Redacted for privacy - COnnecting REpositories · Syntheses and feedings of deuteriated Dap (11a...
Transcript of Redacted for privacy - COnnecting REpositories · Syntheses and feedings of deuteriated Dap (11a...
AN ABSTRACT OF THE THESIS OF
Mu Wang for the degree of Master of Science in Chemistry presented on June
17. 1993.
Title: Studies on the Biosynthesis of Capreomycin
Abstract approved:
Professor Steven . Gould
Capreomycin is an antitubercular antibiotic produced by Streptomyces capreolus. It
consists of four unique structural components: IA, IB, 11.A. and IIB.
The biosynthesis of capreomycin has been studied by administration of isotopically
labeled (14C, 13C and 2H) precursors to S. capreolus A250. The results from feeding
radiolabeled compounds indicated that serine was a most likely precursor. Feedings of
stable isotopically labeled serine (10a and 10b) suggested that the 2,3-diaminopropionate
(Dap) moieties of capreomycin are derived from serine via a dehydroalanyl intermediate.
The level of enrichment at the serine residue of capreomycin IA, 1, compared to those at
the other sites of 1 and at all sites of capreomycin IB, 2, suggested that the serine residue
in 1 derives from serine directly whereas the alanine residue in 2 comes from alanine, and
this was later confirmed by feeding labeled alanine (12a). Therefore, 1 is derived
independently from 2 rather than by a hydroxylation of 2.
Syntheses and feedings of deuteriated Dap (11a and 11b) demonstrated that Dap is
an intermediate in the biosynthesis of capreomycin, and that the P-ureido-dehydroalanine
(Uda) residue is also derived from Dap, presumbly by addition of carbamoyl phosphate.
Redacted for privacy
Preliminary experiments have detected that cell-free extracts of S. capreolus contain
the enzymatic activity which in the presence of the precursor amino acids of capreomycins
catalyzes an ATP /[32P]pyrophosphate exchange reaction. The experiments suggest that the
substrates (precursor amino acids) are activated by the formation of substrate-AMP in
correspondence with a non-ribosomal "thiotemplate" mechanism of peptide biosynthesis.
The further purification and characterization of these enzyme activities are clearly of
interest.
STUDIES ON THE BIOSYNTHESIS OF CAPREOMYCIN
by
Mu Wang
A THESIS
submitted to
Oregon State University
in partial fulfillment of
the requirements for
the degree of
Master of Science
Completed June 17, 1993
Commencement June, 1994
Approved:
Professor of Chemistry in Charge o ajor
Chairman of Department of Chemistry
Date Thesis is Presented: June 17. 1993
Typed by Mu Wang for Mu Wang
Redacted for privacy
Redacted for privacy
Redacted for privacy
To My Wife Jirong , My Son Kevin and My Parents
Bad moments, like good ones, tend to be grouped together.
Edna O'Brien
Acknowledgements
First and foremost, I am deeply indebted to my advisor, Dr. Steven J. Gould, for
his inspiring guidance and unfailing enthusiasm which made this work possible.
I also would like to thank Drs. Donna Minott and Martha Cone for their suggestions
and help when I first started my research in the "Gould Group". My special recognition
goes to Dr. Tom O'Hare, Dr. Nuria Tamayo and Marc Kirchmeier, for their advice, helpful
discussions and most of all for their friendship. Mr. Rodger Kohnert is thanked for hishelp with NMR spectroscopy.
Finally, I wish to express my gratitude to my wife Jirong, my son Kevin and my
parents for their continuous understanding and support
Table of ContentsPage
Introduction 1
Capreomycin: A Cyclic Pentapeptide Antibiotic 1
Structurally Related Peptide Antibiotics 4
Previous Biosynthetic Studies on Capreomycin 6
Objectives and Significance of Present Studies 10
Results and Discussion 13
Fermentation Study 13
Bioassay with Bacillus subtilis 13
Fermentation of S. capreolus 14
Purification of Capreomycins 16
Synthesis 17
DL- [2,3,3- 2H3]Dap (11a) 17
DL-[2-2H]Dap (11b) 18
Hydrolysis of Capreomycin and Purification of Capreomycidine (5) 19
Feeding Experiments 20
L-R1-14C1Serine, L-[U-14C]Alanine and L-[U-14C[Cystine 20
DL-[1-13C]Serine (10a) 20
DL- [3- 13C]Alanine (12a) 24
DL-[2,3,3-2H3]Serine (10b) 27
DL- [2,3,3- 2H3]Dap (11a) 30
DL- [2- 2H]Dap (11b) 30
Exchange Reaction 33
Studies with Cell-free Extract from S. capreolus 35
Preparation and Partial Purification of Cell-free Extract (CI.E) 36
Protein Concentration Determination 37
Enzyme Assay 38
Conclusions and Areas for Further Study 41
Experimental
General
44
44
Table of Contents, Continued
Page
Fermentation of S. capreolus A250 45
Maintenance of S. capreolus A250 45
Production of Capreomycins 45
Isolation of Capreomycins IA and 113 46
Bioassay 47
Synthesis 47
DL-[2,3,3-2H3]Dap (11a) 47
DL- [2- 2H]Dap (11b) 49
Hydrolysis of Capreomycin and Purification of Capreomycidine (5) 49
Feedings 50
Feeding Protocol 50
Calculations in Feeding Experiments 51
L-[U-14C]Serine, L-[U-14C]Alanine and L-[Up4qCystine 51
DL- [1- 13C]Serine (10a) 52
DL- [3- 13C]Alanine (12a) 52
DL-[2,3,3-2H3]Serine (10b) 53
DL-[2,3,3-2H3]Dap (11a) 53
DL- [2- 2H]Dap (11b) 53
Exchange Reaction of Capreomycin 1B under the Fermentation Conditions 54
Cell-free Extract from S. capreolus 55
Preparation and Partial Purification of Cell-free Extract from S. capreolus 55
Protein Concentration Determination 56
Enzyme Activity Assays 56
Bibliography 58
List of Figures
Figure Page
1. Structure of Capreomycins 1
2. Schematic View of Viomycin and Capreomycins 2
3. Proton and Carbon Assignments of Capreomycin IA 3
4. Proton and Carbon Assignments of Capreomycin IB 4
5. Structure of Tuberactinomycins 5
6. Structure of LL-BM547r3 and LL-BM547a 6
7. Proposed Biogenesis of the Streptolidine Moiety of Streptothricin F (7) 7
8. Proposed Biogenesis of Capreomycidine 8
9. Scheme of Ribosomal (a) and Non-ribosomal (b) Biosynthesis of Peptide
Antibiotics 12
10. Dose-dependent Inhibition of B. subtilis Growth by Capreomycin
Concentration Dependence 14
11. Production Curve of Capreomycins 16
12. Typical Chromatogram of Capreomycins 17
13a. Enrichments over Natural Abundance Observed in lb 21
13b. Enrichments over Natural Abundance Observed in 2b 22
14. Proposed Metabolic Pathways for Incorporation of [1-13C]Serine into
r3-Lysine, Capreomycidine and p-Ureido-dehydroalanine 23
15a. Enrichments over Natural Abundance Observed in lc 24
15b. Enrichments over Natural Abundance Observed in 2c 25
16. Proposed Metabolic Pathway for [3-13C]Alanine Feeding 26
17. NMR Spectra of 2d 28
18. NMR Spectra of ld 29
19. Partial NMR Spectra of le and if 31
20. Partial NMR Spectra of 2e and 2f 32
21. Summary of Labeling Patterns of 1 and 2 from Feeding
Experiments with Precursor Amino Acids 34
22. Bradford Assay: BSA Standards 37
23. The Apparent Manner of Capreomycin Assembly 42
List of Tables
Page
Table 1 Results of Radioactive Precursor Feedings 20
Table 2 Specific Activity of [32P]ATP from ATP/[32P]PPi
Exchange Reaction 40
STUDIES ON THE BIOSYNTHESIS OF CAPREOMYCIN
Introduction
Capreomycin: A Cyclic Pentapeptide Antibiotic
The antitubercular antibiotic capreomycin, a metabolite of Streptomyces
capreolus, was first isolated by Herr, et al. in 1960.1 It was subsequently reported to be
composed of two components I and H in 1962,2 and in 1965 was finally shown to
actually contain four components: IA, re, HA and BB (1, 2, 3, and 4, respectively) as
shown in Figure 1.3
RO
H2N14eN
ONH HN
N N NH2
0NH
N
1: R = OH, R' f3- Lysine2: R = H, R' = f3- Lysine3: R = OH, R' =H4: R = H, R' = H
Figure 1. Structure of Capreomycins
In general, capreomycin, which resembles viomycin4 chemically and
pharmacologically, is a second-line agent employed in combination with other
antitubercular drugs. It is active against a number of Gram-positive and Gram-negative
bacteria, but is primarily effective against mycobacteria. In particular, it may be used in
2
place of standard antitubercular drugs such as streptomycin and rifampicin where either
the patient is sensitive to, or the strain of Mycobacterium tuberculosis is resistant to
streptomycin or rifampicin.5,6 The structures of capreomycins were proposed by a
Japanese group in l976,7 but was later revised 8,9 It was concluded that the amino acid
sequence in the cyclic peptide moiety corresponded to that of viomycin. However, the
position linking the branched amino acid, 3- lysine, was surprisingly found to differ from
that of viomycin (Figure 2). When capreomycin was dinitrophenylated and subsequently
hydrolyzed, Na-dinitrophenyl-diaminopropionate (Dap) but not Ni-dinitrophenyl-
diaminopropionate was detected.
Da Pt
Dap1
Sere Sera
Tbd5 Uda4
Viomycin
Dap3 13-Lys
Cpd 5 Uda4
X
Capreomycin IA : SerineCapreomycin IB : Alanine
Abbreviations: Tbd: L-tuberactidine; Dap: 2,3-diaminopropionic acid;
Cpd: L-capreomycidine; Uda: 13-ureido-dehydroalanine.
Figure 2. Schematic View of Viomycin and Capreomycins
3
Detailed examinations of NMR spectra of capreomycins, particularly for the a-
amide and a-methine protons of Dapi and 0-amide proton of Dap3, as well as Edman-
dansyl degradation on reduced capreomycin IA, strongly supported the structure depicted
in Figure 1, in which the 0-lysine residue is linked to the 0-amino group of the Dap3
residue.10 Total syntheses of capreomycins IA and IB have been reported," and
complete assignments of 1H and 13C NMR spectra of both capreomycins IA and IB
(Figure 3 and Figure 4) have been carried out using one and two dimensional NMR
spectroscopy.12
4.41
H2 H
52.37
3.87 HOH
63.80
168.63 56.20
H3.24 40.89
4.1H 0NH
172.52
0 56.60
4.53
1.74
2.12 24.35
3.33
Hsig
4.843.80 337
NH2 1.82 1.82
H31530.48
173 53 4983 24.36
H H3.11
173.7
30 55.04
173.13
0
10621168 79 NH2
10 136.70 14
39.40
H 4.44H
172.83 2.69
1.82 1.82
50.48 Q
NH
37.49 7 155.87
NH2
H 8.05 0
Figure 3. Proton and Carbon Assignments of Capreomycin IA
NH2
H2
19.63CH3 144
324.68
H
3.86 3.71O
3.29 40.96
H ,, ----
4.15n NH H( H 0
1 172.40lisl
H157.92 NH2
H 5.00 10 136.71 14
2.81 2.691.82 1.82
1.71
2.09 24.24
3.34HH
Figure 4. Proton and Carbon Assignments of Capreomycin IB
NH2
4
Structurally Related Peptide Antibiotics
Additional members of this group, called tuberactinomycins (Figure 5), have
subsequently been characterized from other species. Although they are produced by
different species or genera of bacteria, the common structural pattern among these
antibiotics suggests that they are biogenetically related.
H2N
NH2 0N Tr
ONH u/HN
0"N
HN
H
0NH
R2 N NH2H
OH
0
NH2
0
Tuberactinomycin R1 R2
A OH OHB (Viomycin) H OHN OH HO H H
Figure 5. Structure of Tuberactinomycins
5
Tuberactinomycin was initially isolated from fermentation broths of Streptomyces
griseoverticillatus var. tuberacticus, which had been found in the soil at Ohitocho near
Mount Fuji, Japan, by Nagata et al. of the Toyo Jozo Company in 1968.13 The original
strain produced tuberactinomycins A and B, which were separated by ion exchange
chromatography. The chromatographic properties of tuberactinomycin B were found to
be identical to those of the known antibiotic viomycin, whose structure had not been
established at the time.14 From a mutant obtained by nitrosoguanidine treatment of the
original strain, new congeners N and 0 were later isolated.14-16 All congeners of
tuberactinomycin exhibit remarkable and comparable activities against, in particular,
Mycobacterium and Corynebacterium.13-16 Commercially available enviomycin,17
which is used as an antitubercular agent, is composed mainly of tuberactinomycin N.
6
Other antitubercular peptide antibiotics, LL-BM54713 and LL-BM547a (Figure
6), were isolated from a Nocardia species, Leder le culture BM547. Studies against M.
tuberculosis in mice showed LL -BM547 j3 to be as effective as either viomycin or
streptomycin. However, the tolerated dose of LL-BM54713 was 32 times less than that of
streptomycin and 16 times less than that of viomycin. The biological activity of LL-
BM547a is less than that of LL-BM54713 or viomycin.16
NH
HO NNH
LL-BM54713: R =110-CH3-13-arginyl
LL-BM547a: R = H
Figure 6. Structure of LL-BM547r3 and LL-BM547a
Previous Biosynthetic Studies on Capreomycin
The biochemical origins of the residues that constitute capreomycins have
received little attention. The first biosynthetic study of capreomycin was recently carried
out in our research group,18 and focused on L-capreomycidine (5), the most unusual
amino acid component in the capreomycins. The synthesis of capreomycidine has been
reported.19 Its biosynthetic origin was proposed20 to be as an intermediate in the
biosynthesis of streptolidine (6),21 a key component of streptothricin antibiotics,22,23
such as streptothricin F (7, Figure 7) .24,25
OH H
H
Ne NHH H II
0
H2N
COOH
9
0II
H2NCOOH
H
5
HO HN NH2
0 NH2
7
7
Figure 7.18 Proposed Biogenesis of the Streptolidine Moiety (6) [shown in box]
of Streptothricin F (7)
The biosynthesis of 7 has been studied extensively in our group,26-31 and the
intermediacy of 5 was fully consistent with results from the incorporation of a series of
15N-13C-labeled arginines.27,28 However, in further probing the mechanism of formation
of the cyclic guanidine ring of the putative capreomycidine moiety, unusual results were
obtained with a variety of arginines labeled with deuterium at H-2 and/or H-3.28 When
the pentadeuteriated arginine (8a) was fed to a fermentation culture of Streptomyces L-
1689-23,31 loss of deuterium from both H-2 and H-3 was observed in the product thus
obtained, 7a (Scheme 1). This is inconsistent with the original proposal involving a,0-
dehydroarginine (9, Figure 8, pathway a),21 but could still be rationalized by a pathway
via 3-hydroxy- or 3-keto-arginine (Figure 8, pathway b or c).28
8
H2N
H
H2N7 COOH
8
1H2N
H2N COOH
N
0A COOH
H2NOH
H2NH COOH
H2N
H2N
9
Scheme 1
HN NH
HN
H2N COOH
5
Figure 8. Proposed Biogenesis of Capreomycidi
HN. ID
D
H
00HC HD
8a
S. L-1689-23
7a
9
However, when both erythro- and threo-3-hydroxyarginines were synthesized and
tested as intermediates the results were negative,31 neither being incorporated nor
apparently produced de novo. Therefore, the biosynthesis of 5 in S. capreolus became
significant in terms of mechanistic studies of the formation of capreomycidine.
The investigation of the biosynthesis of 5 in S. capreolus involved feeding both
14C-labeled arginine and 2H-labeled arginine to fermentations of S. capreolus and
determining the 14C incorporation and 2H enrichment in isolated capreomycins by liquid
scintillation counting and 2H NMR spectroscopy, respectively. When L-RJ-14C]arginine
was fed, it was incorporated into both IA (8.5% incorporation) and IB (12.7%
incorporation), suggesting that arginine is one of the primary precursors for the
biosynthesis of capreomycins. In order to obtain positional information, a sample of the
deuteriated arginine (8a) was next fed, and the samples of capreomycins IA and IB, la
and 2a, respectively, derived from 8a were analyzed by 2H NMR spectroscopy. The
results, shown in Scheme 2, clearly revealed the retention of one deuterium atom at H-3
and two deuterium atoms at H-5 but not at H-2.
Scheme 2
COOHH H 0
8a la, 2a
S. capreolusH
This result suggested that the capreomycidine (5) residue is derived from arginine
(8) via an a,13-dehydroarginyl intermediate (Figure 8, pathway a). Since the manner of
incorporation of 8a into the capreomycins is apparently quite different than into
streptothricin F, this study did not shed light on the latter process. However, the
10
capreomycin results are readily explained by intramolecular addition of the guanidine to
an a,13-dehydroarginyl residue. Since in the capreomycins capreomycidine is actually
found as part of a peptide structure, and N-acyl-a,13-dehydroamino acids are stable, it is
reasonable to assume that the capreomycidine ring arises from a post-peptide assembly
modification of arginine in a peptide intermediate.
Objectives and Significance of Present Studies
The present studies mainly focused on determining the origins of the remaining
four amino acid residues of the cyclic pentapeptide cores of 1 and 2. The Dap and Uda
residues represent non-protein amino acids. The Dap residue, which is also found in
malonomicin (active against Trypanosoma congolense),32 L-N3-oxalo-Dap (a neurotoxin
responsible for neurolathyrism),33,34 and the tuberactinomycin35 and bleomycin36
families (antituberculosis and antitumor antibiotics, respectively), was proposed to come
from serine via a dehydroalanyl intermediate (Scheme 3).37 We recognized that two
other potential precursors, alanine and cysteine, should also be tested. The Uda residue
may be formed from addition of a guanido-group from arginine to a dehydroalanyl
residue followed by another dehydrogenation (Scheme 4, pathway a), or by addition of an
ureido-group to a Dap followed by another dehydrogenation (Scheme 4, pathway b).
Scheme 3
COOH
10
OH
"-H20" "NH3"
Scheme 4
NH2
11
N)LNH2 4..
NH2
Since 1 and 2 differ only in the presence or absence of a hydroxyl at C-32, it was
possible that 1 was derived by hydroxylation of 2 rather than by substitution of serine for
alanine in the biosynthesis. In an attempt to test this in the previous study,1 8
capreomycin IB obtained from the DL-[l -14C]arginine feeding was fed to a new
fermentation of S. capreolus. However, very little capreomycin IA, with no radioactivity,
was obtained. From this result, it was not clear whether IA is produced independently
from IB or that the administered IB could not enter the cell due to impermeability to the
(potentially toxic) compound. This needed to be further investigated. In addition,
although arginine was found to be the precursor of capreomycidine (5), it was still not
clear whether 5 arises from a pre- or post-peptide assembly modification of arginine. An
in vivo test of labeled 5 would make the whole picture clearer.
After the biosynthetic origins of the four core amino acids in the capreomycins
were determined, the next objective was to investigate the mechanism of the formation of
the peptide. Although a few examples of ribosomal biosynthesis of secondary metabolic
12
peptides have been reported,38 essentially all microbial peptide antibiotics so far studied
at the cell-free level have been shown to be formed by a non-ribosomal mechanism.39,40
Figure 9 summarizes both ribosomal and multienzymatic (non-ribosomal) processes of
peptide biosynthesis.39
In the ribosomal system, amino acids are activated by aminoacyl-tRNA
synthetases. These are symbolized here by their different types of subunit structures.
Each charged tRNA is then lined up at the ribosome according to the mRNA sequence.
The non-ribosomal biosynthesis is not really as well understood. Schematically, in this
system polyenzymes activate amino and hydroxy acids, and the template itself is
aminoacylated, generally, although not exclusively, as a thioester. A transport system
then assembles the lined-up residues, presumably by a sequential mechanism.39 The
latter, which involves multifunctional/multienzyme complexes, may be involved in the
biosynthesis of capreomycins. Thus, the capreomycins present an opportunity to study the
mechanisms and relationships of extensive pre- and post-peptide assembly modification,
in addition to the peptide assembly process itself. Our first target at the cell-free level
focused on finding constituent amino acid activating enzyme(s).
a.
b.
thioester
9s'
Arrinoacyl-tRNA synthetases
transportprotein
Figure 9.39 Scheme of Ribosomal (a) and Non-ribosomal (b) Biosynthesis of Peptide
Antibiotics
Results and Discussion
Fermentation Study
13
The fermentation of S. capreolus and the bioassay conditions, as well as
purification methods for capreomycins had been worked out by Dr. Donna A. Minott in
our laboratory. These procedures18 were utilized, with a few modifications, in all of the
feeding experiments.
Bioassay with Bacillus subtilis
Bioassay with B. subtilis ATCC 6633 was found to be sufficiently sensitive (MIC
3.13 tig/mL) for our purposes and much simpler than the previously reported
turbidometric assay.2 Prior to the bioassay, PA-7 agar (0.5% Bacto-tryptone + 1.5%
Bacto-agar) was inoculated with appropriate amounts of a diluted B. subtilis ATCC 6633
saline suspension. Sterile stainless steel wells were used as reservoirs for the antibiotic
solution. After 12 h of incubation at 37 0C, clear, distinct zones of inhibition had
appeared. A standard bioassay curve was obtained by using different concentrations of
commercially available capreomycin sulfate (Sigma). Plotting the log2 of concentration
versue the diameter of the inhibition zone gave the line shown in Figure 10.
14
20
19
18
17
16
15
14
13
12
11
10
9
8
7
61 2 3 4 5 6 7 8 9 10 11 12 13 14
Log 2 Concentration (pg/mL)
Figure 10. Dose-dependent Inhibition of B. subtilis Growth by Capreomycin
Concentration Dependence
Fermentation of S. capreolus
Fermentation conditions for S. capreolus had been tested by Dr. Donna A. Minott
in our laboratory, and the complex production medium2 was found to be suitable for our
needs. The recipes for the seed medium41 and the production medium are listed below.
15
Seed medium:
Tryptone 0.30 g
Yeast extract 0.15 g
Glucose 0.90 g
MgSO4.7H20 0.18 g
dd. H2O 60 mL (in 250 mL Erlenmeyer flask)
The pH of the solution, initially 6.8, was adjusted to 7.5 using 2% KOH before
sterilization.
Production medium:
Glucose 5.60 g
Bacto-peptone 3.00 g
N-Z Amine A 0.80 g
Blackstrap molasses 2.00 g
Starch (potato) 2.00 g
CaCO3 0.40 g
MgSO4.7H20 1.00 g
Hot tap water 200 mL (in 1000 mL Erlenmeyer flask)
The pH (6.5) was not adjusted.
Generally, the production of capreomycins (IA and IB) was in the range of 2-3
mg/mL as determined by bioassay, of which 20-40% (approximately equal amounts of
IA and IB) was recoverable by the isolation procedure. A study was performed to monitor
the production of capreomycins as a function of time (Figure 11).
50
40
E-a)E
0
maz 20
10
1:1I I I § . t
0 24 48 72 96 120 144 168 192
Time (hrs)
Figure 11. Production Curve of Capreomycins
16
Purification of Capreomycins
After 4 days of incubation (30 °C, 250 rpm), the production of capreomycins
reached a maximum (Figure 11), and the fermentation broth was typically collected after
6 days by filtration to remove the mycelia. Activated charcoal was then added to the
filtrate to bind all of the organic materials. After filtration, the charcoal was washed with
water and 0.05 M HC1, and then eluted with acidic aqueous_ acetone. The resulting filtrate
was concentrated to a small volume and precipitated in acetone. After standing at 4 °C
overnight, the acetone was decanted and the remaining brown oily residue was dissolved
in 6 M HC1, precipitated in methanol, filtered and dried to afford the crude product.
The isolation of IA and 1B was carried out chromatographically on Amber lite CG-
50 by using a gradient (0.4-0.8 M) of ammonium acetate (pH 9.0) monitored by a dual
wavelength (254 and 280 nm) UV detector. A typical chromatogram is shown in Figure
12. Desalting was required next, and this step was first tested by using Sephadex G-10
17
column chromatography followed by recrystallization, but recrystallization gave poor
yields. The yield was later improved dramatically by simply desalting the mixture on a
DEAE-Sepharose CL-6B column (C1- form).
Synthesis
160
140 -
120 -
100 -
80
60
40
20 -
0
Unknown
ABS
Conc. (M)
IB
0
Fraction Number
Figure 12. Typical Chromatogram of Capreomycins
0.9
0.8
- 0.7
0- 0.6 0
80
0.5
0.4
DL- [2,3,3- 2H3]Dap (11a)
DL-Aspartic acid was used as the starting material to prepare DL-[2,3,3-2H3]Dap
(11a). Unlabeled aspartic acid was first converted to DL-[2,3,3-2H3]aspartic acid via an
18
exchange reaction in D20 catalyzed by pyridoxal and Al2(SO4)3.42 The resulting
product was then converted to DL- [2,3,3- 2H3]Dap by Schmidt reaction43 with fuming
H2SO4 and NaN3 (Scheme 5). The product contained deuterium at H-2 (98% enrichment)
and at H-3 (80% enrichment) as determined by 2H NMR spectroscopy.
Scheme 5
H H
11)
NH2
HOOC COOH
Fuming H2S041.NaN3
D20, NaOH
Al2(804)3, Pyridoxal
D DD NH2
H2N COOH
11a
HOOC
NH2
COOH
DL- [2- 2H]Dap (llb)
The first attempt to make DL-[2-2H}Dap (lib) via a cobalt complex was
unsuccessful. No DL-[2-2H]aspartic acid, which would have then been converted to lib
as described above, was obtained. A possible explanation is that deuteriated aspartic acid
could not be released from the aspartato-cobalt complex by chromatography. Both 1H
and 2H NMR spectra suggested the existence of the aspartato-cobalt complex. In the
second attempt, llb was prepared from DL-Dap by an acid-catalyzed exchange reaction
(Scheme 6).45 DL-Dap was mixed with D20 and 35% DC1, and heated at 125 °C in a
sealed pressure tube for 48 h. After removal of 1320 and DC1, the residue was redissolved
in a small volume of water and precipitated in absolute ethanol. The final product showed
a 97% deuterium enrichment at H-2 as determined by 2H NMR spectroscopy.
Scheme 6
H HH NH2
H2N COOH
D20, 35% DCI70-
H DH NH2
H2N COOH
19
Hydrolysis of Capreomycin and Purification of Capreomycidine (5)
The hydrolysis of capreomycin has been reported,1 and the capreomycidine (5)
thus obtained was purified by electrophoresis. In our work, crude capreomycins (1 g)
obtained by fermentation was hydrolyzed in 6 M HC1 at 125 °C for 48 h. The
hydrolysate was then passed through a Dowex 50W-X8 column and the column was
washed with water until the elutant pH was 6-7. Serine and alanine were removed during
this step. The column was then eluted with 0.3 M NH4OH, and ninhyrin-positive
fractions were combined and evaporated to a small volume. This residue was further
purified by chromatography on a Dowex 50W-X8 column using a gradient of
(NH4)2OAc buffer (pH 7-9) to give a 57% recovery of 5 based on a theoretical yield (160
mg, considering 60% of 1 and 2 in crude product) after desalting by DEAE-Sepharose
CL-6B chromatography.
The deuterium-hydrogen exchange reaction at the a-position of 5 was tried using
the same method as that of 11b, but it was unsuccessful. More than one position had
been labeled by 2H. Although it was not a desired product, the multiply deuteriated 5
could still be used as a precursor to test whether or not 5 is a free intermediate in the
biosynthesis of capreomycin.
20
Feeding Experiments
L-[U-14C]Serine, L-M-14C1Alanine and L-[U-14C]Cystine
Administration of L- [U- 14C]serine to fermentations of S. capreolus yielded a
2.1% incorporation for 1 and an 1.6% incorporation for 2, whereas administration of
either L- [U- 14C]alanine or L-[U-14C]cystine yielded less than 1% incorporation into
either 1 or 2. These results, shown in Table 1, indicated that serine is the most likely
precursor not only for the serine residue in 1, but also for the 2,3-diaminopropionate
(Dap, 11) residues of 1 and 2.
Table 1. Results of Radioactive Precursor Feedings
Precursor Specific Activity (dpm/mg)Based
% Incorporation
L-[U -14C1-serine IA 2.04 x 103 IA 2.05IB 1.58 x 103 IB 1.60
[U--[U-14C1-alanine IA5.29 x 102 IA 0.50
IB 9.29 x 102 IB 0.90
;-[U-14C]-cystine IA5.12 x 102 IA 0.46
IB 6.96 x 102 B3 0.631
DL-[1-13C]Serine (10a)
Cultures of S. capreolus were fed DL-[1-13C]serine (10a) to obtain positional
information on the origin of the residues that make up capreomycins. Figures 13a and
13b give the specific enrichments that were determined from the 13C NMR spectra of the
derived lb and 2b, normalized to the C-1 resonance of the natural abundance spectrum.
21
Enrichments (average 1.0%) at C-10, C-17 and C-34 clearly revealed that serine is a
primary precursor for the Dap residues and the .62,3-ureido-Dap (Uda) residue. The
dramatically higher enrichment at C-30 of lb compared to that of at other positions
confirmed that 10a was incorporated directly into the serine residue (Figure 13a). These
results suggest that the capreomycin pathway involves pre-peptide assembly modification
of serine to Dap, rather than post-peptide assembly modification of serine residues, since
approximately the same enrichment at all four carbonyl positions should result if serine
were incorporated directly into each. In contrast to lb, similar enrichments at all four
equivalent positions of 2b were observed. This result appeared to rule out the possibility
that capreomycin IA, 1, is derived by hydroxylation of capreomycin IB, 2. Minor
incorporation of the labeled serine into the guanido-group of the capreomycidine residue,
the ureido-group of the modified Dap residue and into the 13-lysine residue could be
rationalized by the metabolic pathways shown in Figure 14.
0.83%O
H2N ,,,,,3 4
H 1
30
1
0NH / HN 1 7 o
N 1 0 N
II I0) HC..1 NH
N NH2+
0.24%I
0.55%
0 NH2
N 21
1 NH2
0 LT:Zimi
NH2
Figure 13a. Enrichments over Natural Abundance Observed in lb
2
23
CH3-J%CO2-
Pyruvate
OH
H3N" CO2"
Serine
CO2
O 0
H2N CH2- CH2 - OH
Ethanolamine
CH3
H311. 6,02
Alanine
c02-
H3N' CO2
Aspartate
CHO
1I-13N* CO,-
Aspartyl Semialdehyde
H2N - 3: -
Carbamoyl Phosphate
1111
HN - NH2HN C: NH2
(CH2)3 CH
H3N* )1..' CO2" H3N* '11. 002-
Citrulline 0-U reido-dehydroalanine
H3N*. 2-
(C,
C0
(CH2)3
1-1314. "..% CO2-
Diaminopimelate
1NH2
(CH2)4
).H,N co,-a-Lysine
NH2
(CH2)3
H3N` CH
CH2.1CO2"
(3-Lysine
*NH2n
HN - C: NH2
(CH2)3
..1%.I-13N+ CO2
Arginine
-0 `NH3
NH
N oi *NH2H
Capreomycidine"
"c .13c
Figure 14. Proposed Metabolic Pathways for Incorporation of [l-13C1Serine into f3-
Lysine, Capreomycidine and 13-Ureido-dehydroalanine
24
DL-[3-13C]Alanine (12a)
DL-[3-13C]Alanine (12a) mixed with L- [U- 14C]alanine was fed next, and yielded
lc and 2c with the enrichments shown in Figures 15a and 15b. The 3.0% enrichment at
C-32 of 2c in comparison to significantly lower enrichments at other enriched sites was
consistent with the direct incorporation of 12a into 2c. No enrichment was observed at
Dap or Uda residues. Again, secondary incorporation was observed in the
capreomycidine, Uda and 13-lysine residues, and could be rationalized by the metabolic
pathways shown in Figure 16.
NH
NH H' HN
N NH2
NH2
Figure 15a. Enrichments over Natural Abundance Observed in is
25
NH
.0NH H° FIN 0
I HN N NH2
Figure 15b. Enrichments over Natural Abundance Observed in 2c
26
CH3
H3N+
Alanine
H2N- CH2
CH2
CH2.1
H3N+ CH
CH2
c02-(3-Lysine
?1-13
Off` CO2-
Pyruvate
CH3
SCoA
Acetyl CoA
"02C\CH3
d%* CO2
Oxaloacetate
TCA cycle
.02C% -02C% -02C%
CH2 CH2 CH21 1 1
CH2 CH2 CH2
041 0- OA SCoA 0 CO2"
Succinate Succinyl CoA a-Ketoglutarate
CH3 CH3
A ._0 uul 0#.' CO2
Pyruvate Pyruvate
pi
Gri +N H3
NH
N +NH2
Capreomycidine
,3c
CO2
2 2H2N-0:0-F:-
0_
Carbamoyl Phosphate
+NH2
HN-- C- NH2
CH2
CH2
CH2
143N+..°14 CO2
Arginine
02C%
oH2
CH2
H3N+-...k CO2
Glutamic Acid
NH2
CH2
CH2
CH2
H3N+'11. CO2
Omithine
Figure 16. Proposed Metabolic Pathway for [3-13C]Alanine Feeding
27
DL-[2,3,3-2H3]Serine (10b)
In order to determine the mechanism for conversion of serine to Dap, DL-[2,3,3-
2H3]serine (10b) was fed next and yielded ld and 2d. These were analyzed by 2H NMR
spectroscopy in 2H-depleted water, and FIDs were acquired at 298 K. Since the resonance
frequency for H-31 (84.84) overlapped with the small remaining solvent signal, a second
FID was acquired at 345 K with a pre-saturation pulse to diminish the residual water
resonance. The spectrum of 2d (Figure 17b) at 345 K clearly showed resonances for H-12
(88.07), H-19 (83.86 and 83.71), H-26 (81.82), H-32 (81.44) and H-37 (84.15 and
83.29). No deuterium was observed at H-18 (64.43) or H-35 (84.34). Retention of the (3-
hydrogens and loss of the a-hydrogen would be consistent with involvement of a
dehydroalanyl residue in the conversion to 11. The 2H NMR spectrum of ld (Figure 18b)
at 345 K was almost the same as that of 2d except for an additional resonance at H-31
(84.84), which again was consistent with direct incorporation of serine at serine residue in
IA and replacement of serine by alanine at the alanine residue in IB. The enrichment
appearing at H-26 (81.82) could also be rationalized by the metabolic pathway shown in
Figure 14.
b
a
H-19 HODH-32
28
f l 1 i 1 1 i i 1 1 1 T I -I -I 7 I
10 9 8 7 6 5 4 3 2 1 0PPM
Figure 17. NMR Spectra of 2d. (a). 400-MHz 1H NMR spectrum in D20 at 298 K. (b).
61.4-MHz 2H NMR spectrum in 2H-depleted H2O at 345 K showing deuterium
enrichment at H-12, H-19, H-26, H-32, and H-37; residual HOD signal (84.4) was
suppressed by inversion recovery water suppression pulse.
29
I I 1 1 I 1
10 9 8 7 6 5 4 3 2 1 0PPM
Figure 18. NMR Spectra of id. (a). 400-MHz 1H NMR spectrum in D20 at 298 K. (b).
61.4-MHz 2H NMR spectrum in 2H-depleted H2O at 345 K showing deuterium
enrichment at H-12, H-19, H-26, H-31, H-32 and H-37; residual HOD (M.4) was
suppressed by inversion recovery water suppression pulse.
30
DL-[2,3,3-2H3]Dap (11a)
Results from the serine feedings clearly indicated conversion of serine to 11 prior
to peptide assembly. To test this directly, DL- [2,3,3- 2H3]Dap was fed to S. capreolus, and
the derived metabolites, le and 2e, were analyzed by 2H NMR. Both spectra (Figure 19b
and Figure 20b) clearly show deuterium enrichment at H-12, H-19 and H-37. By
comparison with the spectra of id and 2d, deuterium enrichments at H-18 and H-35 were
also recognizable. No enrichment was observed for the serine or alanine residue of le or
2e, respectively. The approximate enrichments were determined by cutting and weighing
traces of the relevant signals from capreomycin and from the t-BuOH that had been
included as a chemical shift reference and for deuterium quantitation. This yielded an
enrichment of 25% at H-12, and an average of 14% for each of the other sites in 2e. The
values for le were 16% and 5%, respectively. 2,3-Diaminopropionic acid is clearly an
intermediate in the biosynthesis of capreomycin, and is thus formed prior to the formation
of the pentapeptide.
DL-[2-211]Dap (11b)
In order to more clearly define the fate of H-2 of 11a, DL- [2- 2H]Dap (11a) was
fed to S. capreolus. The 2H NMR spectra of if and 2f thus obtained (Figure 19c and
Figure 20c) exhibited the expected resonances (E4.3-84.5), but contained less deuterium
at H-18 and H-35 than expected from the previous feeding. However, we subsequently
determined that at these positions the capreomycins suffer slow exchange (ca. 25-50%,
see below) under the fermentation conditions. Absolute levels of apparent incorporation
were observed to vary between experiments more than would typically be observed just
from the vagaries of two different biological experiments.
b
aH-12
HOD
H-19'
H-37'
H-18
H-35
8 7
-6 5
PPM4 3
H-19
H-37
31
Figure 19. Partial NMR Spectra of le and lf. (a). 400-MHz 1H NMR spectrum of 1 in
D20 at 298 K. (b). 61.4-MHz 2H NMR spectrum of le in 2H-depleted H2O at 345 K
showing deuterium enrichment at H-12, H-19 and H-37; residual HOD (84.4) was
suppressed by inversion recovery water suppression pulse. (c). 61.4-MHz 2H NMR
spectrum of if in 2H-depleted H2O at 298 K showing deuterium enrichment at H-18 and
H-35.
b
a
9 8 7 6 5 4 3PPM
32
Figure 20. Partial NMR Spectra of 2e and 2f. (a). 400-MHz 1H NMR spectrum of 2 in
D20 at 298 K. (b). 61.4-MHz 2H NMR spectrum of 2e in 2H-depleted H2O at 345 K
showing deuterium enrichment at H-12, H-19 and H-37; residual HOD (54.4) was
suppressed by inversion recovery water suppression pulse. (c). 61.4-MHz 2H NMR
spectrum of 2f in 21I-depleted H2O at 298 K showing deuterium enrichment at H-18 and
H-35.
33
Exchange Reaction
Deuterium-hydrogen exchange under simulation of the fermentation conditions
was carried out by first using pure capreomycin IB, 2, in D20 at pH 8.0, 30 °C, 250 rpm
(in a shaker) for 7 days. The reaction mixture was then freeze-dried and desalted on a
DEAE-Sepharose CL-6B column. Both 1H and 2H NMR spectra of the sample showed a
similar (ca. 50%) deuterium enrichment each at 84.3-4.4 and 85.0, which correspond to
H-18 (and/or H-35) and H-2 in 2. Next, 36 mg of the resulting product was treated with
H2O and the above procedure was repeated. Approximately 20% of the deuterium was
exchanged back to protium. Another 20 mg of the initial enriched product was mixed
with unlabeled capreomycin IB to make a final deuterium enrichment of 5%,
approximately the expected level of enrichment in our feeding experiment. The sample
was dissolved in H2O and placed in a shaker under the same conditions for 4 days. The
product contained ca. 2% deuterium enrichment at these same positions. The latter result
was consistent with the former result in that ca. 50% deuterium-hydrogen exchange had
occurred under the fermentation conditions.
A summary of all of the feeding experiments is shown in Figure 21.
H2N
NH2
CO2H
11, H2 =H3 =H OH11a, H-2 =H-3 =D11b, H-2= D, H-3= H
H2N CO2H CH3
10a, C-1 =13C H2N)%4CO2H10b, H-2 = H-3 = D
HO2C NH2
NH2
N/L NH
8, H-2 = H-3 = H-5 = H8a, H-2 = H-3 = H-5 = D
H2N3/4,NH
NH
12a, C-3 = 130
H0 NH2
33 NN 21 23 25
0 HHN ° -111H
NH0
5N NHH
la, R=OH, H-3= H-5 =Dlb, R=OH, C-10 = C-17= C-30 =C-34 =13Clc, R =OHld, R=OH, H-12 = H-19= H-31 =H-32 =H-37 =Dle, R =OH, H-12= H-19= H-37 =Dlf, R=OH,H-18 =H-35 =D
2a, R=1-1, H-3 =H-5 =D
2b, R H, C-10 = 0-17 = C-30 = C-34 .13D2c, R = H, C-32 =13C
2d, R = H-12 = H-19 = H-37 = D
2e, R=H,H-12=H-19=H-37=D2f, R=H, H-18 =H-35 =D
34
Figure 21. Summary of Labeling Patterns of 1 and 2 From Feeding Experiments with
Precursor Amino Acids
35
Studies with Cell-free Extract from. S. capreolus
From feeding various labeled precursors to S. capreolus, it was clear that serine
was converted into Dap before a cyclic pentapeptide was formed, and arginine was
possibly incorporated into a linear peptide prior to the formation of capreomycidine.18
Serine and alanine also were directly incorporated into 1 and 2, respectively. Although
the mechanism of the peptide assembly leading to capreomycin was still unknown, based
on the results from studies of other peptide biosynthesis,48-51 it is reasonable to assume
that all the precursor amino acids were activated directly at their respective enzyme sites
before the formation of the cyclic pentapeptide. Therefore, studies at the cell-free level
were required in order to better understand the details of the biosynthesis of capreomycin.
Generally, compounds containing non-protein constituents in the polypeptide
chain may be derived from either modification of the chain or originate from non-
ribosomal systems,40 whereas protein antibiotics (containing only protein amino
acids)46,47 are usually formed via ribosomal mechanisms. In non-ribosomal systems, the
activation of constituent amino acids usually occurs as a result of either ATP
[32P]phosphate exchange,48 corresponding to a "non-thiotemplate" mechanism, or ATP -
[32P]pyrophosphate exchange, corresponding to a "thiotemplate" mechanism as observed
in other peptide antibiotics (Scheme 7),49-51 leading to the formation of an enzyme-
stabilized, non-covalently bound amino or hydroxyl acyladenylate. Upon addition of
[32P]pyrophosphate, the reaction is reversed and [32P]ATP is formed. Assays for
detecting the enzyme activity are usually carried out by using this protocol.
36
Scheme 7
EnzymeATP + substrate substrate-AMP + PP;
Thiotemplate mechanism
Or
EnzymeATP + substrate substrate-ADP + Pi
Non-thiotemplate mechanism
Preparation and Partial Purification of Cell-free Extract (CFE)
Initially, mycelia of S. capreolus were collected from a 5-day fermentation culture
on the basis of capreomycin accumulation (since this corresponded to maximal
capreomycin production). In similar cases involving activating enzymes from other
Streptomyces strains, protein from younger cells was used for enzymatic studies,50 -52
probably due to a higher concentration of enzyme at early stage of cell growth. In our
studies higher enzyme activity was detected when using S. capreolus cells harvested 48 h
after inoculation. Cells were collected by centrifugation, washed sequentially with
buffer, 1 M KC1, 0.8 M NaCl, and buffer. The cells could be stored at -80 0C for at least
two weeks after this step. Typically, one volume of mycelia was then resuspended in two
volumes of pre-cooled buffer, to which protease inhibitor phenylmethanesulfonyl fluoride
(PMSF) had been added, and disrupted by sonication. The CFE was obtained by
centrifugation. Polyethylenimine (PEI) cellulose was then added in order to remove
nucleic acids.53 In early experiments, the final concentration of PEI-cellulose was
usually brought to 0.3%. Subsequently, it was found that the amino acid activating
enzyme activity was increased if the final concentration of PEI-cellulose was reduced to
0.05% 0.1%. The precipitate from the PEI-cellulose procedure was removed by
37
centrifugation, and the resulting supernatant was treated with solid ammonium sulfate. In
one experiment, the solution was brought to 30% (no precipitation occurred at this step),
50% and 90% saturation with solid ammonium sulfate, respectively. In another
experiment, it was brought to 45%, 60% and 80% saturation. The pellets were collected
by centrifugation and desalted on PD-10 (Sephadex G-25) spin columns prior to
enzymatic assays.
Protein Concentration Determination
Protein concentrations were determined by the method of Bradford,54 using
bovine serum albumin (BSA) as the calibration standard. A standard calibration curve is
shown in Figure 22.
0 5
BSA (gg/mL)
10
Figure 22. Bradford Assay: BSA Standards
15
38
Based on the plot shown above, the protein concentration in the preparation of
enzyme for enzyme activity assay was calculated as shown below:
3.38 mg/mL protein in crude CFE.
2.79 mg/mL protein in the supernatant after PEI-cellulose precipitation, 32% of
total protein was removed in this step.
2.19 mg/mL protein in the supernatant of 0-45% (NH4)2SO4 precipitation, 16% of
total protein was removed by (NH4)2SO4.
1.38 mg/mL protein in the supernatant of 46-60% (NH4)2SO4 precipitation, 20%
more of total protein was removed by this step.
Only 0.27 mg/mL protein remained in the supernatant of 61-80% (NH4)2SO4
precipitation. 26% of total proteins were in the pellet.
The protein concentrations for the enzyme assays were as follows:
2.72 mg/mL in 0-45% (NH4)2SO4 pellet.
9.56 mg/mL in 46-60% (NH4)2SO4 pellet.
5.47 mg/mL in 61-80% (NH4)2SO4 pellet.
Enzyme Assay
Amino acid dependent ATP/phosphate and ATP/pyrophosphate exchange
reactions were carried out by a modification of the method of Lee and Lipmann.55 Assay
conditions for enzymatic synthesis of [32PJATP from either [32P]phosphate or
[32P]pyrophosphate were similar to those of Keller.49-51 In a total volume of 500 pL,
250 pL of enzyme, 10 mM phosphate buffer (pH 7.5), 2.5 mM ATP, 4.0 mM MgC12, 10
mM each of substrates, and 50 p.1_, [32P]phosphate or [32P]pyrophosphate (5 x 105 dpm)
were incubated for 30 min at 32 °C. The reaction was stopped by the addition of 10%
trifluoroacetic acid, and radioactive ATP was separated from [32P]phosphate or
[32P]pyrophosphate by adsorbing it onto activated charcoal. [32P]Phosphate and
[32P]pyrophosphate do not bind to charcoal and could be washed out with water, whereas
39
[32P]ATP remained bound to charcoal under these conditions. Radioactive ATP was
eluted from charcoal with either 6 M HC1 or 10% aqueous pyridine and then counted by
liquid scintillation counter. In several control experiments that have been performed in
order to find the ideal assay conditions, 10% aqueous pyridine demonstrated a higher
efficiency in removal of radioactivity from charcoal.
The details of a typical set of assays are shown below:
Vial # Substrate Enzyme
1 Dap Fresh
2 Arginine Fresh
3 Alanine Fresh
4 Serine Fresh
5 Aspartic Acid Fresh
6 Water Fresh
7 Dap Boiled (2 min)
8 Dap No Enzyme
Several early attempts to find the enzyme activity were hampered by a high
background exchange rate. This problem was later resolved by changing the brand of
charcoal (that purchased from either Spectrum Chemical Mfg. Corp. or Eastman showed
the higher background than that of Alpha) and decreasing the amount of charcoal used for
assays. The results from ATP /[32P]phosphate exchange assays with crude CFE suggested
that no exchange had occurred. All of the assays gave the same level of the radioactivity
(a few hundred dpm above the background) after ATP was eluted from charcoal, no
matter which substrate was used. The results from ATP/[32P]pyrophosphate exchange
40
with a variety of enzyme preparations indicated the presence of the amino acid activating
enzyme(s). The highest specific activity was found in the fraction of 0-45% (NH4)2SO4
saturation.
The results from one set of assays are listed in Table 2.
Table 2 . Specific Activity of t2P]ATP from ATP/[321113Pi Exchange Reaction.
(initial 2111)13i: 5.15 x 105 dpm)
l'Op,".12-, substrate
ay472,
/l
Enzymes
Dap Arg Ala Ser
Asp
(Non-precursorsubstrate)
NoSubstrate
(Endo-genous)
Dap
BoiledEnzyme
Dap
NOenzyme
Crude CFE 17515 22189 25089 25965 15858 32024 234 494
0-45% (NH4)2SO4
saturation pellet . 33206 31176 57029 50118 14235 25059 324 370
46-60% (Nli)2SO4
saturation pellet 6460 6418 7372 7565 7180 4753 510 636
61-80% (Nli)2SO4
saturation pellet 3729 3627 4424 3188 4812 4973 406 580
Apparently, the amino acid activating enzyme(s) precipitated at a concentration of
0-45% ammonium sulfate saturation and all of the precursor amino acids were activated
by the enzyme(s). The endogenous exchange rate still remained high in these assays with
the reasons unknown. However, it seemed to decrease after several purification steps. The
non-constituent amino acid of capreomycin, aspartic acid, was also activated in all the
assays performed. This might be accounted for that we chose improper substrate as one of
the controls. It is possible that the enzyme(s) may have a specific site that can activate
aspartic acid which is not utilized in the biosynthesis of capreomycin.
In order to demonstrate that the recovered 32P was associated with ATP produced
in assays with the precursor amino acids, but not with other controls, an autoradiographic
41
study was carried out. An autoradiogram (sensitive to as low as 1000 dpm) of a PEI-
cellulose TLC chromatogram of various exchange reaction mixtures containing different
substrates and enzymes showed the presence of [32P]ATP in assay mixtures with all the
amino acids as substrate, but not with boiled enzyme or incubations lacking enzyme. In
addition, the assay mixtures with the precipitate of 0-45% (NH4)2SO4 saturation showed
the strongest signals on the TLC plate. This was consistent with the results observed from
LSC shown in Table 2. TLC with commercially available [32P]ATP (from NEN) which
was used as a reference showed two spots. One of them had the same Rf value as that of
[32P]pyrophosphate (from NEN). This may be due to degradation of [32P]ATP to AMP
and [32P]pyrophosphate.
Conclusions and Areas for Further Study
We have established that one equivalent of serine and alanine is incorporated
directly into 1 and 2, respectively. However, three of the remaining residues of each core
cyclic pentapeptide are indirectly derived from serine via a 2,3-diaminopropionic acid
intermediate (11). The apparent 0-elimination/replacement mechanism revealed by
incorporation of the deuteriated amino acids is consistent with previous studies on the
biosynthesis of malonomicin56 in particular, and on pyridoxal phosphate-dependent (3-
replacement reactions57 in general. The clean incorporation of added 11 into the Dap and
Uda residues of 1 and 2 is consistent with a pre-peptide assembly modification of serine,
rather than a post-peptide assembly modification (e.g. simple dehydration) such as
presumably results in the dehydro-alanyl residues of berninamycin58 and nosiheptide.59
In contrast with the incorporation of 11, results from the previous study18 of
arginine incorporation are consistent with post-peptide-assembly modification to yield an
N-acyl-dehydroarginyl residue which then cyclizes. The fl-ureido-dehydroalanine residue
most reasonably also comes from post-peptide assembly modifications of a peptide Dap
42
residue. Figure 23 shows two possible peptide assembly modifications which need to be
further investigated in order to get a clearer picture of this biosynthetic process.
H2N Dap-Ala-Dap- Dap-A rgCOOH
H2NDap-Ala-Dap-Dap-CpdCOOH
H2NDap-Ser-Dap-Dap-ArgCOOH
H2N Dap-Ser-Dap-Dap-CpdCOOH
Dap-Ala-Dap-Dap-Cpd
NHCO
Capreomycin IB
Hydroxylation
Dap-Ser-Dap-Dap-Cpd
NHCO-1Capreomycin IA
Figure 23. The Apparent Manner of Capreomycin Assembly
The enzyme activity assay results revealed that the amino acid activating enzyme
activities for serine, arginine, alanine and Dap were present in the soluble supernatant of
the capreomycin producer cells, and that the biosynthesis of capreomycin appears to
involve activation of the component amino acids to produce substrate-AMP, as expected
for a non-ribosomal "thiotemplate" mechanism.40 This result illustrates another example
of the multi-functional/multi-enzyme complexes involved in the biosynthesis of microbial
peptide antibiotics. With some exceptions, most of these appear to involve a pantotheine
arm, presumably for covalent attachment of a growing peptide. Although a number of the
peptides studied have had an unusual amino acid starter unit, the succeeding residues
43
have generally been either intact protein amino acids or ones that have undergone
relatively simple modifications such as a-epimerization or N-methylation within the
peptide synthetase complex.38
The capreomycins present an opportunity to study the mechanisms and
relationships of extensive pre- and post-peptide assembly modification, in addition to the
peptide assembly process itself. Further characterizations of the enzyme(s) are clearly of
interest.
44
Experimental
General
S. capreolus A250 was obtained from the Eli Lilly Company. Bacillus subtilis
ATCC 6633 was obtained from Becton Dickinson Microbiology Systems. Carbon nuclear
magnetic resonance spectra were recorded at 75.5 MHz on a Bruker AC 300 (300 MHz
for 1H) spectrometer, and proton decoupled 2H NMR spectra were run unlocked at 61.4
MHz on a Bruker AM 400 (400 MHz for 1H) spectrometer. All NMR spectra were
obtained using 5 mm NMR tubes, and all samples were prepared in either D20 or 2H-
depleted H2O. The chemical shifts are reported in parts per million (ppm) relative to an
internal standard of t-BuOH (81H=1.28, 82H=1.28, 613C= 31.2).
Radioactivity measurements were carried out using a Beckman Model LS 7800
Liquid Scintillation Counter (LSC) with automatic quench correction and external
standardization to yield disintegrations per minute (dpm).
Amber lite CG-50 ion exchange chromatography was carried out at 4 °C. DEAE-
Sepharose CL-6B column chromatography was performed at room temperature.
Activated charcoal used for work-up of capreomycins was purchased from Spectrum
Chemical Mfg. Corp. Ion exchange resins were purchased from either Bio-Rad
Laboratories or Sigma Chemical Company.
Cell disruption was carried out by using a sonicator (Model W-225R, Heat
System-Ultrasonics, Inc.). Activated charcoal for enzyme activity assays was purchased
from Alpha Products (Johnson-Matthey). Protein assay kit was purchased from Bio-Rad
Laboratories. UV measurements were performed on an IBM 9420 UV-Visible
spectrophotometer.
L- [U- 14C]Serine, L- [U- 14C]alanine and L-[U-14C]cystine were purchased from
ICN. DL- [1- 13C]Serine and DL-[3-13C]alanine were purchased from Cambridge Isotope
45
Laboratories. Deuterium depleted water for 2H NMR was purchased from Aldrich
Chemical Company.
Fermentation of S. capreolus A250
Maintenance of S. capreolus A250
The agar medium (200 mL) contained dextrin (Difco, 2.0 g), yeast extract (Difco,
0.2 g), N-Z amine A (ICN Biochemicals, 0.4 g), beef extract (Difco, 0.2 g), CoC12.6H20
(0.002 g), BBL granulated agar (Bacton Dickinson Microbiology Systems, 4.0 g), and
deionized water. The pH was adjusted to 7.0 with 1 N NaOH prior to sterilization. Agar
in a petri plate was streaked with S. capreolus A250 spore suspension that had been
stored at -80 °C, and was then incubated at 28 °C for 7 days. An isolated colony was
suspended in 0.9% sterile NaC1, and a portion of the suspension (0.25 mL) was used to
inoculate another agar plate for confluent growth. After incubation at 28 °C for 27 days, a
sterile solution of 0.1% Tween-80 (10 mL) was added and the spores were suspended
with the aid of a sterile loop. The spore suspension was then transferred to a sterile
centrifuge bottle and centrifuged at 3000 rpm for 10 min. The pellet was resuspended in
20% sterile glycerol (12 mL) and 1.2 mL aliquots were distributed to cryovials and
preserved in an ultra-freezer (-80 °C) for long-term storage.
Production of Capreomycins
A portion of a spore suspension of S. capreolust8 (0.25 mL) was used to inoculate
a seed medium (60 mL) containing tryptone (Difco, 0.30 g), yeast extract (0.15 g),
glucose (0.90 g), MgSO4 -7H20 (0.18 g), FeSO4-7H20 (0.06 mg), CaC12.2H20 (0.015 g)
and deionized water. The pH was adjusted to 7.5 with 2% KOH before sterilization.
The flask (250 mL) was incubated at 30 °C, 250 rpm for 42 h, and a portion of the
seed culture (5% v/v) was used to inoculate production medium (200 mL in 1 L
46
Erlenmeyer flask) containing glucose (5.60 g), Bactopeptone (Difco, 3.00 g), N-Z amine
A (0.80 g), blackstrap molasses (purchased from local food co-op, 2.00 g), potato starch
(Sigma, 2.00 g), MgSO4.7H20 (1.00 g), CaCO3 (0.40 g), and hot tap water. The pH (6.5)
was unadjusted. The fermentation (200 mL) was incubated at 30 °C, 250 rpm for 6 days.
Isolation of Capreomycins IA and IB
The fermentation broth (200 mL) was harvested after 6 days of incubation by
filtration through Celite. The filter cake was washed with water (100 mL), and activated
charcoal (9.0 g) was added to the combined filtrate and washings. The mixture was
allowed to stand for 40 min at room temperature. The charcoal was filtered through
freshly prepared Celite and the filtrate was discarded. Successive washings of the filter
cake with water (60 mL) and 0.05 M HC1 (60 mL) were also discarded. The filter cake
was next eluted with acidic aqueous acetone (20% acetone, 80% 0.05 M HC1, v/v, 250
mL) in a period of 15 min. The acetone eluate was then concentrated in vacuo to a small
volume (ca. 5 mL) and transferred, with stirring, to acetone (80 mL) and the cloudy
mixture allowed to stand at 4 0C overnight. The light brown solution was decanted, and
the resulting reddish-brown oily residue was dissolved in 6 M HC1 (3 mL). The solution
was filtered into stirred Me0H (36 mL), and the precipitate was then collected by
filtration.
A column of Amber lite CG-50 (100-200 mesh, N'H4+ form) (1.0 x 42.5 cm) was
equilibrated in 0.4 M NH4OAc, buffered with NH4OH to pH 9.0 at 4 0C. The sample
(450 mg) was loaded onto the column in a small volume (3 mL) of the same buffer. The
column was eluted first with a gradient from 0.4 to 0.8 M NH40Ac (225 mL each in Bio-
Rad gradient former) and then isocratically with 0.8 M NH40Ac (300 mL). Fractions (8
mL each) were collected at 60 -min intervals and were monitored by an UV detector (254
and 280 nm) (Pharmacia). On the basis of the UV profile, fractions containing either IA
47
or IB were combined and freeze-dried. The acetate salt was chromatographed on a
column of DEAE Sepharose CL-6B (Sigma, 1.0 x 30 cm, form) eluted with water.
Bioassay
Sterile PA-7 agar (1.5% Bacto-agar and 0.5% Bacto-tryptone) was thawed and
then equilibrated at 55 °C in a water bath. Agar (20 mL) was dispensed into a petri plate,
allowed to solidify and another 6 mL of the same agar containing a diluted B. subtilis
ATCC 6633 saline suspension (0.027% final concentration) was then added and used for
the bioassay.
Sterile stainless steel wells (i.d. 5 mm, up to 6 per plate) were placed evenly on an
agar plate prepared as above, and the solution (100 [tL) to be assayed was placed in each
well. The plate was incubated at 37 °C for 24 h. The diameter of the inhibition zone was
then measured and used to calculate the concentration of the antibiotic from the standard
curve.
Synthesis
DL42,3,3-2H3]Dap (11a)
DL-Aspartic acid (2.66 g, 20 mmol), NaOH (1.60 g, 40 mmol), pyridoxal
hydrochloride (0.41 g, 2 mmol) and Al2(SO4)3.18H20 (0.33 g, 0.5 mmol) were washed
into a 50-mL flask with D20 (10 mL). The sample was vortexed vigorously and
lyophilized. Triethylamine (1.4 mL, 10 mmol) and D20 (20 mL) were added to the
sample, and the mixture heated at 50 °C for two days in an oil bath covered with
aluminum foil to minimize photodecomposition of the pyridoxal. The mixture was cooled
and the contents were washed into a 500-mL flask containing water (400 mL), sodium
oxalate (0.54 g, 4 mmol), 3-methylpyrazole (0.8 mL,10 mmol) and 1 M HC1 (35 mL, 35
48
mmol). The resultant mixture was then loaded onto a Dowex 50W-X8 column (H+ form,
100-200 mesh, 2.5 x 24 cm). After washing the column with water, the sample was
eluted with 0.1 M NH4OH. Ninhydrin positive fractions were combined, rotary
evaporated, redissolved in water, and re-evaporated to dryness. The residue was dissolved
in 25% DC1 (16 mL) and added to a flask which was then evacuated and sealed. After
heating for 8 days at 125 °C, the mixture was cooled, diluted with water, and evaporated
to dryness in vacuo . The sample was redissolved in water and taken to dryness to remove
excess HC1, yielding 0.80 g (30%) DL-[2,3,3-2H3] aspartic acid. mp: 301 °C dec. (lit-6°
>300 °C dec.); Integration of the peak at 8 4.09 and 8 2.95 from 2H NMR spectrum
showed the deuterations at H-2 (98%) and H-3 (97%); 1H NMR spectrum which was
obtained by using the same amount of sample used for 2H NMR study showed no
resonance at the corresponding chemical shift.
DL-[2,3,3-2H3]Aspartic acid (0.77 g, 5.66 mmol) was placed in a 250-mL 3-neck
flask and dissolved in 20% fuming sulfuric acid (5 mL) with cooling followed by the
addition of dry chloroform (15 mL). The mixture was heated to reflux, and sodium azide
(0.80 g, 12.32 mmol) was added in small amounts with vigorous stirring over 3 h. The
mixture was stirred for an additional 4 h, then cooled in an ice bath, and the chloroform
layer was poured off. The thick paste-like residue was poured onto crushed ice (ca. 30 g)
and diluted to 100 mL. The solution was then passed through a Dowex 50W-X8 column
(H+ form, 100-200 mesh, 2.5 x 11 cm) and washed first with 150 mL 0.5 M HC1, and
then with water to neutrality. The adsorbed lla was eluted with 10% ethanolic ammonia
(4.0 M), and the ninhydrin positive fractions were combined and evaporated to a small
volume (ca. 2 mL). Ethanol (95%, 20 mL) was added slowly and most of the llaHC1
precipitated. The precipitate was filtered, washed with 95% ethanol (10 mL) and dried.
The filtrate was concentrated to ca. 5 mL, and more 11a-HC1 was recovered. The
combined yield was 0.41 g (69%) as a white powder. mp: 234 °C (lit.61 236-237 °C);
49
2H NMR analysis indicated 98% deuteriation at H-2 and 80% at H-3. 1H NMR spectrum
showed only a very small peak at 8 3.5.
DL- [2- 2H]Dap (lib)
A solution of DL-diaminopropionic acid monohydrochloride (2.00 g, 19 mmol),
D20 (20 mL), and DC1 (35% in D20, 1 mL) was placed in a pressure tube, sealed and
heated in an oil bath at 125 0C for 24 h. After cooling, the solvent was removed in vacuo
and the resulting residue was dried under vacuum for 2 h. The residue was dissolved in
water (50 mL), concentrated to a small volume (ca. 3 mL), and precipitated in absolute
ethanol (100 mL). After standing at 4 0C for 4 h, the precipitate was collected by
filtration, washed with ethanol, and dried under vacuum to give 1.94 g (96%) of DL-[2-
21-11-Dap-HC1. mp: 234 0C; 1H NMR (D20, 300 MHz) 8 4.07 (t, < 0.03H, J=7.06 Hz), 8
3.50 (s, 2H); 2H NMR analysis indicated 97% deuteration at H-2.
Hydrolysis of Capreomycin and Purification of Capreomycidine (5)
Capreomycin (1.0 g, crude product from fermentation) was refluxed at 125 0C in
20 mL of 6 M HC1 over 48 h. The solvent was removed by evaporation under vacuum,
and the residue treated with water (100 mL) and again evaporated to dryness. The residue
was then taken up in 75 mL of water and passed over a Dowex 50W-X8 column (200-
400 mesh, H+ form, 2.5 x 12 cm).
The column was eluted with water until pH 6-7, and then eluted with 0.3 M
NH4OH until no ninhydrin-positive materials came off the column. All of the ninhydrin
positive fractions were combined and concentrated. The residue was then applied to a
Dowex 50W-X8 column (200-400 mesh, H+ form, 2.5 x 12 cm) and eluted with a pH
gradient (pH 7-9) of 0.4 M NH4OAc. Elution progress was followed by silica gel TLC
using the system CHC13 / Me0H / NH4OH / H2O (1:4:2:1). Fractions 5-20 (8-10 mL per
50
fraction) contained Dap and 21-25 contained a mixture of Dap and capreomycidine.
Fractions 26-30 contained capreomycidine exclusively while fractions 38-48 contained
another ninhydrin-positive material which was not identified.
Fractions 26-30 were combined and freeze-dried to give 110 mg of residue which
was then charged to a DEAE-Sepharose CL-6B column (2.5 x 15 cm, form). Elution
was carried out using water. Fraction volumes were 7-8 mL and the ninhydrin-positive
material appeared in fractions 3-9. These fractions were combined and freeze-dried to
yield 91.2 mg (57% based on theoretical yield) of pure capreomycidine. TLC: (CHC13-
Me0H-28%NH4OH-H20, 1:4:2:1, on silica gel, 0.25% ethanolic ninhydrin spray) Rf
0.46; 1H NMR (D20, 300 MHz) 8 4.13 (m, 1H), 8 3.95 (d, 1H, J=5.3 Hz), 8 3.47 (m,
2H), 8 2.20 (m, 1H), 6 2.02 (m, 1H); 13C NMR (D20, 75.5 MHz) 8 172.45, 155.91,
58.10, 50.23, 37.75, 23.43.
Feedings
Feeding Protocol
Labeled compounds, except DL- [2- 2H]Dap, were pulse-fed in thirds to production
cultures at 12, 32 and 56 h after inoculation with a seed culture. DL-[2-2H]Dap was fed in
four pulses at 12, 32, 56 and 80 h after inoculation.
Generally, a small volume (2-3 gL) of a stock solution of radioactive compound
was diluted with deionized water in a 10 mL volumetric flask. Duplicate samples of 50
IL each were diluted to 5 mL in volumetric flasks. Duplicate samples of 100 gL each
were taken for scintillation counting and the remainder was used for the feeding. The
stable isotopically labeled compounds were weighed out, and dissolved in 10 mL water
for the feeding. Each precursor solution was slowly introduced into the production broth
at the appropriate time using a syringe fitted with a MSI membrane filter (size 0.22 gm).
51
Calculations in Feeding Experiments
Various calculations were performed to determine the % incorporation, %
enrichment and specific activity based on the amount of precursor fed, bioassay and
radioactivity data. The formulae used for these calculations are given below:
Specific activity isolated =radioactivity (dpm) in metabolite isolated
mmole of metabolite isolated
specific activity isolated x mmole of metabolite produced*% Incorporation = x 100
total radioactivity (dpm) in precursor fed
* based on bioassay
gmole of 211 observed by NMR integration% 2H Enrichment = x 100
gmole of product used for 2H NMR
% 13C Enrichment =II Jr
)1r If
where Il = The absolute intensity of the signal to be measured in thelabeled sample
= The absolute intensity of the same signal (same chemicalshift) as above in the natural abundance sample
Ir = The absolute intensity of the reference signal which should notbe enriched in the labeled sample
= The absolute intensity of the same reference signal in thenatural abundance sample
L-[U-14C]Serine, L- [U- 14C]Alanine and LtU-14C]Cystine
A single seed culture was used to inoculate three production broths. Aqueous
solutions (10 mL each) of L- [U- 14Clserine (10.73 pCi), L- [U- 14C]alanine (11.02 pCi)
and L- [U- 14C]cystine (11.98 pCi) were each fed in three equal portions to separate 200 -
mL production cultures at 12, 32, and 56 h after inoculation. After a total of 6 days,
52
bioassay of the fermentations indicated ca. 480 mg of total capreomycins for each culture.
Work-up and purification yielded 113.5 mg of IA (specific activity 1.36 x 106
dpm/mmol, 2.1% incorporation based on one-half of the bioassay mass) and 177.0 mg of
lB (1.56 x 106 dpm/mmol, 1.6% incorporation based on one-half of the bioassay mass)
from the L- [U- 14C]serine feeding. Similarly, 78.5 mg of IA (3.54 x 105 dpm/mmol, 0.5%
incorporation) and 70.8 mg of 1B (6.07 x 105 dpm/mmol, 0.9% incorporation) from the L-
[U -14C]alanine feeding, and 133.7 mg of IA (3.42 x 105 dpm/mmol, 0.5% incorporation)
and 107.2 mg of IB (4.54 x 105 dpm/mmol, 0.6% incorporation) from the L-[U-
14c] cystine feeding were obtained.
DL- [l- 13C]Serine (10a)
DL- [1- 13C]Serine (10a, 350 mg, 3.30 mmol) was fed in thirds to a 200-mL
production culture at 12, 32, and 56 h after inoculation with a seed culture. Bioassay
indicated 490 mg total capreomycins. This was worked up as usual, and 340 mg of crude
product was obtained after methanol precipitation. Only the center fractions from each
peak obtained from the Amberlite CG-50 chromatography were combined, and pure lb
(79.9 mg) and 2b (45.4 mg) were collected after the final desalting with the DEAE
Sepharose CL-6B column. A portion of each (40.0 mg) and t-BuOH (25 ilL) were
dissolved in D20 (0.5 mL), and the 13C NMR spectra (75.5 MHz) obtained with sweep
width 18,518 Hz, 64 K data points, 300 pulse width, 1.769 sec aquisition time, and
36,378 scans (lb) or 25,600 scans (2b). Each FID was transformed with 3.0 Hz line
broadening.
DL-[3-13CJAlanine (12a)
A mixture of DL-[3-13C]alanine (12a, 445 mg, 4.94 mmol) and L- [U- 14C]alanine
(9.52 liCi) in water (10 mL) was pulse-fed to a 200-mL production culture as described
above. After work-up and purification, the center fractions yielded 74.7 mg pure is (1.59
53
x 105 dpm/mmol, 0.3% incorporation) and 118.5 mg pure 2c (3.35 x 105 dpm/mmol,
0.7% incorporation). The 13C NMR spectra were acquired with 22,050 scans for lc and
18,450 scans for 2c.
DL-[2,3,3-2H3]Serine (10b)
A mixture of D L-[2,3,3-2H3]serine (10b, 440 mg, 4.15 mmol) and L-[U-
14C]serine (8.06 pCi) in water (10 mL) were pulse-fed to a 200-mL production broth as
described above. After work-up and purification, 152.2 mg pure 1 d (2.63 x 105
dpm/mmol, 0.8% incorporation) and 135.3 mg pure 2d (4.57 x 105 dpm/mmol, 1.5%
incorporation) were obtained. Samples of each (20.0 mg ld and 31.7 mg 2d) in
deuterium-depleted water (0.5 mL) plus t-BuOH (25 !IL) were analyzed by 2H NMR
(61.4 MHz) at 298 K and 345 K, as previously described.18 36,860 scans at 298 K and
14,679 scans at 345 K were acquired for ld, and 54,180 scans at 298 K and 12,288 scans
at 345 K for 2d.
DL-[2,3,3-2H3]Dap (11a)
DL-[2,3,3-2H3]Diaminopropionic acid (11a, 410 mg, 3.94 mmol) in water (10
mL) was pulse-fed to a 200-mL production broth as described above. After work-up and
purification, 47.3 mg pure le and 32.6 mg pure 2e were obtained. Each sample in
deuterium-depleted water (0.5 mL) plus t-BuOH (25 "IL) was analyzed by 2H NMR (61.4
MHz) at 298 K and 330 K, is previously described.18 1,892 scans at 298 K and 11,264
scans at 330 K were acquired for le, and 2,183 scans at 298 K and 13,312 scans at 330 K
for 2e.
DL-[2-211]Dap (11b)
DL- [2- 2H]Diaminopropionic acid (11b, 800 mg, 7.70 mmol in 30 mL water, pH
was adjusted to 7 with Na2CO3 prior to feeding) was fed in four equal portions to a 200-
54
mL production broth as described above. The work-up was carried out without the
methanol precipitation step. After purification with Amber lite CG-50 column (NH4+
form, 100-200 mesh, 1.0 x 42.5 cm), 62.3 mg if and 54.0 mg 2f were obtained. Desalting
step was omitted in this experiment. Samples of each (26.6 mg if and 54.0 mg 2f) in
deuterium-depleted water (0.5 mL) plus t-BuOH (50 pl.) were analyzed by 2FI NMR
(61.4 MHz) at 298 K; 36,115 scans were acquired for lf, and 27,500 scans for 2f.
Exchange Reaction of Capreomycin 11B under the Fermentation Conditions
Pure unlabeled capreomycin B3 (100 mg) was dissolved in 10 mL of D20 and
adjusted pH to 8.0 by addition of sodium carbonate. The mixture was then placed in a
shaker and incubated for 7 days at 30 °C, 250 rpm. After removal of the solvent, the
residues were redissolved in a small volume of water and passed through a DEAE-
Sepharose CL-6B column (2.5 x 15 cm, Cl- form), eluting with water. The ninhydrin-
positive fractions were combined and freeze-dried to give 92 mg of product. A portion of
this product (17.7 mg) was used for NMR studies. Both 1H and 2H NMR analysis showed
ca. 46% deuterium enrichment at 84.3-4.4 (H-18 or H-35) and 85.0 (H-2), respectively.
Another portion of the deuteriated product (36 mg) was treated with H2O by the
same procedure as described above to see if any deuterium could be exchanged with the
protons in the solvent. After 4 days, ca. 28% deuterium enrichment remained. This
suggested ca. 40% exchange. In the third experiment, the deuteriated product (20 mg)
was mixed with unlabeled capreomycin IB (180 mg), dissolved in 20 mL of H2O and
adjusted pH to 8.0 with sodium carbonate. 2H NMR at this stage showed ca. 5%
deuterium enrichment at the corresponding positions. The mixture was then shaken under
the same conditions as described above for 4 days. After work-up and desalting steps, 70
mg of sample was used for 2H NMR analysis, and this indicated ca. 2% deuterium
enrichment at both 84.4 and 85.0.
55
Cell-free Extract from S. capreolus
Preparation and Partial Purification of Cell-free Extract from S. capreolus
Buffer A: 10 mM KH2PO4, 10 mM K2HPO4, 5 mM dithiothreitol (DTT) and
15% glycerol (v/v), pH 7.5.
Buffer B: buffer A + 1mM EDTA.
All operations were carried out at 4 °C.
Step 1: Cells from 200 mL of a 48 h fermentation of S. capreolus were harvested
by centrifugation (4 °C, 13800 g, 20 min), and then washed (200 mL each) sequentially
with buffer A, 1.0 M KC1, 0.8 M NaC1, and buffer A. After each wash the cells were
centrifuged as above. The washed cells (22 g, wet wt.) were then resuspended in buffer B
(40 mL), and phenylmethanesulfonyl fluoride (PMSF) (16 mg in 1 mL isopropanol) was
subsequently added. The cells were then disrupted by sonication (power level 7 with
micro-tip, 80% duty cycle, pulsed for 6 x 20 s in total 12 min). Cell debris was removed
by centrifugation (4 °C, 13400 g, 20 min), and the supernatant was decanted to afford a
crude cell-free extract (CFE, 45 mL).
Step 2: Polyethylenimine (PEI) cellulose (Sigma, 50% in water, w/w) was added
to the CFE to a final concentration of 0.05%. After gentle stirring for 20 min, the cloudy
solution was centrifuged (4 °C, 13400 g, 20 min) and the precipitate was discarded.
Step 3: The supernatant was brought to 45% saturation by addition of solid
ammonium sulfate (Sigma, Grade III). The suspension was stirred for 30 min, and the
precipitate was collected by centrifugation (4 °C, 13400 g, 20 min) and stored at -80 °C.
The resulting supernatant was brought to 60% saturation with solid ammonium sulfate
and was stirred for additional 30 min. The pellet was obtained by centrifugation (4 °C,
13400 g, 20 min) and stored at -80 °C. The 60% (NH4)2SO4 saturated supernatant was
further brought to 80% saturation with solid ammonium sulfate and was stirred for 30
56
min. The pellet was collected by centrifugation (4 °C, 13400 g, 20 min) and stored at -80
°C.
Step 4: Buffer B (2 mL) was added to each of the three pellets which had been
stored at -80 °C for 12 h, and the dissolved pellets were loaded onto PD-10 spin-columns
packed with ca. 5 mL of Sephadex G-25 (pre-equilibrated in buffer B). The elutants were
transferred to three vials. The columns were washed once with 1 mL of buffer B. The
washings were combined with the corresponding elutants and used for enzyme activity
assays.
Protein Concentration Determination
Samples to be assayed were prepared as follows: to an 1.5-mL microfuge tube,
200 1.1.1, of the dye reagent and 800 RI, of protein solution (in most cases, a diluted
solution) were mixed and vortexed. After standed for 5 min, the samples were measured
at 595 nm by an UV-spectrophotometer individually, and protein concentrations were
determined by a standard calibration curve prepared from BSA (Figure 22).
Enzyme Activity Assays
The amino acid activating enzyme activity was assayed using an ATP -
[32P]pyrophosphate exchange reaction. The assay mixture contained 10 mM phosphate
buffer, pH 7.5, 2.5 mM ATP, 4.0 mM MgC12, 5.0 mM dithiothreitol (DTT), 1.0 mM
EDTA, 10 mM substrate, 50 1.11. [32P]pyrophosphate (5.15 x 105 dpm), and 250 pi.
enzyme solution, and incubated at 32 °C for 30 mM.
Trifluoroacetic acid (100 pl. of 10% ) was added to quench the exchange reaction
and the proteins were removed by centrifugation (7000 rpm, 2 min). The supernatant was
transferred to new tubes followed by addition of 50 µL of charcoal suspension (3% in 0.1
M Na4P207). The mixture was then vortexed, centrifuged, and the supernatant was
removed. The remaining charcoal residue was washed three times with water (3 x 800
57
!IL) and centrifuged each time. Finally, 100 pL of 10% aqueous pyridine was added to
the charcoal, and the mixture was vortexed and centrifuged at 7000 rpm for 4 min. A
portion of this solution (50 ilL) was used for LSC.
The rest of the solution was used for autoradiographic study. A PEI-cellulose TLC
plate was spotted with ATP first and then with the solution (50 gL), and developed in 0.5
M LiC1 solution. The air-dried plate was then exposed to a Kodak X-ray film in a light-
resistant cassette. The cassette was placed at -80 0C for 12 h before development of the
film.
58
Bibliography
1. Herr Jr., E. B.; Haney, M. E.; Pittenger, G. E.; Higgens, C. E.: "Isolation and
Characterization of a New Peptide Antibiotic." Proc. Indiana Acad. Sci. 1960,
69, 134.
2. Stark, W. M.; Higgens, C. E.; Wolfe, R. N.; Hoehn, M. M.; McGuire, J. M.:
"Capreomycin, A New Antimycobacterial Agent Produced by Streptomyces
capreolus sp. n." Antimicrob. Agents and Chemother. 1962, 596-606.
3. Stark, W. M.; Boeck, L. D.: "Biosynthesis of Capreomycin I in a Chemically
Defined Medium." Antimicrob. Agents and Chemother. 1965, 157-164.
4. Bycroft, B. W.; Cameron, D.; Hassanaliu-Walji, A.; Johnson, A. W.; Webb, T.:
"The Total Structure of Viomycin, a Tuberculostatic Peptide Antibiotic."
Experientia 1971, 27, 501-503.
5. Martindale, W. The Extra Pharmacopoeia; Blacow, N. W., Ed.; Hazel! Watson
and Viney Ltd. Great Britain, 1972, pp. 1585-1586.
6. Herr Jr., E. B.; Redstone, M. 0.: "Chemical and Physical Characterization of
Capreomycin." Annal. N.Y. Acad. Sci. 1966, 135, 940-946.
7. Shiba, T.; Nomoto, S.; Wakamiya, T.: "The Chemical Structure of Capreomycin."
Experientia 1976, 32, 1109-1111.
8. Nomoto, S.; Teshirna, T.; Wakamiya, T.; Shiba, T.: "The Revised Structure of
Capreomycin." J. Antibiot. 1977, 30, 955-959.
9. Shiba, T.; Nomoto, S.; Teshima, T.; Wakamiya, T.: "Revised Structure and Total
Synthesis of Capreomycin." Tetrahedron Lett. 1976, 3907-3910.
59
10. Shiba, T.: "Structure and Synthesis of Capreomycin." in Bioactive Peptides
Produced by Microorganisms; Umezawa, H.; Takita, T.; Shiba, T., Eds.; Halsted
Press: New York, 1978; pp. 84-85.
11. Wakamiya, T.; Shiba, T.; Kaneko, T.: "Chemical Studies on Tuberactinomycin. I.
The Structure of Tuberactidine, Guanidino Amino Acid Component." Tetrahedron
Lett. 1970, 3497-3500.
12. Minott, D. A.; Gould, S. J. Unpublished results.
13. Nagata, A.; Ando, T.; Izumi, R.; Sakakibara, H.; Take, T.; Hazano, K.; Abe, J.:
"Studies on Tuberactinomycin (Tuberactin), A New Antibiotic. I. Taxonomy of
Producing Strain, Isolation and Characterization." J. Antibiot. 1968, 21, 681-687.
14. Izumi, R.; Noda, T.; Ando, A.; Teruo, T.; Nagata, A.: "Studies on
Tuberactinomycin. Ill. Isolation and Characterization of Two Minor Components,
Tuberactinomycin B and Tuberactinomycin 0." J. Antibiot. 1972, 25, 201-207.
15. Ando, T.; Matsuura, K.; Izumi, R.; Noda, T.; Take, T.; Nagata, A.; Abe, J.: "Studies
on Tuberactinomycin. II. Isolation and Properties of Tuberactinomycin N, A New
Tuberactinomycin Group Antibiotic." J. Antibiot. 1971, 24, 680-686.
16. McGahren, W. J.; Morton, G. 0.; Kunstamann, M. P.; Ellestad, G. A.: "Carbon-13
Nuclear Magnetic Resonance Studies on a New Antitubercular Peptide Antibiotic
LL-BM54713." J. Org. Chem. 1977, 42, 1282-1286.
17. Hamakawa, H.; Ohno, H.; Shimizu, T.: "General Pharmacological Studies of
Tuberactinomycin N, A New Antibiotic Drug." Oyo Yakuri (Japan), 1974, 8, 817-
833, C.A. 82,1975, 51530y.
18. Gould, S. J.; Minott, D. A.: "Biosynthesis of Capreomycin. 1. Incorporation of
Arginine." J. Org. Chem. 1992, 57, 5214-5217.
19. Shiba, T.; Ukita, T.; Mizuno, K.; Teshima, T.; Wakamiya, T.: "Total Synthesis of L-
Capreomycidine." Tetrahedron Lett. 1977, 2681-2684.
60
20. Bycroft, B. W.: "Structural Relationships in Microbial Peptides." Nature 1969,
224, 595-597.
21. Bycroft, B. W.; King, T. J.: "Crystal Structure of Streptolidine, a Guanidine-
containing Amino-acid." J. Chem. Soc., Chem. Commun. 1972, 652-654.
22. Khokhlov, A. S.: "Streptothricins and Related Antibiotics." J. Chromatogr. Libr.
1978,15, 617-713.
23. Borders, D. B.: "Ion-exchange Chromatography of Streptothricin-like Antibiotics."
in Methods in Enzymology; Hash, J. H., Ed.; Academic Press: New York, 1975; pp.
256-263.
24. Waksman, S. A.; Woodruff, H. B.: "Streptothricin, A New Selective Bacteriostatic
and Bactericidal Agent Particularly Active Against Gram-negative Bacteria." Proc.
Soc. Exp. Biol. Med. 1942, 49, 207-210.
25. Van Tamelen, E. E.; Dyer, J. R.; Whalen, H. A.; Carter, H. E.; Whitfield, G. B., Jr.:
"Constitution of the Streptolin-Streptothricin Group of Streptomyces Antibiotics."
J. Am. Chem. Soc. 1961, 83, 4295-4296.
26. Gould, S. J.; Martinkus, K. J.; Tann, C.-H.: "Biosynthesis of Streptothricin F. 1.
Observing The Interaction of Primary and Secondary Metabolism With [1,2-
13C2] Acetate." J. Am. Chem. Soc. 1981, 103, 2871-2874.
27. Gould, S. J.; Martinkus, K. J.; Tann, C.-H.: "Studies of Nitrogen Metabolism Using
13C NMR Spectroscopy. 2. Incorporation of L- [guanido -13C, 15N2]Arginine and
DL-[guanido-13C, 2-15N]Arginine into Streptothricin F." J. Am. Chem. Soc. 1981,
103, 4639-4640.
28. Martinkus, K. J.; Tann, C.-H.; Gould, S. J.: "The Biosynthesis of The Streptolidine
Moiety in Streptothricin F." Tetrahedron 1983, 39, 3493-3505.
29. Thiruvengadam, T. K.; Gould, S. J.; Aberhart, D. J.; Lin, H.-J.: "Biosynthesis of
Streptothricin F. 5. Formation of 13-Lysine by Streptomyces L-1689-23." J. Am.
Chem. Soc. 1983, /05, 5470-5476.
61
30. Paliniswamy, V. A.; Gould, S. J.: "Biosynthesis of Streptothricin F. Part 6.
Formation and Intermediacy of D-Glucosamine in Streptomyces L-1689-23." J.
Chem. Soc. Perkin Trans 1 1988, 2283-2286.
31. Gould, S. J.; Lee, J.; Wityak, J.: "Biosynthesis of Streptothricin F. 7. The Fate of the
Arginine Hydrogens." Bioorg. Chem. 1991,19, 333-350.
32. Batelaan, J. G.; Barnick, J. W. F. K.; van der Baan, J. L.; Bickelhaupt, F.: "The
Structure of The Antibiotic K16. I. The Dipeptide Side Chain." Tetrahedron Lett.
1972, 3103-3107.
33. Rao, S. L. N.; Adiga, P. R.; Sharma, P. S.: "The Isolation and Characterization of (3-
N-Oxalyl-L-a,f3-Diaminopropionic Acid: A Neurotoxin from the Seeds of Lathyrus
sativus." Biochemistry 1964, 3, 432-436.
34. Harrison, F. L.; Nunn, P. B.; Hill, R. R.: "Synthesis of a-and 13-N-Oxalyl-L-oc,13-
Diaminopropionic Acids and Their Isolation from Seeds of Lathyrus sativus."
Phytochemistry 1977,16, 1211-1215.
35. Nomoto, S.; Shiba, T.: "Syntheses of Capreomycin Analogs in Relation to Their
Antibacterial Activities." Bull. Chem. Soc. Jpn. 1979, 52, 1709-1715.
36. Takita, T.; Muraoka, Y.; Nakatani, T.; Fujii, A.; litaka, Y.; Umezawa, H.:
"Chemistry of Bleomycin. XXI. Metal-complex of Bleomycin and Its Implication
for the Mechanism of Bleomycin Action." J. Antibiot. 1978, 31, 1073-1077.
37. van der Baan, J. L.; Barnick, J. W. F. K.; Bickelhaupt, F.: "Biosynthesis of L-2,3-
Diaminopropanoic Acid." J. Chem. Soc. Perkin Trans 1 1984, 2809-2813.
38. Kleinkauf, H.; von Dohren, H.: "Bioactive Peptides Resent Advances and Trends."
in Biochemistry of Peptide Antibiotics; Kleinkauf, H.; von Dohren, H., Eds.; Walter
de Gruyter and Co.: New York, 1990; pp.1-31.
39. Kleinkauf, H.; von Dohren, H.: "Nonribosomal Biosynthesis of Peptide
Antibiotics." Eur. J. Biochem. 1990, /92, 1-15.
62
40. Kleinkauf, H.; von Dohren, H.: "Biosynthesis of Peptide Antibiotics." Ann. Rev.
Microbiol. 1987, 41, 259-289.
41. Boeck, L. D.: "Development of a Chemically Defined Medium For Biosynthesis of
Capreomycin by Streptomyces capreolus." Butler Univ. Bot. Studies 1964, 14, 1-
20.
42. Richards, F. M.; Le Master, D. M.: "A General Procedure for the Preparation of a,(3-
Labeled Amino Acids." J. Labelled Comp. Radiopharm. 1982, 19, 639-646.
43. Rao, S. L. N.: "Chemical Synthesis of NO-Oxalyl-L-a,I3-Diaminopropionic Acid
and Optical Specificity in Its Neurotoxic Action." Biochemistry 1975,14, 5218-
5221.
44. Keyes, W. E.; Legg, J. I.: "Deuteration at the a-C of L-Aspartic Acid via the
Template Action of A Dissymmetric Cobalt (III) Complex." J. Am. Chem. Soc.
1973, 95, 3431-3432.
45. Martinkus, K. J. Ph. D Thesis, 1982, pp. 186. (University of Connecticut).
46. Berdy, J. in Handbook of Antibiotic Compounds. Vol. IV, Part 1&2, 1980, Boca
Raton, Fla: CRC, pp. 558, pp. 356.
47. Miyashiro, S.; Udaka, S.: "Screening and Some Properties of New Macromolecular
Peptide Antibiotics." J. Antibiot. 1982, 35, 1319-1325.
48. Ghosh, S. K.; Mukhopadhyay, N. K.; Majumder, S.; Bose, S. K.: "Fractionation of
the Mycobacillin-synthesizing Enzyme System." Biochem. J. 1983, 215, 539-543.
49. Zocher, R.; Nihira, T.; Paul, E.; Madry, N.; Peeters, H.; Kleinkauf, H.; Keller, U.:
"Biosynthesis of Cyclosporin A: Partial Purification and Properties of a
Multifunctional Enzyme from Tolypocladium inflatum." Biochemistry 1986, 25,
550-553.
63
50. Glund, K.; Schlumbohm, W.; Bapat, M.; Keller, U.: "Biosynthesis of Quinoxaline
Antibiotics: Purification and Characterization of the Quinoxaline-2-carboxylic Acid
Activating Enzyme from Streptomyces triostinicus." Biochemistry 1990, 29, 3522-
3527.
51. Keller, U.; Kleinkauf, H.; Zocher, R.: "4- Methyl -3- hydroxyanthranilic Acid
Activating Enzyme from Actinomycin-producing Streptomyces chrysomallus."
Biochemistry 1984, 23, 1479-1484.
52. Nielsen, J. B.; Hsu, M-J.; Byrne, K. M.; Kaplan, L.: "Biosynthesis of the
Immunosuppressant Immunomycin: The Enzymology of Pipecolate Incorporation."
Biochemistry 1991, 30, 5789-5796.
53. Shibata, T.; Cunningham, R. P.; Radding, C. M.: "Homologous Pairing in Genetic
Recombination: Purification and Characterization of Escherichia coli rec A
Protein." J. Biol. Chem. 1981, 256, 7557-7564.
54. Bradford, M. M.: "A Rapid and Sensitive Method for the Quantitation of
Microgram Quantities of Protein Utilizing the Principle of Protein-Dye Binding."
Anal. Biochem. 1976, 72, 248-254.
55. Lee, S.; Lipmann, F.: "Tyrocidine Synthetase System." Methods Enzymol. 1975,
43, 585-602.
56. Schipper, D.; van der Boon, J. L.; Bickelhaupt, F.: "Biosynthesis of Malonomicin.
Part 1. 13C Nuclear Magnetic Resonance Spectrum and Feeding Experiments with
13C-Labeled Precursors." J. Chem. Soc. Perkin Trans 1 1979, 2017-2022.
57. Floss, H. G.; Vederas, J. C.: "Stereochemistry of Pyridoxal Phosphate-catalyzed
Reactions." in Stereochemistry; Tamm, C., Ed.; Elsevier: New York, 1982; pp.161-
199.
58. Pearce, C. J.; Rinehart, Jr, K. L.: "Berninamycin Biosynthesis. 1. Origin of the
Dehydroalanine Residues." J. Am. Chem. Soc. 1979, 99, 5069-5070.
64
59. Houck, D. R.; Chen, L. C.; Keller, P. J.; Beale, J. M.; Floss, H. G.: "Biosynthesis of
the Modified Peptide Antibiotic Nosiheptide in Streptomyces actuosus." J. Am.
Chem. Soc. 1988, 110, 5800-5806.
60. Budavari, S.; et al. Ed. Merck Index 11th Edition, 1989, Merck & CO., Inc.,
Rahway, N. J. U.S.A., pp. 862.
61. Budavari, S.; et al. Ed. Merck Index 11th Edition, 1989, Merck & CO., Inc.,
Rahway, N. J. U.S.A., pp. 2962.