-BamHI: 5`-TCT CGG ATC CTA ATA CAT
CCT GCC AG-3`; oligo 3`-PheAT-BamHI: 5`-ATC GGA TCC ATT
TGG TCT ATA CAA C-3`; and oligo 3`-PheATS-BamHI: 5`-ATA GGA
TCC TAA TTC AAT AGA CCA GTC C-3`.
® DNA polymerase, 10 
reaction buffer from
New England Biolabs (Schwalbach, Germany), and different pairs of
primers. The reaction conditions were: 0.2-20 ng of chromosomal
DNA from Bacillus brevis ATCC 9999, 1-10 pmol of primers
(equimolar), 300 µM deoxyribonucleoside triphosphates
(dNTPs), 10 mM potassium chloride, 10 mM ammonium
sulfate, 20 mM Tris/HCl, pH 8.8 at 25 °C), 6 mM magnesium sulfate, 0.1% Triton X-100, and 1 unit of Deep
Vent® DNA polymerase in a total volume of 100
µl(28) . tag fusion sites between vector and truncated grsA fragments by sequencing using the chain termination method of
Sanger et al.(29) . The sequences of the primers used,
which were also synthesized by Dr. M. Krause, were as follows: oligo
5`-promotor: 5`-GGC GTA TCA CGA GGC CC-3`; 3`-His tag: 5`-ACG CCC GGC
GGC AAC CG-3`.Expression of the His
Modified expression plasmids containing the
amplified grsA fragments were transformed in Escherichia
coli M15(pREP4)(30) . The transformants were used to
inoculate 2 -tagged GrsA
Derivatives YT medium (31) supplemented with ampicillin
(100 µg/ml), kanamycin (25 µg/ml), and magnesium chloride (10
mM). Cells were grown at 30 °C with moderate shaking until A
reached 1.5-1.8. IPTG was added to a
final concentration of 1.5 mM, and cultures were grown for an
additional 3 h. The extent of expression was analyzed by
SDS-polyacrylamide gel electrophoresis (PAGE) using the method of
Laemmli(32) .Enzyme Purification
All operations were carried
out at 4 °C except the chromatography, which was performed at room
temperature. Cells were collected by centrifugation for 15 min at 6,000
g, and resuspended in sonication buffer (50 mM sodium phosphate, pH 8.0 at 23 °C, 300 mM sodium
chloride) at 3 volumes/g of wet weight. Subsequently, the samples were
lysed by freezing, thawing, and sonication using a Branson Sonifier at
middle output. The cell debris was pelleted by centrifugation at 40,000
g for 30 min, and the supernatant obtained was
directly applied to a Ni
-charged chelating Superose
HR 10/2 column, previously equilibrated with sonication buffer. Both
the immobilized metal ion affinity chromatography column and the fast
performance liquid chromatography system were purchased from Pharmacia
LKB Biotechnology Inc. (Freiburg, Germany). The proteins were eluted by
applying imidazole gradients ranging over 0-250 mM in
sonication buffer. Samples were analyzed by SDS-PAGE (10%), pooled,
dialyzed against assay buffer (50 mM sodium phosphate pH 7.8
at 4 °C, 10 mM magnesium chloride, 2 mM dithioerythritol, and 1 mM EDTA), and measured using the
procedure of Bradford (33) .ATP-PP
Essentially,
ATP-PP
Exchange exchange of the GrsA derivatives was measured as
described by Lee and Lipmann(34) . The label, tetrasodium
[
P]pyrophophate (16.06 Ci/mmol), was purchased
from DuPont NEN (Bad Homburg, Germany). An assay mix contained 20 pmol
of enzyme, 0.02-0.2 mM ATP, 0.02-0.2 mMD-Phe, 50 mM sodium phosphate, pH 7.8 at 23
°C, 1 mM dithioerythritol, 0.1 mM EDTA, 1 mM magnesium acetate, 0.1 mM tetrasodium pyrophosphate, and
0.15 µCi of [P]pyrophosphate in a total
volume of 100 µl. Determination of the substrate specificity was
performed under saturating conditions with 0.2 mM of the
cognate and non-cognate amino acids, respectively.Thioester Formation and Cleavage
Covalent binding
of the substrate amino acid phenylalanine was tested as described
previously by Gocht et al.(20) . L-[U-
C]Phe (474 mCi/mmol) was purchased
from Amersham/Buchler (Braunschweig, Germany). Each assay mix contained
0.1 nmol of enzyme, 0 or 2 mM ATP, 50 mM sodium
phosphate, pH 7.8 at 23 °C, 1 mM dithioerythritol, 0.1
mM EDTA, 10 mM magnesium chloride, and 2 µCi of
[C]Phe in a total volume of 100 µl. The
reactions were allowed to proceed for 30 min at 37 °C. They were
stopped by addition of 2 ml of ice-cold 7% trichloroacetic acid (TCA),
and further incubated on ice for 30 min. Trichloroacetic
acid-precipitated proteins were collected on glass fiber filters
(GF-92; Schleicher & Schuell, Dassel, Germany), and washed with an
excess of trichloroacetic acid. After addition of 5 ml of scintillation
mixture (Rotiszint Eco Plus; Roth, Karlsruhe, Germany) retarded
enzyme-bound [C]Phe was counted in a 1900CA
Tri-Carb liquid scintillation analyzer (Packard).Cleavage of the Thioesterified Phenylalanine
The
cleavage of the enzyme-phenylalanine complexes was done as described by
Ullrich et al.(35) . Trichloroacetic acid-precipitated
complexes were washed five times with trichloroacetic acid and two
times with ethanol to remove any traces of unbound label. The
vacuum-dried pellets were suspended in 100 µl of cleavage solution
(80% formic acid and 6% hydrogen peroxide) and incubated at 56 °C
for 30 min. The reaction mixtures were dried again, extracted with 50%
ethanol several times, and subjected to thin-layer chromatography (TLC)
for analysis.Thin-layer Chromatography
Analysis of
protein-[C]Phe cleavage products was done by
chiral plate high performance thin-layer chromatography (2.5 cm
10-cm chiral high performance thin layer chromatography plates; Merck,
Darmstadt, Germany) with solvent system A (methanol-water-acetonitrile;
1:1:4)(23) . The label was identified by autoradiography and by
scanning on a Berthold LB2723 thin-layer scanner II.
Amplification and Cloning of the grsA
Fragments
To delineate the minimal extension of an amino acid
activating domain, we have amplified truncated fragments of the grsA gene. As shown in Fig. 1, we chose the 3` end
oligonucleotides for the PCR in consideration of structural
conservations found in peptide synthetases. The first fragment
amplified comprises 2421 base pairs and encoded the 92.5-kDa mutant
PheATS-His (Fig. 1c). It carries a deletion of 291 aa
at the C-terminal end of the native peptide synthetase GrsA (1098 aa).
This protein is devoid of a putative racemization module. The second
amplified grsA fragment of 1968 base pairs encoded the
76.2-kDa deletion mutant PheAT-His. This construct, comprising 656 aa,
reduced the primary structure of the intact GrsA protein to a conserved
region, typical for adenylate- and thioester-forming enzymes. Also
constructed was a third GrsA deletion. Here an additional 100 aa from
the C-terminal end of PheAT-His was removed, resulting in the 556-aa
(64.3-kDa) fragment PheA-His. This GrsA mutant is devoid of a putative
thioester binding site and possesses the primary structure of a
superfamily of adenylate-forming enzymes(7) . Fig. 1c summarizes the extent of the GrsA deletion
mutants derived.Expression and Purification of Truncated GrsA
Derivatives
High level expression of the GrsA derivatives was
obtained, using freshly transformed E. coli cells containing
pREP4 and pQE60 derivatives. Optimization of the fermentation
parameters of E. coli strain M15 with respect to the culture
medium, the production time, and the concentration of IPTG revealed the
conditions described under ``Experimental Procedures.'' A
very important factor in the solubility of the expressed recombinant
GrsA derivatives was the addition of magnesium chloride. We found that
a concentration of 10 mM Mg in the culture
medium increased the yield of soluble protein from 10% to more than
75%, constituting magnesium as a cofactor of peptide synthetases during
ATP hydrolysis. Using optimized conditions, we found high level
expression with a predominant soluble cytoplasmic location of the GrsA
derivatives (Fig. 2a).
(lane 1)
and 3 h after induction with IPTG (lane 2). Lane 3 shows the dialyzed PheATS-His pool after affinity purification on
chelating Superose. b, the protein pools of PheATS-His (lane 1), PheAT-His (lane2), and PheA-His (lane3) after immobilized metal ion affinity
chromatography are shown.
ATP-PP
The first step of
amino acid activation during non-ribosomal peptide biosynthesis is the
ATP-dependent substrate activation as acyladenylate(36) . We
determined the influence of the introduced C-terminal deletions in GrsA
on Phe-dependent ATP-PP Exchange exchange. As shown in Table 1, only slight alteration in exchange activities were
observed for the GrsA deletions, when compared with the native GrsA
protein. The deletion of a putative racemase module in mutant
PheATS-His induced the activity to nearly 120% of the wild type level
for L-Phe-dependent ATP-PP exchange. The deletions
in GrsA derivatives PheAT-His and PheA-His caused no reduction in
exchange activities and revealed normal levels for the L-Phe-dependent reaction. Interestingly, even the 556-aa
construct PheA-His, possessing nearly half the size of GrsA (1098 aa),
restored the same adenylation activity as the wild type protein (Table 1).
exchange reaction of the GrsA derivatives. The ATP-PP exchange reaction was measured for the three deletion mutants in
dependence on the cognate amino acid phenylalanine, the Phe analogue
3-phenylpropionic acid (3-PPA), and the amino acid residues
that comprise the peptide antibiotic gramicidin S (Pro, Val, Orn, and
Leu, respectively). The highest exchange activity of each mutant was
defined as 100%, and the values obtained for the other amino acid were
settled with those.
Binding Affinities of the Truncated GrsA
Derivatives
We investigated the binding affinities of the mutant
GrsA proteins and the substrates ATP and phenylalanine, respectively.
The kinetic constants (K
)
were determined from Lineweaver-Burk plots in dependence on the
concentration of both reaction partners (37) (data not shown).
Plots obtained at different substrate concentrations intersect at
abscissa values of
-1/K. The kinetic
constants determined for wild type GrsA and mutants are summarized in Table 1. values obtained also
correspond to those determined previously for the four GrsB domains and
TycA(20, 38) . Limited deviations in the binding
affinities for ATP were observed. For the deletion mutants PheATS-His
to PheA-His, the range was between 1.1 and 0.8 mM, indicating
a slight increase in the ATP affinity for the shorter fragments. For
the cognate amino acid phenylalanine, a 20-fold higher binding affinity
could be detected for the GrsA deletions, and almost identical K values were determined
(
60 µM; Table 1). Furthermore, the points of
intersection within the different plots are located in the upper left
quadrant, indicating a random binding of the substrates on GrsA and
derivatives as shown previously(38) . The substrates ATP and
Phe associates independently with the protein, and even if both ATP and
phenylalanine are linked, the formation of the acyladenylate can be
performed.Covalent Binding of Phenylalanine
The second step
of amino acid activation during non-ribosomal peptide synthesis
accomplishes covalent binding of the carboxyl-activated amino acid as
thioester (36) . As reported, the core 6 motif LGGHSL was found
to be involved in the formation of the
thioester(5, 20) . In order to assign the location of
a putative thioester module, the ability of the constructed GrsA
deletions in thioester formation was tested.
C]phenylalanine in the presence
and absence of ATP. The vertical bars represent the amount of
covalently bound [C]phenylalanine (in fmol) per
pmol of enzyme.
C]Phe from GrsA and derivatives was performed
in 80% performic acid and 6% hydrogen peroxide, indicating the
thioester nature of this linkage(35) . To determine the
racemization capacity of the GrsA deletions, the
protein-[C]Phe complexes were cleaved and
subjected to chiral plate TLC. The D- and the L-forms
of the cleaved phenylalanine were distinguished by their R
values of 0.49 and 0.59 in solvent A,
respectively(20) . Fig. 5shows the autoradiography and
the radioactivity scan of the chiral high performance thin layer
chromatography plate. The wild type GrsA protein revealed racemase
activity as determined by signals for the D- and L-enantiomer, respectively (Fig. 5, lane 2).
The C-terminal 291-aa deletion of PheATS-His (Fig. 5, lane
3), which is devoid of a putative racemase module, shows only L-Phe and no D-isomer, indicating for the first time
by biochemical studies a possible role of this C-terminally located
region in racemization of the covalently bound amino acid.
C]phenylalanine cleavage products by
chiral plate TLC. a, autoradiography of separated cleavage
products. Commercially available L-[C]Phe, slightly contaminated with D-[C]Phe, was used as control (lane
1). Cleavage products (D-Phe, L-Phe, and N-formyl-Phe; (35) ) of the wild type protein (lane2) and the GrsA deletion mutant PheATS-His (lane3) are shown. b, radioactivity scan of lanes2 and 3.
exchange activity and specificity, their substrate affinity,
their capacity for covalent binding of phenylalanine, and their ability
to racemize the cognate amino acid. exchange activity, revealing a core fragment of this
activity consisting of 556 aa. This adenylate-forming area comprises
the core sequences 1-5 and possesses nearly half the size of
native GrsA (1098 aa residues). With the exception of core 1, these
core motifs are believed to be involved in ATP binding and
hydrolysis(20, 21, 22) . The extent of the
present core motifs coincides with the conserved regions reported for a
family of adenylate-forming enzymes, including EntE from E.
coli, luciferase from P. pyralis, 4-coumarate:CoA ligase
from rat, and acetyl:CoA synthetase from A.
nidulans(2, 7) . Interestingly, the substrates of
these enzymes are entirely carboxyl acid derivatives; however, even the
64.3-kDa core fragment (PheA-His) only activates phenylalanine but not
the phenylalanine carboxyl acid derivative 3-phenylpropionic acid.
These findings and the exclusive activation of the cognate amino acid
strongly suggest the location of the amino acid recognition site within
the first 556 aa of native GrsA enzyme. We also investigated the
binding affinities between truncated GrsA deletions and the substrates
ATP and Phe, respectively. The Michaelis-Menten constants K determined revealed only
slight alteration in the binding affinities and corresponded to values
published for several other native peptide
synthetases(20, 38) .C]Phe. A
similar finding was also observed for the peptide synthetase TycA, when
expressed in E. coli(20) .
)
-D-galactopyranoside; PAGE,
polyacrylamide gel electrophoresis; PheA, PheAT, and PheATS, deletion
mutants of the phenylalanine (Phe)-activating GrsA protein containing
adenylation (A), thioester formation (T), and spacer (S) modules,
respectively.
We thank Inge Schüler for
technical assistance and Stefan Borchert, Kürsad
Turgay, Martin Gocht, and Oliver de Peyer for discussion and comments
on the manuscript.
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  A. D. Berti, N. J. Greve, Q. H. Christensen, and M. G. Thomas Identification of a Biosynthetic Gene Cluster and the Six Associated Lipopeptides Involved in Swarming Motility of Pseudomonas syringae pv. tomato DC3000 J. Bacteriol., September 1, 2007; 189(17): 6312 - 6323. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Rachid, D. Krug, K. J. Weissman, and R. Muller Biosynthesis of (R)-beta-Tyrosine and Its Incorporation into the Highly Cytotoxic Chondramides Produced by Chondromyces crocatus J. Biol. Chem., July 27, 2007; 282(30): 21810 - 21817. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C.-Y. Wu, C.-L. Chen, Y.-H. Lee, Y.-C. Cheng, Y.-C. Wu, H.-Y. Shu, F. Gotz, and S.-T. Liu Nonribosomal Synthesis of Fengycin on an Enzyme Complex Formed by Fengycin Synthetases J. Biol. Chem., February 23, 2007; 282(8): 5608 - 5616. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. A. Chan, M. T. Boyne II, A. M. Podevels, A. K. Klimowicz, J. Handelsman, N. L. Kelleher, and M. G. Thomas Hydroxymalonyl-acyl carrier protein (ACP) and aminomalonyl-ACP are two additional type I polyketide synthase extender units PNAS, September 26, 2006; 103(39): 14349 - 14354. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B.-N. Lee, S. Kroken, D. Y. T. Chou, B. Robbertse, O. C. Yoder, and B. G. Turgeon Functional Analysis of All Nonribosomal Peptide Synthetases in Cochliobolus heterostrophus Reveals a Factor, NPS6, Involved in Virulence and Resistance to Oxidative Stress Eukaryot. Cell, March 1, 2005; 4(3): 545 - 555. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Kessler, H. Schuhmann, S. Morneweg, U. Linne, and M. A. Marahiel The Linear Pentadecapeptide Gramicidin Is Assembled by Four Multimodular Nonribosomal Peptide Synthetases That Comprise 16 Modules with 56 Catalytic Domains J. Biol. Chem., February 27, 2004; 279(9): 7413 - 7419. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S.-W. Lu, S. Kroken, B.-N. Lee, B. Robbertse, A. C. L. Churchill, O. C. Yoder, and B. G. Turgeon A novel class of gene controlling virulence in plant pathogenic ascomycete fungi PNAS, May 13, 2003; 100(10): 5980 - 5985. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. F. Ackerley, T. T. Caradoc-Davies, and I. L. Lamont Substrate Specificity of the Nonribosomal Peptide Synthetase PvdD from Pseudomonas aeruginosa J. Bacteriol., May 1, 2003; 185(9): 2848 - 2855. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. J. Hijarrubia, J. F. Aparicio, and J. F. Martin Domain Structure Characterization of the Multifunctional alpha -Aminoadipate Reductase from Penicillium chrysogenum by Limited Proteolysis. ACTIVATION OF alpha -AMINOADIPATE DOES NOT REQUIRE THE PEPTIDYL CARRIER PROTEIN BOX OR THE REDUCTION DOMAIN J. Biol. Chem., February 28, 2003; 278(10): 8250 - 8256. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Velkov and A. Lawen Mapping and Molecular Modeling of S-Adenosyl-L-methionine Binding Sites in N-Methyltransferase Domains of the Multifunctional Polypeptide Cyclosporin Synthetase J. Biol. Chem., January 3, 2003; 278(2): 1137 - 1148. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. R. Mofid, R. Finking, and M. A. Marahiel Recognition of Hybrid Peptidyl Carrier Proteins/Acyl Carrier Proteins in Nonribosomal Peptide Synthetase Modules by the 4'-Phophopantetheinyl Transferases AcpS and Sfp J. Biol. Chem., May 3, 2002; 277(19): 17023 - 17031. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. J. Welch, S. Chai, and J. H. Crosa The Overlapping angB and angG Genes Are Encoded within the trans-Acting Factor Region of the Virulence Plasmid in Vibrio anguillarum: Essential Role in Siderophore Biosynthesis J. Bacteriol., December 1, 2000; 182(23): 6762 - 6773. [Abstract] [Full Text]
|
|
|
|
|
|
B. Silakowski, H. U. Schairer, H. Ehret, B. Kunze, S. Weinig, G. Nordsiek, P. Brandt, H. Blocker, G. Hofle, S. Beyer, et al. New Lessons for Combinatorial Biosynthesis from Myxobacteria. THE MYXOTHIAZOL BIOSYNTHETIC GENE CLUSTER OF Stigmatella aurantiaca DW4/3-1 J. Biol. Chem., December 24, 1999; 274(52): 37391 - 37399. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. M. Wertheimer, W. Verweij, Q. Chen, L. M. Crosa, M. Nagasawa, M. E. Tolmasky, L. A. Actis, and J. H. Crosa Characterization of the angR Gene of Vibrio anguillarum: Essential Role in Virulence Infect. Immun., December 1, 1999; 67(12): 6496 - 6509. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. H. Duitman, L. W. Hamoen, M. Rembold, G. Venema, H. Seitz, W. Saenger, F. Bernhard, R. Reinhardt, M. Schmidt, C. Ullrich, et al. The mycosubtilin synthetase of Bacillus subtilis ATCC6633: A multifunctional hybrid between a peptide synthetase, an amino transferase, and a fatty acid synthase PNAS, November 9, 1999; 96(23): 13294 - 13299. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T.-P. Lin, C.-L. Chen, L.-K. Chang, J. S.-M. Tschen, and S.-T. Liu Functional and Transcriptional Analyses of a Fengycin Synthetase Gene, fenC, from Bacillus subtilis J. Bacteriol., August 15, 1999; 181(16): 5060 - 5067. [Abstract] [Full Text]
|
|
|
|
|
|
D. Konz, S. Doekel, and M. A. Marahiel Molecular and Biochemical Characterization of the Protein Template Controlling Biosynthesis of the Lipopeptide Lichenysin J. Bacteriol., January 1, 1999; 181(1): 133 - 140. [Abstract] [Full Text]
|
|
|
|
|
|
T. Stachelhaus, H. D. Mootz, V. Bergendahl, and M. A. Marahiel Peptide Bond Formation in Nonribosomal Peptide Biosynthesis. CATALYTIC ROLE OF THE CONDENSATION DOMAIN J. Biol. Chem., August 28, 1998; 273(35): 22773 - 22781. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G.-H. Lin, C.-L. Chen, J. S.-M. Tschen, S.-S. Tsay, Y.-S. Chang, and S.-T. Liu Molecular Cloning and Characterization of Fengycin Synthetase Gene fenB from Bacillus subtilis J. Bacteriol., March 1, 1998; 180(5): 1338 - 1341. [Abstract] [Full Text]
|
|
|
|
|
|
A. Elsner, H. Engert, W. Saenger, L. Hamoen, G. Venema, and F. Bernhard Substrate Specificity of Hybrid Modules from Peptide Synthetases J. Biol. Chem., February 21, 1997; 272(8): 4814 - 4819. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Stein, J. Vater, V. Kruft, A. Otto, B. Wittmann-Liebold, P. Franke, M. Panico, R. McDowell, and H. R. Morris The Multiple Carrier Model of Nonribosomal Peptide Biosynthesis at Modular Multienzymatic Templates J. Biol. Chem., June 28, 1996; 271(26): 15428 - 15435. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T Stachelhaus, A Schneider, and M. Marahiel Rational design of peptide antibiotics by targeted replacement of bacterial and fungal domains Science, July 7, 1995; 269(5220): 69 - 72. [Abstract] [PDF]
|
|
|
|
|
|
H.-P. Stuible and E. Kombrink Identification of the Substrate Specificity-conferring Amino Acid Residues of 4-Coumarate:Coenzyme A Ligase Allows the Rational Design of Mutant Enzymes with New Catalytic Properties J. Biol. Chem., July 13, 2001; 276(29): 26893 - 26897. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASBMB Journals | Molecular and Cellular Proteomics |
| Journal of Lipid Research | ASBMB Today |