dissertation submittedto the faculty ofthe ...€¦ · l-lysine-u-mc was incorporated into both the...
TRANSCRIPT
STUDIES IN THE BIOSYNTHESIS OF VIRGINIAMYCIN S1
4Josephine W. Reed
i
Dissertation submitted to the faculty of the
Virginia Polytechnic Institute and State University
in partial fultillment of the requirements for the degree of
DOCTOR OF PHILOSOPHY
in
CHEMISTRY H
APPROVED: · _ _I
· —¢' Y
' ’ ,;__Ü
.:,4 ~..Ü
David G. I. Kingston, Chairman
/ /‘
‘ I 4 'U / 4 ,ig „
Harold Bellin ‘
Milos Hudlickyé E I
I
/ g ·Vi
_ V
/ ·. . ' -
‘,
”. , :;
· ' .4 Norman G. Lewis Robert H. White
February, 1988Blacksburg, Virginia
STUDIES IN THE BIOSYNTHESIS OF VIRGINIAMYCIN S1
by
. Josephine W. Reed
ICommittee Chairman: David G.I. Kingston
Chemistry
(ABSTRACT)
Some aspects of the biosynthesis of three amino acid residues in virginiamycin S1
have been studied in Srrcpromyccs virginiae by the incorporation of amino acid precursors
labeled with either radioactive or stable isotopes.
L-Lysine-U-MC was incorporated into both the 4-oxo-L-pipecolic acid and 3-
hydroxypicolinic acid residues. The formation of the heterocyclic ring of both of these
amino acids was shown to occur with retention of the nitrogen from the s—amino group of
lysine, as shown by the incorporation of DL-lysine—6J3C-6-UN. In addition, the 3-
hydroxypicolinic acid residue incorporated deuterium from (2RS, SR)-lysine-5-d1, but not
from (2RS, 5S)-lysine-5-d1. This fmding indicates that the 5-pro-(R) hydrogen of L-lysine
is retained during the biogenesis of 3-hydroxypicolinic acid.
In the conversion ofL-phenylalanine to L-phenylglycine, the amino group moves to
the benzylic position. This process could proceed either by an intermolecular mechanism,
in which the original nitrogen is lost, or by an intramolecular mechanism, in which that
nitrogen is retained. Administration of (RS)-phenylalanine·3-BC-UN resulted in its
incorporation with loss of the labeled nitrogen. The process therefore occurs by an
intermolecular mechanism.
ACKNOWLEDGMENTS
Thanks are due to all of those who have helped me achieve this goal, with special thanks to:
TABLE OF CONTENTS
1. Introduction.................................................................................... 1
1.1. Purpose...................................................................................1
1.2. Review of the Literature................................................................ 2
2 . Results and Discussion................................................................... 17
2.1. The Origin of the 4-Oxo-L-pipecolic Acid Residue................................ 17
2.2. The Origin of the 3·Hydroxypicolinic Acid Residue.............................. 31
2.3. The Origin of the L-Phenylglycine Residue.........................................44
3. Experimental..................................................................................53
3.1. General..................................................................................53
3.2. Culture Conditions.....................................................................54
3.3. Isolation of Virginiamycin S1........................................................ 55
3.4. Analysis of Virginiamycin S1 Amino Acids by GC/MS.......................... 55
3.5. Incorporation of L-Methionine-mczhyl-MC.........................................56
3.6. Incorporation of L-Aspartic·U-MC Acid............................................57
3.7. Incorporation of L-Lysine-U·“C....................................................58
3.8. Incorporation of DL·Lysine-6-UC-6—1·’N...........................................58
3.9. Synthesis of 4-Oxo-DL-pipecolic Acid Hydrochloride............................ 59
3.10. Synthesis of Virginiamycin S1 Amino Acid Derivatives for HPLC............. 61
3.11. Separation of Virginiamycin S1 Amino Acids by HPLC..........................63
3.12. Synthesis of (2RS, 5R)-Lysine-5-dl Dihydrochloride............................64
3.13. Synthesis of (2RS, SS)-Lysine-5-d; Hydrochloride.............................. 68
3.14. Incorporation of (2RS, SR)-Lysine-5-dl ...........................................70
3.15. Incorporation of (2RS, 5S)-Lysine-5-dl............................................71
3.16. Synthesis of DL-Phenylalanine-3-13C·15N Hydrochloride....................... 71
3.17. Incorporation of DL·Phenyla1anine-3-1-IC-ÜN Hydrochloride...................75
4. Conclusion....................................................................................76
S. References.....................................................................................77
iv
Appendix...........................................................................................84
Vita.......................................................................................„......... 115
v
LIST OF FIGURES
1. Chemical structure of virginiamycin group A antibiotics.................................... 4
2. Chemical structure of virginiamycin group B-I antibiotics.................................. 6
3. Chemical structure of virginiamycin group B-II antibiotics................................. 8
4. Partial BC-NMR spectrum ofVS; labeled with (2RS)·lysine-6·l3C·6-ÜN:the signal at 38.6 ppm..........................................................................24
5 . Partial HPLC chromatogram ofN-benzoyl derivatives from VS; hydrolysate..........29
6. Partial 13C·NMR spectrum of VS; labeled with (2RS)-lysine·6-13C-6-15N:the signal at 139.6 ppm........................................................................ 38
7. Partial BC-NMR spectrum ofVS; labeled with DL-phenyla1anine·3-BC-15N:the signal at 56.1 ppm..........................................................................51
LIST OF SCHEMES
1. Two pathways from L-lysine to pipecolic acid............................................ 18
2. Synthesis of DL·lysine-6-UC-6-15N.......................................................23
3. Synthesis of 4·oxo-DL—pipecolic acid hydrochloride.....................................27
4. Pathway to nicotinic acid and picolinic acid from L-aspartic acid andglycera1dehyde·3-phosphate.................................................................32
5. Pathway to nicotinic acid and picolinic acid from L·tryptophan.........................33
6. The 2,6-diaminopimelate pathway of lysine biosynthesis............................... 36
7. The synthesis of (2RS, SR)-lysine-5-d;...................................................40
8. Possible mechanisms for the conversion of lysine to ß—lysine..........................46
9. Possible mechanisms for the conversion of phenylalanine to phenylglycine..........48
10. Synthesis of DL·pheny1alanine-3-UC-!-SN................................................ 49
vii
LIST OF TABLES
1. Organisms producing the virginiamycin family of antibiotics.............................. 3
2. Distribution of radioactivity in virginiamycin S; components............................. 16
3 . Incorporation of radioactivity in virginiamycin S;..........................................22
4. Relative abundance of the ions for the base peak (M-COOBu) ofbutylN·trifluoroacetyl-4-oxopipecolic acid from VS; labeled with (ZRS)-lysine·5-13C-6J5N.....................................................................................25
5 . Relative abundance of the ions for the base peak (M—COOBu+H) of butyl3-hydroxypicolinate from VS; labeled with (2RS)·lysine·6-UC-6-HN.................39
6. Relative abundance of the ions for the base peak (M—COOBu+I-I) of butyl3·hydroxypico1inate from VS; labeled with (2RS, 5R)-lysine-5-d; (I) and(2RS, SS)-1ysine·5-d; (II).....................................................................42
7. Relative abundance of the ions for the base peak (M-COOBu) ofN-t1·ifluoroacety1-N-metl1y1·L-phenylala11ine butyl ester (I) andN-trifluoroacetyl-L-phenylglycirie butyl ester (H) from VS; labeled with(2RS)-phenylalanine-3-HC-15N..............................................................52
1. INTRODUCTION
1 . 1 . Purpose E
Virginiamycin S1 (1) is a cyclic peptidolactone antibiotic produced by the soil
bacterium Srreptomyces virginiae. Several of its component amino acids are quite
unusual, and their biosynthetic origins are unknown or incompletely understood. It is the
aim of this project to determine the origin of the 4—oxo-L·pipeco1ic acid moiety and to
investigate some of the mechanistic aspects of the biosynthesis of the L-phenylglycine and
3—hydroxypico1inic acid residues.
MOEH
OH M. O O N
)\/k N ·o . o H 2 ¤5 H
O NH O
OH
1\/
1
1
2
1 . 2 . Review of the Literature
Virginiamycin S1 (1) is a member of the virginiamycin family of antibiotics, which
is made up of two major types of compounds, type A and type B. Together they have the
unique property of exhibiting a synergistic inhibitory action on sensitive organisms.
Because of this feature, antibiotics of this family are often referred to as
"synergimycins."1 A number of these antibiotics have been isolated from various
sources, almost always as mixtures of A and B components (Table 1); patricins A and B
(Figure 2) are chemically synthesized.21
The type A compounds are polyunsaturated cyclic peptolides having a molccular
weight of about 500. Characteristic of their structure is a substituted aminodecanoic acid
and an unusual oxazole system. Six compounds of this group have been isolated. 'I'heir
structures are shown in Figure 1, along with different names that have been assigned these
structures by various investigators.
The type B antibiotics are cyclic peptidolactones with a molecular weight of about
800. Most of the organisms that produce the virginiamycins provide a mixture of type B
components having similar structures. This group of antibiotics is divided into two
subgroups: group B-I (Figure 2), which includes those antibiotics (such as virginiamycin
S1) that contain seven amino acid residues, and group B-H (Figure 3), which includesl
those (such as etamycin) that contain eight amino acid residues,
When virginiamycin was first isolated by De Somer and Van Dijck in 1955 from the
culture broth of a bacterial species isolated from a Belgian soil sample thought to be
Srrepzomyccs virginiae,51 it was shown to be a mixture of four biologically active
components. Infrared spectroscopy suggested that it was similar to streptogramin, which
had been isolated in 1953 by Chamey et al.6 By 1957, the individual components had been
isolated in pure form and their properties identified.52 One of these components was
Table 1. Organisms producing the virginiamycin family of antibioticsz
Microoranism pp p p pp ppp _ pp
Streptomyccsyirginiae virginiamycins (staphylomycins) 3
S.- alborectus virginiamycins 4
S. Ioidensis vcmamycins 5
S. graminofaciens strcptogramins 6
S. osrreogriseus ostreogrycins 7
S. mirakaensis mikamycins 8
S. pristinae pristir1amyci11s 9
S. olivaceous PA114A, B (syncrgistius) 10
S. congancnsis F1370A, B 11
S. griseus vi1idog1isei11(etamycir1) 12S. griseoviridus griscoviridin
S. griseoviridus P8648 neoviridogriscins 13griseoviridin
Strepromyces sp. ctamycin 14
Aczinoplancs philippinensis A-2315 A, B, C 15
A. auranricolor plauricins 16
Acrinoplanes sp. A17002A, B, C, F 17
Aczinomadurajlava madumycins 18
Micromonospora sp. vemamycins 19
Actinomyces dagheszanicus amibiotic 6613 (ctamycin) 20
4
OH•OH
O NH O NHO O
N N
, N E ‘.N-". ~" . äi
. _ /\ °/\ °
Virgi11iamyci11M1 (22) Virgi11iamycinM2 (22)V¢mamyci11A (5) Ostreogxycin G (27)Strcptogrami11A (23) Pristinamyci11HB (26)Ostreogryci11A (24)MikamycinA (25)Pristinamyci11HA (26)Syncrgistin A-I (10)
OH
O NH
NS O
ä 0 °
Griscoviridin (28, 29) ~
Figure 1. Chemical structure of virginiamycin group A antibiotics
5
R
ORe
NO
ONH $
é o/\ 0
Antibiotic (Ref.)
A-2315A (30) R1 = R3 = H, R2 = R4 = OHCP35763 (31)Madumycin H (18)
”
A17002F (17)
A-2315B (30) R1 = H, R2 = OH, R3R4 = OCP36926 (31)Madumyciul (18)
A17002C (17) R1 =R2 = R3 = H, R4 = OH
Figure 1. Continued
6
/ O H
N'“ R1
NH H NO 4n
N
' O O OO O
H
N H Ö N Rg\„„Rs
ANTIBIOTIC (REF) R1 R2 R3 X (other substitutions)
Virginiamycin S1 (32, 33) C2H5 CH; H 4-oxopipecolic acid
Virginiamycin Sg (34) CZHS H H 4—hyd1·oxypipccolic acid
Virginiamycin Sg, (34) CZHS CH; H 5-hydroxy-4-oxopipccolic acid
Virginiamyciu S4 (34) CH; CH; H 4»oxopipccolic acid
Virginiamycin S5 (35) CZHS CH; H allo ·4-hydroxypipecolic acid(alaninc not phenylglycinc)
Strcptogramin B (36) CZHS CH; N(CH;)2 4-oxopipecolic acid °
Mikamycin IA (37)PA1 14B1 (38)Pristinamycin LA (26)Vemamycin Ba (39)Ostrcogxycin B (40)
Pristi11amycinIC (26) CH; CH; N(CH;)2 4-oxopipccolic acidVcmamycin By (39)Ostreogrycin B1 (40)
Figure 2. Chemical structure of virginiamycin group B-I antibiotics
7
ANTIBIOTIC (REF) R1 R2 Rs X (other substitutions)
Pristinamycin IB (26) C2H5 CH; NHCH; 4-oxopipecolic acidVemamycin Bß (39)Ostrcogrycin B2 (40)
Vemamycin B8 (39) CH; CH; NHCH; 4»oxopipeco1ic acid
Ostreogrycin B; (41) CZHS CH; N(CH;)2 5-hydroxy-4-oxopipecolic acid
Vcmamycin C (42) C2H; CH; N(CHs)2 aspartic acid_ Doricin
Patricin A (21) C2H; CH; H proline
Patricin B (21) CZHS CH; H pipecolic acid
Plauracin BI (31) CH(CH;)2 CH; H pipecolic acid(p-McO—pheny1g1ycir1c)
Figure 2. Continued
8
’ ‘\ 0H„ /
ä IQH H X
u o‘ H
. 0 0 0o o H
NcH„ gn, NcH,>—-{ o
0 H 0NcH,
H2
Antibiotic (Ref.) ' R1 R2 X
Ncoviridogriseinl (43) C2H5 CH; H
Neoviridogxiscin H (44) CH; CH; HEtamycin (45)
NeovirdogriseinHI (43) CZH5 CH; OH
Viridogriscin (46) CH; CH; OHEtamycin (47)6613 (20)N6OViI'idOgI'iS¢iI‘lIV (48)
Ncoviridogrisein-MP (49) CH; CH; CH;
Ncoviridogriscin-C1 (50) CH; CH; C1
Viridogrisein H (35) CH; H QH
Figure 3. Chemical structure of virginiamycin group B-H antibiotics
9
designated factor S (later called virginiamycin S;) because of its activity against Bacillus
subrilis. It is a weak acid, with a pKa of 9.0 in ethanol and 7.7 in a 1:2 mixture of
dimethylformamide and water. It is insoluble in water and petroleum ether, but soluble in
most other organic solvents.
A communication by Vanderhaeghe and Parmentier reported the structure of
virginiamycin S1 (1) in 1959.32 They published a full account of its structure
determination the following year.33 The structure of virginiamycin S1 was determined by
hydrolysis, partial hydrolysis, and Edman degradation; the antibiotic was shown to
consist of a lactone ring containing, in order, L-threonine, D-or-amino-n-butyric acid, L-
proline, N-methyl-L-phenylalanine, 4-oxo-L·pipecolic acid, and L·phenylg1ycine residues.
The hydroxyl group of threonine forms the lactone with the carboxyl function of
phenylglycine, and the amino group of threonine is acylated with 3-hydroxypicolinic acid.
The primary structure proposed by Vanderhaeghe and Parmentier was subsequently
conf'u·med by crystal structure determination of the methanolic solvate.33 The tertiary
structure of the macrocycle was shown to be constrained by a transannular hydrogen bond
between the carboxyl oxygen of the N-methylphenylalanine residue and the amide
nitrogen of the phenylglycine residue. Further confirmation of its structure came from
mass spectral studies carried out by Kiryushkin et al.3‘I and later by Compemolle et a1.55
The solution conformation of virginiamycin S1 and virginiamycin S4 has been
reported.56 Features of its conforrnation, determined from IH and I3C nuclear magnetic
resonance studies, include a bend in the proline-N-methylphenylalanine-4—oxopipecolate-
phenylglycine region and the aforementioned hydrogen bond between N-
methylphenylalanine and phenylglycine, as well as a weaker hydrogen bond between the
lactone oxygen and the amide nitrogen of or-aminobutyric acid. The benzyl group of N-
methylphenylalanine is folded over the 4-oxopipecolic acid ring; the 3-hydroxypicolinic
acid residue lies outside the region of the peptidolactone ring. Comparison of the results
ß10
of a similar study of the solution conformations of allohydroxy- and deoxyvirginiamycin
S (obtained by sodium borohydride reduction of the parent antibiotic) to that of
virginiamycin S1 shows that the latter has the most rigid conformation in the series.57
Comparing the biological activity of these antibiotics (as determined by Janssen et a1.),58
the investigators concluded that virginiamycin S1 has the optimal conformation for binding
to the bacterial ribosome, at least in the cases where the type B antibiotics are active
without the presence of type A antibiotics.57
Virginiarnycin S1 has recently been synthesized by Kessler et al. by two different
routes.59 In one method the cyclic hexapeptide was constructed and then coupled with 3-
hydroxypicolinic acid. In the second, a linear hexapeptide containing the 3-
hydroxypipecolate was cyclized. The benzyl ether of virginiamycin S1 has been
synthesized in a similar fashion.6°
The virginiamycin-like antibiotics inhibit the growth of bacteria. Gram-positive
microorganisms are more sensitive (minimum inhibitory concentration, 0.1-5 ug/mL)
than gram-negative bacteria (minimum inhibitory concentration, 5-200 pg/mL), although
some gram-positive bacteria (Mycobacreria) are quite resistant and some gram-negative
ones (Haemophilus and Neisseria) are susceptible. The difference in activity is due to the
differences in the permeability of the bacterial cells; ribosomes from gram-negative
organisms are just as sensitive as those from gram-positive organisms in cell-free
systems.61
Individually the A components and the B components stop bacterial growth
reversibly; however, a mixture of the two causes a 10- to 100-fold increase in inhibition as
well as loss of cell viability, depending on the organism tested. The maximum activity is
seen in 1:1 to 2:1 mixtures of components A and B, ratios usually found in nature.1 Early
work with the virginiamycins has shown that a 70:30 mixture is the most active against
gram-positive microbes, except in Bacillus subrilzls, where the effect is reversed.62
11
The viigiaiamycihe have limited effect on eucaryotic organisms. ln Euglena, type
A compounds block chlorophyll synthesis and chloroplast multiplication, causing a
reversible bleaching of the cells and inhibition of photoautolrophic growth. Type B
compounds alone show no appreciable effect, but a mixture of A and B antibiotics causes
permanent b1eaching.1· 61
The virginiamycins are widely used today as feed additives in animal husbandry as
performance promoters for cattle, swine, and poultry. Growth of the animals is enhanced
by the inhibition of the intestinal tlora, especially grarn—positive bacteria.61 A number of
characteristics has led to the success of the virginiamycins in this application:63 (1)
activity confined to gram·positive microorganisms; (2) low solubility and low absorption
through the intestinal wall and therefore little accumulation in animal tissues; (3) relative
stability at acidic pH; (4) high safety standards related to its synergistic activity; for
example, lowered probability of resistance induction and absence of toxicity or residue
problems. In addition to their use as feed additives, the virginiamycins are used in
veterinary medicine, especially in the treatment of swine dysentery.6"· 66
Because it is possible to transmit to humans, through meat, bacterial strains
carrying plasmids with antibiotic-resistance factors, antibiotics used as growth promoters
are not used in the treatment of human infections. For this reason, the virginiamycins are
used only in the field of animal husbandry today.61 For several years, however, they
were used successfully in a number of applications in human medicine, including the
treatment of whooping cough66·67 and "staphylococcal infections of the skin.68
Some work has been done on the structure—activity relationships of the type B
· virginiamycins. Nature already provides a variety of analogs, for the most part in the
pipecolyl and the N·methylphenylalanine residues: all these congeners provide the same
— antibiotic activity and ability to bind to bacterial ribosomes.6° A number of derivatives of
pristinamycin IA have been prepared by a group of French workers: some of these have
12
assorted substituents introduced at the 5·position of 4·oxopipecolic acid,7°· 71 and in
others the ketone functionality of the same residue was replaced by various amine
derivatives.72 In all cases the activity of the derivatives against Szaphylococcus aureus
was in the range of 0.1-125 ug/mL, comparable to the activity of the naturally occurring
virginiamycins. Derivatives at the phenolic functionality have also been prepared.6°
When pristinamycin IA is acetylated, the antibiotic activity is preserved; however,
esterification by an allyl moiety leads to complete loss of its biological activity.
Apparently the acetylated derivative is hydrolyzed readily to the natural antibiotic. A linear
molecule, obtained by hydrolysis of the lactone ring of pristinamycin IA by means of an
enzyme isolated from a resistant strain of Sraphylococcus aureus, is devoid of antibiotic
activity.77
The virginiamycin·like antibiotics act by inhibition ofprotein synthesis.61 In vivo
and in vitro studies have shown that both type A and type B antibiotics bind to the SOS
subunit of the ribosome, the site of protein synthesis in the cell.7"‘7‘ The type B
antibiotics bind to the ribosome in stoichiometric amounts. The type A antibiotics,
however, bind in sub·stoichiometric amounts: when these detach from the ribosome,
alterations produced in the particle persist.77 It has been shown that the donor and
acceptor sites of peptidyl transferase on the 50S subunit are permanently inactivated on
transient incubation with virginiamycin M.78 In cell-free systems interaction of
virginiamycin M with the ribosomes enhances their affinity for virginiamycin S six—fold.76
Apparently the type A compounds cause a conformational change in the ribosome that
increases its affinity for the type B antibiotics. At this time it has not been determined
which step of protein synthesis is the specific target of the type B virginiamycins.7°
Although they are quite different in structure, virginiamycin S; and the other type
B antibiotics are related to the macrolidc antibiotics (eg., erythromycin, leucomycin,
spiramycin, tylosin) and the lincosamide antibiotics in their mode of action and in their
13
cellular targets.8° All have reversible activity on bacteria and bind to the SOS ribosomal
subunit in 1:1 complexes.7° The macrolides compete with virginiamycin S in the binding
reaction to the ribosomes; this suggests that the binding sites of these two groups of
antibiotics overlap.81 When type A virginiarnycins are present, the conformational change
they trigger'was
found to lower the affinity of ribosomes for the macrolides while
enhancing their affinity for type B antibiotics.79
The intrinsic fluorescence of the type B virginiamycins due to the 3-
hydroxypicolinyl portion of the antibiotic has allowed investigators to determine the
location of the ribosomal binding site by fluorescence energy transfer.82 In addition,
fluorescence quenching studies have shown that the binding site on the surface of the
ribosome is in the shape of a well that accommodates the 3-hydroxypicolinyl residue.
Access of the antibiotic to this site seems to be controlled by both hydrophobic interactions
° and electmstatic forces.85
Because ions are responsible for the shape of ribosomal particles, the presence of
monovalent and bivalent cations affect the activity of the virginiamycin-like antibiotics.79
The binding of type B antibiotics to the ribosome is dependent on the presence of NH4+
or K+ and also Mg2+ if monovalent ions are present. The conformational change of the
SOS subunit induced by the type A antibiotics requires either NI-14+ or K+ and either
Mg2+ or Ca2+.8"· 85 The interaction of type B antibiotics with cations (protons and
alkaline earth cations) has also been studied.86·87 These cations form a complex with the
3-hydroxypicolinic acid portion of the antibiotic, which can act as a dibasic acid.
Virginiamycin S1 has also been shown to facilitate the transport of protons and other
cations across phospholipid bilayer membranes in much the same fashion as the natural
protein cation carriers in membranes.87 However, the role of this phenomenon in the
activity of the type B antibiotics has not been shown.
The biogenesis of a number of medium-sized peptide and depsipeptide antibiotics
14
has been shown to proceed by a multienzyme thiotemplate mechanismgs and not by the
ribosomal mechanism that produces proteins. Etamycin formation is not inhibitcd by the
presence of chloramphenicol, an antibiotic that inhibits protein synthesis; this finding
suggests that, it too is formed by the multienzyme thiotemplate mechanism.89 Presumably
the other type B virginiamycins are formed by a similar mechanism. The speciticityofthese
templates are not as strict as that of the messenger ribonucleic acid templates
mediating the synthesis ofproteins, and similar amino acids can replace each other, so that
a variety of antibiotic analogs are produced by a single organism under a variety of growth
conditions. This phenomenon is seen in the organisms producing the type B
virginiamycins. Srreptomyces virginiac, for example, produces five related type B
antibiotics. (See Figure 2.) Scientists have been able to take advantage of this in a
procedure known as directed biosynthesis. For example, the organisms that produce
etamycin normally synthesize only that antibiotic, but the addition of a particular amino
acid to the culture in high concentration or indeed as the only nitrogen source has resulted
in the production of a number of novel congeners where the added amino acid has
replaced an amino acid residue (usually 3-hydroxyproline) in the antibiotic.25- 43- 44- 49-
50 (See Figure 3.)
Before Molinero began the study of the study of the biosynthesis of virginiamycin
$1,90 the only type B antibiotic whose biosynthesis had been studied was etamycin.91- 92
The origin of each of the constituent amino acids of that antibiotic was established by the
incorporation of radioactively labeled amino acids. Because etamycin bears some
resemblance to virginiamycin S1, the results of these studies provided a starting place for
planning a similar study of this antibiotic. Etamycin and virginiamycin S1 both contain a
threonine residue and a 3·hydroxypicolinic acid residue. In addition, the phenylsarcosine
residue of etamycin is merely the N-methyl derivative of phenylglycine, found in
virginiamycin S1. And both etamycin and virginiamycin S1 contain N-methyl amino acids.
15
It is not surprising that radioactively labeled L-threonine was efficiently incorporated in the
corresponding portion of both antibiotics. In contrast to the 3-hydroxypicolinic acid
residue in the antimycobacterial antibiotic pyridomycin, in which it was shown that L-
aspartate, glycerol, and pyruvate were efficient precursors,°3 L-lysine was shown to be the
best precursor for the same residue in both etamycin andvirginiarnycin S1. The results in
etamycin were confirmed in a later biosynthetic study, which also showed that 5—hydroxy·
L-pipecolic acid and 5-hydroxy-DL-lysine were incorporated into the 3-hydroxypicolinic
acid residue.9‘ As in the phenylsarcosine portion of etamycin, the phenylglycine residue of
virginiamycin S1 arises from L-phenylalanine. The N-methyl groups of both antibiotics
were labeled by L·methionine-merhyl-13C. Molinero's work also established that the L-
proline residue incorporates L-proline·U-MC and that the N-methyl-L-phenylalanine
residue incorporates L-phenylalanine-U-MC. The biogenetic origins of the remaining
residues of virginiamycin S1 are less clear. Of the amino acids incorporated into the
antibiotic, L-threonine seems to be the best precursor of the D-a-aminobutyric acid residue.
And small but significant amounts of labeled L-aspartic acid, L-lysine, and L-methionine
seemed to be incorporated into the 4-oxo-L-pipecolic acid. The results of Molinero's
research are summarized in Table 2.
16
H}
Q- Q Q ·==2 -1 -0-0 -0 Q2 ¤~ .2kl E 3 °?
"‘°.E§ 2 Q $„ SQ 2 T9
° ' 2E1:I--· 2*2 -2
°·62*I q- -¤·"‘ ·x·
‘° O -7* >~.¤.E * '^ N I: ·- co>~.
u"">~--—><
cg -¤¤,g··o
2* I cs--E6-=>.gp
” *7 *7 N Y YQ Y°‘ N Ö_öi><?'*‘F
_=3 o c c --6 6 5 5 gllällqmlm
II Ilällägäg-Ü
.¤:
S Im „ N „ NN „ Q C 2<¤2c6:
c: I °1 2 Q :**2 Q Q Q
* IW
S -·.
g E I Q Q! Q <t -0-0 ~¤ ·— cE_
§ I v c 3 o NG c oo ~o
o I° I
-•vz E . Q Q! Q! *t ·1<t *1 Q ·—·
=¤_
I c o o tg mc ·n <f>, 1
IN
2 ,_1
*6E ¤ r g_5 an | *1 *0 Q *1 QQ Q Q Q §G (6 •-•$ •—•Q¤'·
E-
>. I- Q! Q! Q Q! ·—·¢ QQ Q ¤~ -2f: o ~o N vw oioi „-E qi ,.1 o
blI-IG
0wu':
an§**1
G*’~
ä•1•·'g
¤-eé Q: ¤<e Q! <>0 Ä E ä äüI I,. mix ·'* •·•
E G GO Ö Ö Öoz);-‘
••- V=‘E>
°
“1>„ .Z·”
2 22 2äg $3 U 0 -2'3 2NN
2 g 6 ag *7 § ggg -2 U*2 Ö L1 L> 2"' E Ö
· ···· » Q Z"- ä °§- 9 E
E _2 — ¤„-„ ¤ . ¤- <»>< = 2Sa —- 2 S 2* 6 Q 8 mg 2 °>~. Q -,- aä
,;_- I: ···* ·¤cu I:-' 0 5 --• ···· ' ä,. Q •-·
E E E ** % äéä E E :2 mw 5 2C6 ß-I E" ( Av Q,. < I -X- -16 -16
I I I I I I I I-]* *•-'I •-l •-'I I-] I-I I-] 5] G -16
2. RESULTS AND DISCUSSION
2 . 1 . The Origin of the 4-Oxo·L-pipecolic Acid Residue
4-Oxo-L-pipecolic acid (2) is an unusual amino acid; to date it has been detected
only in the virginiamycin type B antibiotics. The related amino acids pipecolic acid (3) and
4-hydroxypipecolic acid (4), however, are more common. Pipecolic acid is widely
distributed in plants and has been isolated from the fungus Neurospora and rats.°5 4-
Hydroxypipecolic acid is also found in several plant species;96 in addition, it replaces 4-
oxo-L-pipecolic acid in two of the congeners of virginiamycin S1—vi1·giniamycins $234
and 85,35 both of which are coproduced with virginiamycin S1.
Q OH
QQH QCOOH COOH
H H H
2 3 4
The biosynthesis of both of these amino acids has been studied. Gupta and
Spencer have shown that pipecolic acid in rats, in Neurospora crassa, and in two higher
plants (Phaseolus vulgaris and Sedum acre) is formed from lysine with the retention of the
8-amino nitrogen.97 They have proposed that the conversion of lysine into pipecolic acid
proceeds via 6-amino-ot-ketocaproic acid (5) and A1-piperideine-2·carboxylic acid (6)
(Scheme 1, pathway A). Other researchers have observed similar results in the study of the
origin of pipecolic acid.98·”·1°° On the other hand, Fowden has observed that pipecolic
acid in Acacia arises from the loss of the e—amino nitrogen: the pathway in this case
proceeds by way of ot-aminoadipic-5-semialdehyde (7) and A1-piperideine-6-carboxylic
acid (8) (Scheme 1, pathway B).1°‘ The biosynthesis of 4-hydroxypipecolic acid has also
17“
18
L·LysineE
fl IlHQN o COQH o HQN COZHs 7
Erg\co,H CO2H
6 8
H3
Scheme Two pathways from L-lysine to pipecolic acid
19
been studied in Acacia.l°1•1°2 Unlike 5-hydroxypipecolic acid, which is produced from 5-
hydroxylysineßos 4-hydroxypipecolic acid arises from the hydroxylation of pipecolic acid,
not from 4-hydroxylysine. Indeed, 4-hydroxylysine is rarely observed in nature. (L-thr6o-
Y-hydroxylysine has been isolated as a component of the peptide antibiotics cerexin A and
B_)104
It is not surprising that the biogenetic origin of 4-oxo-L-pipecolic acid has
particularly interested us. Of the seven amino acid components comprising virginiamycin
S1, however, 4-oxo-L-pipecolic acid has been most difficult to study. It is the only one of
the seven that is not cornmercially available; in addition, the amino acid is sensitive to the
basic conditions used in many of the traditional amino acid derivatization procedures. The
development, therefore, of a suitable derivative for use in its separation by HPLC from the
other components of the antibiotic and in its quantification has been tedious. Because of
these problems, at the conclusion of Molinero's work, there was still no satisfactory
answer to the question ofitsIn
his biosynthetic studies of virginiamycin S1,°° Molinero hydrolyzed the
radiolabeled antibiotic, then prepared N-benzoyl derivatives of the constituent amino acids
(except 3-hydroxypicolinic acid, the nitrogen of which cannot be acylated.) These
derivatives were separated by reversed-phase HPLC, collected, and the radioactivity of
each counted. In the case of 4-0xO-L-pipecolic acid, its derivative was produced in such
small yield (4% as compared to 40-50% for the others) that the peak seen on the HPLC
chromatogram rises barely above the baseline. In addition, Molinero was unable to obtain
the N-benzoyl-4-oxopipecolic acid standard in crystalline form.
The results of these studies showed that L-lysine-U-MC, L-aspartate-UJ"C, and L-
methionine-merhyl-14C were incorporated into the 4-oxo-L-pipecolic acid part of
virginiamycin S1, although none of these were incorporated in dramatic proportions.
Certainly the lysine result seemed valid on the basis of the studies on pipecolic acid and 4-
20
hydroxypipecolic acid. And L-aspartate is a reasonable precursor; a number of mechanisms
can be imagined that would transform it into 4-oxopipecolic acid. But it seemed unlikelyI
that L-methionine be a precursor, as methionine usually provides methyl groups, and 4-
oxopipecolicacid has, of course, no methyl groups.‘
The frrst task in the second phase of the study of the biogenesis of virginiamycin S;
was to devise an improved procedure for the derivatization of4-oxopipecolic acid, then to
repeat the experiments in which radiolabcled L-lysine, L—aSpartatc, and L-methionine are
incorporated into virginiamycin S1. Because of its relative instability in basic solution,
however, the amino acid continued to resist attempts at derivatization. Anxious to
reexamine the questionable incorporations, we decided to do the incorporation experiments
and to separate the 4-oxo-L-pipecolic acid from the other amino acids in an underivatized
form by ion-exchange chromatography.33 .
In the experiment in which L-methionine-methyl-MC was fed, the label was, as
before, incorporated into the antibiotic with a specific incorporation of 2.4 x 10*2.
(Specific incorporation = 100 X specific activity of VS;/specific activity of precursor.)
(The N-methyl·L-phenylalanine portion incorporates labeled methionine; we assume the N-
methyl group comes from the S-methyl of methionine.) The 4-oxo-L·pipeco1ic acid
isolated by ion-exchange chromatography showed virtually no radioactivity: less than 2%
of the radioactivity of the intact virginiamycin molecule was located in the 4·oxo-L-
pipecolic acid moiety. Virginiamycin S1 isolated following the incorporation of L·lysine-
U·1"C showed a specific incorporation of 9.8 x 10*3, a figure comparable to the result of
Molinero's experiment. However, when L-aspartate-U-MC was administered, the
observed specific incorporation (5.9 x 10*4) differed dramatically from that seen in the
earlier experiment and was significantly lower than the incorporations of labeled methionine
and lysine. From this experiment, it can be concluded that L-aspartic acid is not a precursor
to any part of virginiamycin S1. Because it is unlikely that there exist two major
21
biosynthetic routes to 4-oxopipecolic acid and because the results from the labeled lysine
incorporations are consistent, one can only conclude that L-aspartic acid is not a precursor
to 4-oxo-L-pipecolic acid or any other part of the antibiotic. The results of the above
experiments are summaxized in Table 3.
A later experiment in which DL-lysine-6-UC-6-UN (9) was fed to S. virginiae
confirrned that L-lysine is indeed a precursor to 4-oxo·L-pipecolic acid. Also this particular
experiment answered the question of which nitrogen of lysine is incorporated into 4·oxo-L-
pipecolic acid: does the cyclization of lysine proceed by pathway A or pathway B (Scheme
1). The doubly labeled lysine was prepared by M. B. Purvis1°5 by a synthetic strategy
(Scheme 2) based on the one that was used to prepare (2RS, 5R)·lysine-5-d1 and (2RS,
5S)·lysine-5-d1, which were used in an experiment to study the mechanism of formation of
the 3—hydroxypicolinic acid portion of virginiamycin S;. (See Section 2.2.) The BC-
NMR spectrum of the virginiamycin S1 isolated following the feeding of DL-lysine-6-13C-
6-ÜN showed that the resonance at 36.8 ppm due to C-6 of the 4·oxo-L·pipecolic acid
residue was enhanced and appeared as a doublet (UCN = 10 Hz) (Figure 4). This result
clearly shows that the the e-nitrogen of lysine is retained in 4-oxo·L-pipecolic acid in
virginiamycin S1. It was contirmed by GC/MS analysis of the N-trifluoroacetyl butyl
esters of the amino acids obtained by acid hydrolysis of the labeled virginiamycin S1. The
data shown in Table 4 are the relative abundances of the ions (from the loss of the
butoxycarbonyl group) due to the base peak of the 4-oxopipecolic acid derivative. These
data show that this peak consists of 1% singly labeled species and 6% doubly labeled
species. Incidentally, we did not detect the incorporation of the deuterated lysines (Section
2.2) into 4-oxo-L-pipecolic acid. The deuterium is located at the 5-position, adjacent to the
carbonyl group, and the hydrogens at this position apparently exchange in the slightly
acidic aqueous fermentation medium or during the derivatimtion process.
Although the answers to our questions were apparent, it was still our aim to
22
**. c~0 vw: ..*1; Ä
6 °é 6
' vz
*2,5 o0
_EM
___
vz"‘ ¤~
¢E =\/ ·ä
\-
2 ä,5 *5 c
:
Q V! E
•-«
'; E ·2 „ E °0\ V7 oo °^ gg g
d, 0 °" <'-• b 0
{5* x- UN¤.
Eck
t:‘·* ;5 ZP
8 u "*E
2 ‘ä
.2 ·»¤-• 3
.s 6.¤ ,g
an r-3 ·
.*.: 8 c ·¢ g ä
> E ,5 °! **2 3 2
r:c '^ · "'
•;
ä>
>.E
32=* ZPä -2
U
cu6
.2 ä9
'¤ z: ^Q 9.::
N uf0
:.
.2ÜQ
L)g II
*-1*.
c:
g_ 4. N S .9
s„.Q Ü
"‘ ES
c K; ¤ u ¤
ä 2 °*= Ö
I-!
• 3; U.o Q
° ' E. ä 9
•
"‘ E
M .9 ·° ,5g*2
ä Ü .9 8 ···
ä g 3* 2 E §_
u¤ö
·—* .-J '·¤·
23
0 OCO Et CO Et2 i. „__|Ä\/°"°CO;Et CO;Et
O 0
Oc°2EI lu“_ lll.
c°2EI
0
OCO;Et _ CO;Et
N,_|/\/\gM, -—--9;-»Nwiacißn
cO2Et CÖQEÜ
O O
V . H°2c\I/\/\I3CHgI5NHg .NH;
9
Scheme 2. Synthesis of DL-lysine-6-BC-6-HN.104 i. CH;=CHCHO, NaOMc;ii. 9-BBN; iii. MSCI, Et3N; iv. Na13C15N, DMSO; v. H2, PtOg; H3O+
24
38.0 39.0(ppm)
Figure 4. Partial BC-NMR spectrum of VS; labeled with (2RS)-lysine-6-BC-6-15N: the signal at 38.6 ppm _
25
Table 4. Relative abundance of the ions for the base peak (M—COOBu) of
butyl N·trifluoroacetyl-4-oxopipecolate from VS; labeled with (2RS ) -lysine-6-HC-6-HN
„„„ll 196Labeled 100 12 7
Unlabeled 100 11 1
C¤rr¤¢t¤dl¤b¢l•=d* 111 1 Ä11—···· 931*
Relative abundances of the ions from the labeled compound corrected for the naturallyoccurring A+1 and A+2 contributions of the unlabeled and singly labeled peaks.
* * Mole % unlabeled, singly labeled, and doubly labeled species, respectively
26
develop a satisfactory method of producinga derivative of 4-oxo-L—pipecolic acid that
would allow us to separate it from the other amino acids in the antibiotic by HPLC. It
‘ should therefore be a UV-active compound to allow for its detection in our HPLC system.
But because we did not have a readily available supply of the amino acid, its development
was difficult. At first, we obtained a small supply of by hydrolyzing a portion of
virginiamycin S1, then subjecting the hydrolysate to ion-exchange chromatography. In this
manner, we were able to obtain about 220 mg of reasonably pure 4-oxo-L-pipecolic acid
hydrochloride. As this supply diminished in what seemed to be endless futile attempts in
making a suitable derivative, the appeal of devising a synthesis of this amino acid
increased. A search of the literature revealed that 4-oxopipecolic acid had been synthesized
previously by Clark-Lewis and Mortimer in low yield (1.5% in 3 steps) from methyl 4-
chloropicolinate via 4-hydroxypipecolic acid.1°‘ The yield was improved by French
workers to 16% by the use of different conditions in the step in which the aromatic
picolinic acid derivative is reduced to 4—hydroxypipecolic acid.1°7 Additional reading of
the literature uncovered a synthesis ofpiperideine 11, which was used as an intermediate in
syntheses of the piperidine alkaloids (+)-4-hydroxysedamine and (+)-4-
hydroxyallosedamineßos It seemed an ideal starting material for the synthesis of the
desired amino acid.
The synthesis of 4-oxo-DL-pipecolic acid hydrochloride was undertaken as outlined
in Scheme 3. Piperideine 11 was prepared in 2 steps from the commercially available 4-
piperidone ethylene ketal (10). Addition of the elements of HCN, followed by hydrolysis
of the nitrile provided protected 4-oxopipecolic acid (13) in 35% overall yield.
Deprotection of the keto group proved to be somewhat troublesome, however. The first
attempt at hydrolysis apparently succeeded, but the process of stripping the aqueous acid
on a rotary evaporator only caused the ketal to re·form. Neutralization of the hydrolysis
reaction mixture followed by ion—exchange chromatography afforded 4-oxo-DL-pipecolic
27
O O · O O Q Q
N NN/
H ICI
10 11
0 0 0 o
iii. ‘Ü
iv. ÜN CN N COQHH H
12 13
0——“ Ö_ N co Hcl + H2
2
14
Scheme 3. Synthesis of 4-oxo-DL-pipecolic acid hydrochloride. i. NCS;
ii. KOg,18-crown·6; iii. KCN, pH 5; iv. Ba(OH)2•8HgO; v. 2NHC1
28
acid hydrochloride (14) in modest yield. At the present time the yield of the last step of the
synthesis has not been optimized. Because of the poor yield in the last step, the synthesis
does not compete favorably with the procedure already existing in the literature. If the
isolation of the final product were to be improved, however, this synthesis would provide a
convenient source of 4-oxo-DL-pipecolic acid. .
Ironically, the problem of isolating the synthetic amino acid provided a solution to
the synthesis of a good derivative of that amino acid. The protected 4-oxopipecolic acid
underwent acylation with benzoyl chloride to yield its N-benzoyl derivative (15).
Preparation of the derivative in the virginiamycin S1 hydrolysate was accomplished by the
addition of a drop of ethylene glycol to the rnixture before removal of the aqueous acid on
the rotary evaporator, then acylation with benzoyl chloride by the Schotten—Baumann
method. The new derivative separated nicely from the other amino acid derivatives with the
same HPLC conditions that had been worked out previously.9° The magnitude of the peak
in the HPLC chromatogram is comparable to those of the other virginiarnycin S1 amino
acid derivatives. Figure 5 shows a representative chromatogram.
Ph/R0
1 5
This work establishes that L-lysine is a precursor to 4-oxo-L-pipecolic acid in
virginiamycin S1. It shows that the e-amino group of the lysine molecule is retained in the
cyclization to the heterocyclic compound. Additional biosynthetic studies would further
elucidate the mechanism by which lysine is converted to 4-oxo-L-pipecolic acid.
29
1. N-benzoyl-L-thrconinc
2. Ngbcnzoyl-L-prolinc
3. N—bcnzoyl·D-oc-aminobutyric acid
4. N-bcnzoyl-4-oxo-L-pipecolic acid cthylcnc kctal
5. Bcnzoic acid
6. N-bcnzoyl-L·pher1yla1anine
...1 ... -...2 -...-3.4 5
· 0 5 10 15 20 25
· (minutes)
Figure 5. Partial HPLC chromatogram of N-benzoyl derivatives from VS;
hydrolysate
30
Incorporations of pipecolic acid, 4-hydroxypipecolic acid, and 4-hydroxylysine would
establish whether these compounds are intermediates in the process. Also the relationship
between the origin of this amino acid and those of its congeners (4-hydroxypipecolic acid
and 5—hydro:gy—4··oxopipecolic acid) would be an interesting question to examine.
31
2.2. The Origin of the 3-Hydroxypicolinic Acid Residue
The pyridine nucleus is found in a number of bacterial metabolites having widely
disparate origins. For example, fusaric acid (16) is derived from aspartic acid and
acetate.1°° The pyridine nucleus in proferrorosamine A (17) is derived from lysine by way
ofpicolinic acid,11° whereas phomazarin (18) arises from a polyketide chainßu
\\ I ,
HOgC N N
1 6 1 7
IO O H
I\ O H
wu u/ cozuOH O
1 8
There are two biosynthetic pathways to picolinic acid and its isomer nicotinic acid
lmown to operate in different microorganisms.112 In one pathway (Scheme 4), aspartic
acid and glyceraldehyde-3-phosphate are condensed to form a heterocyclic ring, which is
then transformed to quinolinic acid (19). Quinolinic acid can then be converted to either
picolinic acid (20) or nicotinic acid (21). In the other pathway (Scheme 5), tryptophan
undergoes an interesting series of reactions to produce quinolinic acid and then picolinic
acid or nicotinic acid.
The biosynthesis of the 3-hydroxypicolinic acid (22) has been studied in the
antibiotics pyridomycin?3 and in etamycin.9192 In pyridomycin, 3·hydroxypico1inic acid
32
°"op
H °°=H H cozu‘ + ——_”
L
H CO2H CO2H
\NCOQH N COZH
N CO2H
19
/ . /N COQH N
20 21
Scheme 4. Pathway to nicotinic acid and picolinic acid from L-aspartatic
acid and glyceraldehyde-3-phosphate
33
NH2 ÜNH
o co,¤·• O OOzH
H
O COQH
NH: C02H CO2H
NH NH OHC NH2 Z Hozc ZOH OH
OgHN CO2H
19
\ \N C02H N
2 0 2 1
Schemc 5. Pathway to nicotinic acid and picolinic acid from L-tryptophan
34
originates from L-aspartic acid and glycerol or pyruvate by what could be a mechanism
similar to that of the biogenesis of picolinic acid. In etamycin, a virginiamycin type B
antibiotic, L-lysine is a precursor of the same residue, and L-aspartatc is not incorporated in
any signiticant amount. Further studies have shown that 5-hydroxylysine (mixed
isomers)113 and 5-hydroxypipecolic acid (from Sigma, isolated from Phoenix dacrylüfero,
dates, and thus of the (2S, 5R) configuration)113 are also precursors?4
O HI N/
CO2H
2 2
Molinero's work°° on virginiamycin S1 showed that both L-aspartic acid and L-
lysine are incorporated into 3—hydroxypico1inic acid (Table 1). Molinero also fed labeled
tryptophan to S. virginiae, and he found no incorporation into the antibiotic. However, the
tryptophan he used was labeled at C-3; if it were incorporated into the 3-hydroxypicolinyl
portion of virginiamycin S1 by a mechanism like that shown in Scheme 5, then the labeled
carbon would be lost, and no incorporation would be detected. It seems likely though that”
the mechanism in virginiamycin S1 is more like that in pyridomycin or etamycin, The more
appropriately labeled tryptophan-arJ"C was administered in the cases of both pyridomycin
and etamycin, and virtually no incorporation was observed.
In etamycin, where lysine was incorporated approximately 10 times more efficiently
than aspartate, and in pyridomycin, where no incorporation was detected at all, the
biogenetic precursors are clear-cut. In virginiamycin S1, however, both precursors were
incorporated into the 3-hydroxypicolinic acid residue in comparable magnitudes. Molinero
354
attempted to explain the disparity by suggesting that L-lysine was the true precursor and
that L·aspartate was being converted to L-lysine by way of the 2,6·diaminopimelate
pathway (Scheme 6), known to operate in bacteria.114 If this were the case, however, one
would expect to see a labeling pattern more like the one that is seen in etamycin, where the
incorporation of aspartate is significantly less than that of lysine.
In conjunction with the study of the origin of 4-oxo·L-pipecolic acid (see Section
2.1) the labeled L-aspartic acid experiment was repeated. It was found that the specific
incorporation of radiolabeled aspartate was extremely low—5.9 x 104. This finding
suggests that L-aspartate is not a significant precursor to any part of the virginiamycin S1
molecule. It is in agreement with the f'mdings in the closely related antibiotic etamycin.
Thus L-lysine is the principal precursor to 3-hydroxypicolinic acid in virginiarnycin S1.
The questions that we have undertaken to answer in this project concem the
mechanism by which L-lysine is transformed into 3-hydroxypicolinic acid. First, which of
the two nitrogens ofL-lysine is retained in the cyclization to the heterocyclic ring system?
Second, is there any stereospecificity in the loss of the hydrogens at C-4, C-5, and C—6 of
lysine as the aromatization process takes place? In other words, is the pro(R) orpro(S)
hydrogen retained preferentially at each of the three positions?
The first question is related, of course, to the one asked in the case of the origin of
4-oxo-L-pipecolic acid. The answers to both questions were obtained in the same
experiment. Thus (2RS)-lysine·6-BC-15N1°4 was administered to growing cultures of S.
virginiae, and evidence of incorporation was sought in the 3-hydroxypicolinic acid residue
of virginiamycin S1. The 13C—NMR spectrum of the labeled virginiamycin S1 showed an
intense peak at 139.6 ppm due to the labeled C-6 in the 3-hydroxypicolinic acid residue.
Because the one·bond coupling constants are very small between carbon and nitrogen in
aromatic systems,115 it was not expected that any splitting would be observed even if the
nitrogen adjacent to this carbon were labeled. Examination of the peak under high
36
lCO;H CHO—§-•—»
I H;N CO;H H;N/[CO;H
L·¤sp¤rti¢ Mid L-aspartic acid semialdehyde
CHOÄ
+ Ä 1-1-—>ONll
HO;CY\/YCO;H—-1——> ———-——>
Hqzg \N CO;H NH2 NH2
HO;C\/\/YCO;H-—-§-—>
NH;
L-lysine
Scheme 6. The 2,6-diaminopimelate pathway of lysine biosynthesis
37
resolution conditions showed the peak to be broad, with evidence of a shoulder (Figure 6).
During the same GC/MS experiment in which doubly labeled 4-oxo-L-pipecolic
acid was studied, we were able to determine the isotopic labeling pattern of 3-
hydroxypicolinic acid. The antibiotic was hydrolyzed, and the resulting amino acids were
derivatized as the N-trifluoroacetyl butyl esters. Under the reaction conditions 3-
hydroxypicolinic acid was esterified; because the nitrogen is part of the aromatic system, it
is not acylated; neither is the phenolic oxygen. The base peak in the mass spectrum of the
. butyl ester of 3-hydroxypicolinic acid is at m/z 95, which corresponds to the protonated ion
in which the butoxycarbonyl group has been lost. The results are shown in Table 5.
The table shows that 3% of the ions for this peak are singly labeled, and 6% of the
peaks are doubly labeled. The results indicate that the mechanism of ring closure of lysine
to the heterocyclic ring proceeds with substantial loss of the ot-nitrogen and retention of the‘
8-nitrogen. If the conversion of lysine to 3—hydroxypicolinic acid goes by way of a
pipecolic acid intermediate as has been shown in etamycin,9‘ the mechanism would be
quite similar, up to a point, to that of the formation of4-oxo-L-pipecolic acid.
The second question posed about the mechanism of the formation of 3-
hydroxypicolinic acid in virginiamycin S1 concems the stereospecificity involved in the
removal of hydrogens in the aromatization process. There are three carbons of lysine that
retain hydrogens when it is transformed into 3·hydroxypicolinic acid: C-4, C-5, and C-6.
We chose to examine the process at C-5 by administering lysines stcreospecifically
deuterated at this carbon and observing the incorporation ofeach isomer.
Because these labeled lysines are not commercially available, it was necessary to
devise a synthesis of lysine by which C-5 could be labeled stereospecitically with
deuterium. A synthesis of stereospecifically deuterated ornithine by Townsend and
coworkers116·117 provided an intermediate that could be converted into the required
compounds. The synthesis of (2RS, 5R)·lysine-5-611 is outlined in Scheme 7. Deuterated
38
139.8 139.6 _. 139.4 _
(ppm)
Figure Partial 13C·NMR spectrum of VS; labeled with (2RS)-lysine-6-13C·6-15N: the signal at 139.6 ppm
39
Table 5. Relative abundance of the ions for the base peak (M-COOBu+H)
of butyl 3-hydroxypicolinate from VS; Iabeled with (2RS)-lysine-6-13C-6-15N
87ll 8 7L¤b¤1¢d 12-3 7.8
88888888 190 8.8 lCorrected 1abeled* 100 3.5 6.2
8-- 91 8 Ä* Relative abundances of the ions from the labeled compound corrected for the naturally
occurring A+1 and A+2 contributions of the unlabeled and singly labeled peaks.
* * Mole % unlabeled, singly labeled, and doubly labeled species, respectively
40
. 0 OCO2Et
iCÖ2Et CHO
CO2Et CO2Et
O - O’ 23
° ¤CO2Et
ii. iii.CO2Et
O24
° o ° ¤CO2Et • CÖQEÜ g
COQEI . CO2Et
O 0
25 26-S
O DCO2Et
·
C02Et
O27-S
O DCO2Et D
vii,COZEI
ONH2
28-R 29-R
Scheme 7. The synthesis of (2RS, SR)-lysine-5-dl: i. CH2=CHCHO,
NaOMe; ii. NaBD4; iii. PCC; iv. (+)—oz-pinene-9-BBN; v. MsC1, Et3N; vi.
KCN, DMSO; vii. Hg, PtO2; H3O+
41
aldehyde 25 was obtained by the reaction of diethyl phthalimidomalonate with acrolein,
reduction with sodium borodeuteride, then oxidation with pyridinium chlorochromate.
(Use of transition metal oxidants is known to produce large primary kinetic
isotopes.)110110 The aldehyde obtained in this manner was approximately 85-90%
labeled with deuterium (as determined by its NMR spectrum.) Reduction of the labeled
aldehyde with Midland's reagent, (+)-ot·pinene-9-borobicyclo[3.3.l.]nonane (9-BBN),
afforded the (S)·alcohol 26-S. Reduction of aldehydes with this reagent provides the
chiral alcohol with an enantiomeric excess approaching 100%.120 (R)-Nitrile 28-R was
obtained by displacement of the mesylate by cyanide ion. Hydrogenation of the nitrile over
PtO2 followed by acid hydrolysis afforded (2RS, SR)-lysine-5-d1(29·R). The (S)-
deuterated amino acid (29-S) was obtained in the same marmer, except the reduction of the
deuterated aldehyde was effected by (—)-ot—pinene—9-BBN.
Virginiamycin S1, obtained from fermentations in which each of the labeled lysines
was administered, was analyzed by GC/MS in the same manner as before. The results of
this analysis are shown in Table 6. The results show that the deuterium from (2RS, 5R)-
lysine-5-d1 is incorporated into the 3-hydroxypicolinic acid portion of the antibiotic,
whereas that from (2RS, 5S)-lysine-5-dl is not. The 5-pro(S) hydrogen of lysine is
therefore lost in its transformation. If the mechanism proceeds by way of 5-hydroxylysine
as has been implicatedlin the biosynthesis of etamycin,04 then the 5-pro(R) hydrogen is
retained when C-5 is hydroxylated.
One question that may arise is whether both the D and the L isomers of lysine are
incorporated. We know that the L-lysine is incorporated. If the D isomer is incorporated, it
is likely that it would be isomerized to the L configuration, as enzymes capable of
interconverting D- and L-amino acids are widely distributed in rnicroorganisms. Lysine
racemases have been found in Proreus vulgaris and a number of other microorganisms.121
There are a number of interesting experiments that would further elucidate this
T42
Table 6. Relative abundance of the ions for the base peak (M—COOBu+H)
of butyl 3Jhydroxypicolinate from VS; labeled with (2RS,5R)-lysine-5-d;
(I) and (2RS,5S)lysine·5-d; (II)
1..6 l
161661661 100 14.11161666166 100 8.8
5.6666
_
II.
,.6161661611 100 7.91161666166 100 8.8
Corrected labe1ed* 100 -0.9
666100*Relative abundances of the ions from the labeled compound corrected for the naturallyoccurring A+1 contribution of the unlabeled peaks.
* * Mole % unlabeled and labeled species, respectively
43
interesting process in virginiamycin S1. Lysines stereospecifically deuterated at C·4 and C-
5 could be incorporated. Studies like those done on etarnycin could be done in
virginiamycin S1 to establish whether S-hydroxylysine and S-hydroxypipecolic acid are
precursors. 5-hydroxylysine is incorporated, then (SR)- and (SS)·hydroxylysines could
be fed to determine which isomer is the preferred precursorr Similar experiments could be
performed with labeled S·hydroxypipeco1ic acid. A positive result from that experiment
coupled with the information obtained from the present study would answer the question of
whether the hydroxylation of lysine occurs with retention or inversion of the configuration
of the S-pro-(R) hydrogen that is retained from lysine.
44
2.3. The Origin of the L-Phenylglycine Residue
The mechanism of the conversion of L-phenylalanine to L·phenylg1ycine has not
been previously studied. However, biosynthetic studies of the N-methyl derivative of L-
phenylglycine, L-phenylsarcosine, were undertaken in the virginiamycin—like antibiotic
etamycin.92 In etamycin, L-phenylsarcosine arises from L·phenylalanine with the loss of
the carboxyl group and the preservation of the remainder of the carbon skeleton without
rearrangement. This is in contrast with the biosynthesis of tropic acid (30), a component
of the tropane alkaloids hyoscyamine and scopolamine. Tropic acid also originates from L-
phenylalanine, whose carboxyl group migrates to the prochiral C—3 position with retention
of conüguration.122
|-| $H;, _ H. $COOH_ ,_ , c¤-szonNH;
30
Biosynthetic studies on related systems provide insight into possible mechanisms
for the transformation of L-phenylalanine to L-phenylglycine. For example, the B-lactam
antibiotic nocardicin A contains L·p·hydroxyphenylglycine, which has been shown to be
derived from L-tyrosine with clean loss of its carboxyl group.123 Townsend and Brown
report that labeled ß·hydroxytyrosine and p-hydroxymandelic acid are incorporated into
nocardicin A,124 which suggests that tyrosine is hydroxylated to [5-hydroxytyrosine or, if
transamination has already occurred, the corresponding ß-hydroxyketo acid. Oxidative
decarboxylaüon would then produce p-hydroxymandelic acid; finally oxidation and
transamination would yieldp—hydroxyphenylglycine.
The antibiotic vancomycin, produced by the organism Szrepromyces oriemalis,
contains p-hydroxyphenylglycine and m-chloro-ß-hydroxytyrosine residues, both of which
45
are derived from tyrosine.125 The related antibiotic avoparcin has a similar origin.126 The
presence of the ß—hydroxy compound in these systems adds credence to Townsend's
argument that tyrosine is initially hydroxylated in the biosynthesis of these amino acid .
residues. INone of the above studies broached the question of the nitrogen migration that
occurs in the biogenesis of L-phenylglycine and related compounds. However, the 1,2-
migration of nitrogen in the formation of some ß·amino acids from the corresponding ot-
amino acids provides some suggestions for possible mechanisms for this process'. In the
antibiotics edeine A and edeine B}27 produced in cultures of Bacillus brcvis, the ß-
tyrosine residue is formed by the action of an 0t,ß-mutase on tyrosine with the loss of the
3-pro(S) hydrogen. The original ot-nitrogen is also lost, as is any labeled hydrogen at the
2-position; this Suggcsts an interrnediate in which a Schiff base is formed between the
amino group of tyrosine and a carbonyl group at the active site of the enzyme. Whereas the
loss of the 3·pro(S) hydrogen shows a relationship to the mechanism of the ammonia-lyase
enzymes (one of which, for example, catalyzes the formation of cinnarnic acid from
phenylalanine), the latter observation is inconsistent with that kind of mechanism because
an1rnonia·lyase reactions do not involve the exchange of ot·hydrogens.
The conversion of lysine to ß-lysine, a component of the antibiotic streptothricin F,
produced by Strepzomyces L-1689-23, has been shown to occur by the intramolecular
migration of the ot-nitrogen to C-3 with inversion of configuration.128 There is a
concomitant migration of the 3-pro(R) hydrogen to C-2, also with inversion of
configuration, but by an intermolecular process. The same transformation occurs in
Clostridium subterminale by the enzyme 2,3-aminomutase.129 Two possible mechanisms
are shown in Scheme 8.
The aspect of the transformation of L-phenylalanine to L-phenylglycine thatIparticularly interested us was the migration of the nitrogen. The process could be an
46
Enz-x0O “ H
NHm-12N: b
\0
I \op
N1a \b
ENz-x|.g ENZ·XH
00IE \ 0+
0 0I \
opI \
op1 1N0
H NHo/II\I/I\/\/ ’
I NNM\o
I \op
1N
ß-lysine
Scheme 8. Possible mechanisms for the conversion of lysine to ß-lysinelzs
47
intermolecular process whereby the original nitrogen is lost—a number of mechanisms can
be envisioned for an intermolecular process. Some of these are outlined in Scheme 9. In
pathway A, phenylalanine is hydroxylated to ß-hydroxyphenylalanine, which is degraded
to mandelic acid. Phenylglyoxylic acid would arise from further oxidation, and finally
transamination would yield phenylglycine. Phenylalanine, in pathway B, would first
undergo transamination to phenylpyruvic acid, then oxidative decarboxylation to yield
mandelic acid. In the last example, pathway C, phenylalanine would be decarboxylated toB
phenylethylamine, then hydroxylated, and oxidized to yield mandelic acid. If the nitrogen
were transferred by an intramolecular process, a mechanism analogous to that proposed for
the conversion of lysine to ß·lysine (Scheme 8) could be operating.
We undertook then to determine whether the mechanism proceeds by an inter- or
intramolecular mechanism. To this end, we planned to administer phenylalanine-3-BC-
BN (41) to S. virginiac. If the mechanism were an intramolecular process, then the
labeled nitrogen would be retained in the phenylglycine molecule, bonded to the labeled
carbon. In this case, BC—BN coupling of a few Hertz would be observed in the BC-
NMR spectrum of the labeled virginiamycin S1. If the mechanism were intermolecular, on
the other hand, no coupling would be seen.
The doubly labeled phenylalanine was synthesized by minor modification of
classical methods (Scheme 10). Benzoic-carboxy-BC acid, prepared from phenyl
magnesium bromide and BCO; obtained from labeled barium carbonate, was reduced and
reoxidizedB° to afford benzaldehyde-carboxy-BC (31). The labeled benzaldehyde was
condensed with N·acetylglycine-BN to produce azlactone 33.Bl Hydrolysis of the
azlactone produced doubly labeled ot-acetamidocinnamic acid (34), which was
subsequently hydrogenated and hydro1yzedB2 to yield the desired DL-phenylalanine-3-
BCJ5N hydrochloride (35). ~
48
CO;H
B c· A
NH;
OH
co,uNH;
ou/
_
O@)Lco,u
NH;
Scheme 9. Possible mechanisms for the conversion of phenylalanine to
phenylglycine
49
I ‘°c•-no (ww+
@/lßuncocu,
31 32
O
a. aa.OCH;
33
?°m-ncocr-1, '!°u¤-1,cu
34 35
Scheme 10. Synthesis of DL-phenylalanine-3-UCJSN: i. Na0Ac, Ac2O; ii.H20, acctone; iii. H2, 5% Pd/C; H;0+
50
Following fermentation in a medium enriched with the labeled phenylalanine,
virginiamycin S1 (7.6 mg) was isolated. Its 13C·NMR spectrum showed intense
resonances at 36.7 ppm and 56.1 ppm due to C·3 of the N-methyl-L-phenylalanine residue
and C-2 of the L-phenylglycine residue respectively. Examination of the peak at 56.1 ppm
under high resolution conditions showed no splitting due to coupling to 15N (Figure 7).
Because it is possible that the nitrogen could be washed out by transamination, as
either phenylalanine or phenylglycine, failure to detect labeled nitrogen does not in itselfI
prove that the process of interest proceeds by an intermolecular process. Fortunately the
virginiamycin molecule contains an internal standard (N·methyl·L·phenylalanine) that can
tell us whether any transamination occurs in phenylalanine (although it is not possible to
determine to what extent it might occur in phenylglyeine). Mass spectrometry was able to
give us the required information. The labeled virginiamycin S1 was hydrolyzed, and the
resulting amino acids were converted to their N-tritluoroacetyl butyl ester derivatives,
which were analyzed by GC/MS. The base peak for N-trifluoroacetyl—N-methyl-L-
phenylalanine butyl ester is due to the ion produced by the loss of the butoxycarbonyl
group. The data in Table 7 shows that this ion (therefore the intact amino acid) consists of
14% singly labeled and 8% doubly labeled species. It is apparent that some transamination
has occurred, but if comparable transamination had occurred in phenylglycine and
rearrangement were intramolecular, it would be possible to see splitting in the peak at 56.1
ppm. In addition, the derivative ofphenylglycine was observed in the GC/MS experiment;
all of the ions of the base peak (due also the the loss of the butoxycarbonyl group) were
either unlabeled (86%) or singly labeled (14%). (See Table 7.) It has been shown,
therefore, that the conversion of L-phenylalanine to L-phenylglycine proceeds by an
A intermolecular mechanism.
51
.gäiggeegggeg"'-
I·-
?$--„’
Tee? ·‘ Ä .=;- ..
_; iäiärgi
;.
56.0 56.1 56.2 56.3(ppm)
Figure 7. Partial BC-NMR spectrum of VS; labeled with DL-phenylalanine-3-UC-15N: the signal at 56.1 ppm
52
Table 7. Relative abundance of the ions for the base peaks (M—COOBu) of
N-trifluoroacetyl-N-methyl-L-phenylalanine butyl ester (I) and N-trifluoro-
acetyl-L·phenylglycine butyl ester (II) from VS; labeled with (2RS)-
phenylalanine·3-"CJ-‘N
1.mx: 230
2311.6186166100 31.2 13.6U8118b61681 100 13.0 1.2
comm 188661681* 100 18.2 10.08888 78 14 Ä
11.
Labßlßd 100 44.9 7.3U111881861611 ‘ 100 28.3 3.4
Corrected1abe1ed* 16.6 -0.7
8.88Ä 14 Ä* Relative abundances of the ions from tl1e labeled compound corrected for the naturally
occuuiug A+1 and A+2 contributions of the unlabeled and singly labeled peaks.
* * Mole % unlabeled, singly labeled, and doubly labeled specics, respectively
3. EXPERIMENTAL
3.1. General
Melting points were deterrnined on a Kofler block and were uncorrected.
Radioactivity was deterrnined by means of a Beckman LS—3800 or Beckman LS- 100 liquid
scintillation counter. The liquid scintillation cocktail was Beckman Readi-Solv MP or a
cocktail consisting of 2,5—diphenyloxazole (4 g), p-bis(2-(4-methyl-5-phenoxazolyl))-
benzene (0.2 g), and Triton X-100 (333 mL) in enough toluene to 1 L.K
The following chromatographic materials were used: for analytical thin—1ayer
chromatography (TLC), Merck silica gel 60 F254 on aluminum, 0.2 mm thiclcness; for
preparative scale TLC, Analtech Uni-Taper silica gel GF plates or Analtech silica gel GF
(500 pm); for column chromatography (flash), Merck silica gel 60 (230-400 mesh); for
ion-exchange chromatography, Dowex 50W (8% cross·linked, dry mesh 200-400) or
Dowex 2 (8% cross-linked, dry mesh 100-200) resin. The following system was
employed for high performance liquid chromatography (HPLC): a Waters Associates
M6000A pump, a Valco six-port injection valve, and a Waters Associates 440 absorbance
detector (at 254 nm). In some cases a Tracor 980 A solvent programmer was utilized. For
analytical HPLC, Waters Nova-Pak Radial-Pak columns (C8 or C18) were used with a
Waters Associates RCM-100 radial compression module, for preparative scale I—IPLC, a 30
x 1 cm column packed with LiChrosorb RP-8 (10 pm).
Nuclear magnetic resonance (NMR) spectra were deterrnined on a Bruker WP·270
or a Bruker WP-200 spectrometer. Mass spectra were obtained on a Finnigan—MAT 112
mass spectrometer or a VG Analytical 7070E mass spectrometer. The gas chromatographs
used in GC/MS analyses were a Varian 2100 (coupled to the Finnigan instrument) or a
Hewlett-Packard 5790A (coupled to the VG instrument). Infrared (IR) spectra were
deterrnined on a Perkin-Elmer 710B or a Perkin—Elmer 283B infrared spectrophotometer.
i53
541
3.2. Culture Conditionsd
The microorganism Strepzomyces virginiae strain 1830 was obtained from
SmithK1ine Animal Health Products, West Chester, Pennsylvania, as either a pellet or a
slant. The strain was maintained on either soluble-starch-agar or potato—g1ucose—agar
slants. The solub1e·starch-agar slants were prepared by combining com starch (10 g),
(NH4)2SÜ4 (2 g), KZHPÜ4 (1 g), MgSÜ4•7H2O (1 g), Nacl (1 g), and CaCÜ3 (3 g) in 1
L water. The pH was adjusted to 6.8 with 2 N HC1. Agar (20 g) was added, and the
mixture was dispensed into test tubes and sterilized i11 an autoclave. 30 min at 120 °C (15
psi). Each sterile slant was inoculated with a spore suspension (0.2 mL) obtained from
either a pellet or another slant and incubated 6-8 days at 25-28 °C. The slants were stored
at 4 °C until needed. The potato-g1ucose—agar slants were prepared as followsz Cut-up
potatoes (200 g) were boiled in water for 30 min and the mixture filtered through
cheesecloth. After glucose (10 g) was added to the filtrate, water was added to bring the
volume to 1 L, and the pH was adjusted to 7.4 with 1 N NaOH. Following the addition of
agar (20 g), the mixture was dispensed into test tubes, then sterilized and inoculated as
before.
The vegetative inoculum was prepared by transferring a spore suspension from a
slant to a baffled 250-mL Erlenmeyer flask containing 30 mL of medium STA-2.133 STA-
2 was prepared as fo11ows: A mixture of com-steep liquor (36.5 g) in 1 L tap water was
adjusted to pH 7.5 with NaOH. After the addition of peanut-oil cake (8 g), the mixture
was boiled for 2 min, then filtered. Glucose (50 g), MnSO4 (0.01 g), and CaCO3 (5 g)
were added. The medium was dispensed into baftled 250-mL Erlenmeyer tlasks and
· sterilized for 30 min at 120 °C. The inoculum was incubated at room temperature on a
rotary shaker (Lab-Line Orbit Environ-Shaker, Lab—Line Insuuments) at 330 rpm for 72 h.
— After this time, the vegetative inoculum was transferred to the production medium STA-
14133 in a fermentor (MultiGen 2-L fermentor, New Brunswick Scientific). Medium STA-
55
14 was prepared as followsz One liter tap water containing com-steep liquor (36 g) and
ycast autolysate (5 g) was adjusted to pH 7.9 with NaOH. Peanut·oi1 cake (10 g) was
added; the mixture was boiled for 2 min, then filtered. Linseed oil (10 g), glucose (5 g),
glycerol (25 g), and CaCO3 (5 g) were added. The mixture was transferred to the
fermentor vessel and sterilized in an autoclave for 45 min at 120 °C (15 psi). The broth
was aerated at 1.25 L/min and stirred at 450 rpm. Precursors were added after 8-10 h,
introduced through a 0.45 um filter. Fermentation was continued for a total time of 48 h.
In some runs, pH was maintained between 6.5 and 6.8 by the addition of sterile 1 N
NaOH.
3.3. Isolation of Virginiamycin S1
Virginiamycin S1 was isolated from the culture broth after the completion of the
incubation period. The broth was filtered with the aid of Hyflo Super Cel in order to
remove the mycelia, then the filtrate was extracted twice with a one—third volume of hexane,
then three times with a one-half volume of ethyl acetate. The ethyl acctatc layers were
combined, washed with a one-half volume of water, dried over MgSO4, and evaporated at
reduced pressure. Virginiamycin S1 was isolated by flash chromatography with CHCI3-
MeOH (99: 1) and shown to be >90% pure by analytical I—IPLC.
3.4. Analysis of Virginiamycin Sl Amino Acids by GC/MS
Virginiamycin S; to be analyzed by GC/MS was hydrolyzed in 1 rnl.. 6 N HC1 at
105 °C for 24 h. N-Trifluoroacetyl amino acid butyl esters of the virginiamycin S1 amino
acids were prepared in the following manner: After evaporation to dryness, the residue
was dissolved in dry 3-4 N HC1 in 1-butanol (prepared either by bubbling dry HC1
through dry 1—butanol or by adding acetyl chloride to 1-butanol) and heated at 110 °C for 3
56
h (15 h in some cases). The excess reagent was evaporated under a stream of N2, then 1
rnL dichloromethane and 1 mL tritluoroacetic anhydride were added. The reaction mixture
stood at room temperature for 24 h, then evaporated to dryness under a stream of N2. The _
residue was dissolved in a minimal amount of dichloromethane for analysis by GC/MS.
Derivatives of unlabeled amino acids (25-40 mg of each amino acid of interest) to
use as standards were prepared in the same way. The derivative of 3-hydroxypicolinic acid
was prepared on a larger scale: 3-Hydroxypicolinic acid (110 mg, 0.79 mmol) in 2 mL 4
N HC1 in dry 1·butanol was heated ovemight at 110 °C. There appeared to be a
considerable amount of starting material remaining at the end of this period The rnixture
was evaporated to dryness; the residue was dissolved in 2 mL distilled water and brought
to pH 6 with NaOH. The product (28 mg, 15% yield) was isolated by extraction of the
aqueous mixture with an equal volume of ethyl acetate, evaporation to dryness, then flash
chromatography (5 g silica gel; ethyl acetate-hexane, 6:4). The ester was dissolved in 1
mL dichloromethane and 1 rnL trifluoroacetic anhydride and allowed to stand for 23 h.
Only the ester was detected (TLC; ethyl acetate-hexane, 6:4) at the end of the reaction time.
Butyl 3-hydroxypicolinatez IH NMR (CDC13) 5 0.98 (t, 3H, J = 7.4 Hz), 1.48 (m, 2H),
1.86 (m, 2H), 4.78 (t, 2H, J = 6.9 Hz), 7.40 (m, 21-I), 8.30 (dd, 1H, J = 1.8 Hz, J = 4.0
Hz), 10.80 (s, 1I-I).’
3.5. Incorporation of L-Methionine-methyl-I‘C
S. virginiae 1830 was grown in media STA-2 and STA- 14 as previously described.
Eight hours after the production medium STA-14 was inoculated, 27 },tCi L-methionine-
merhyl-MC (New England Nuclear; specific activity, 45.7 mCi/mmol) in 4.4 mL sterile
0.01 N HC1 was added to the fermentation medium. The pH of the medium was
maintained in the range of 6.5-6.8. After 48 h, the fermentation was halted, and
57
virginiamycin S1 was isolated as previously described. The antibiotic was further purified
by HPLC (acetonitrile—water, 1:1; flow rate, 6 mL/min). Pure virginiamycin S1 (5.9 mg)
was obtained in this manner. A portion (10%) of the antibiotic was counted. The specific
activity of the virginiamycin S; was detemtined to be 1.1 x 10*2 mCi/nunol. The specific
incorporation was then 2.4 x 10‘2. A portion of the antibiotic (3.2 mg) was hydrolyzed in
1 mL 6 N HC1 at 104 °C for 24 h. The hydrolysate was evaporated to dryness at reduced
pressure; excess HC1 was removed by storing the residue over NaOH pellets in an
evacuated desiccator overnight. The residue was dissolved in 1 mL water and placed on a
column (1 x 44 cm) containing Dowex·50W ion-exchange resin (H+ form). The column
was eluted with 1 N HC1; 1-mL fractions were collected The fractions containing 4·oxo—L-
pipecolic acid (52-64) were combined and evaporated to dryness. The isolated compound
showed a single spot on TLC and co-chrornatographed (R5 = 0.41) with authentic 4·oxo—L-
pipecolic acid hydrochloride (1-propano1—water, 7:3). It had a specific activity of 1.9 x
10** mCi/m1no1 (on the assumption that 100% of the amino acid was recovered); 1.7% of
the total radioactivity ofvirginiamycin S1 was located in the 4-oxo-L-pipecolic acid residue.
3.6. Incorporation of L-Aspartic-U-1"C Acid
This experiment was performed in a similar manner to the one previously described
for the incorporation of L·methionine·merhylJ"C. L-Aspartic-UJ"C acid (ICN, 49.8 uCi;
specific activity, 184 mCi/mmol) and 0.18 g L-threonine (added to enhance production)
were dissolved in 5.0 1nL 0.01 N HC1 and added to medium STA-14 8 h after inoculation.
Virginiamycin S1 (3.6 mg) was isolated by flash chromatography as previously described
and had a specific activity of 1.1 x 10*3 mCi/mmol. Specific incorporation was 5.9 x 10**.
4-Oxo-L-pipecolic acid obtained as before had a specific activity of 7.9 x 10*5 mCi/mmol;
7.1 % of the radioactivity of the labeled antibiotic was found in the 4-oxo·L·pipeco1ic acid
residue.
58
3.7. Incorporation of L-Lysine·U-MC
The incorporation ofL-lysine-U·“C was undertaken in a similar manner to those of
the other radiolabeled precursors. L·Lysine-UJ"C•HCl (ICN, 49.8 ttCi; specific activity,
275 mCi/mmol) and L-threonine (0.10 g) in 4.5 mL 0.01 N HC1 were added to medium
STA-14 8 h after inoculation. Virginiamycin S1 was isolated by flash chromatography as
previously described. It had a specific activity of 2.7 x 10*2 mCi/mmol; specific
incorporation was 9.8 x 10*3.
3.8. Incorporation of DL-Lysine-6-13C-6-ÜN
The fermentation and preliminary puritication of the antibiotic took place at the
laboratories of SmithKline-RIT, Rixensart, Belgium. S. virginiae 5722 was grown in a
vegetative medium and transferred to a production medium (40 mL in each of 29 250-mL
Erlenmeyer flasks) after 2 days. After 24 h, 200 p.L of a solution of 65 mg DL-lysine-6-
ÜC-6-ÜN,1°"prepared from NaÜCÜN (Prochem, BOC Limited; 90% ÜC, 99% ÜN;
the intermediate ethyl 2·carbethoxy-2-phthalimido-5-cyanopentanoate-6J3C—6-ÜN was
determined by MS to be 76.1% doubly labeled and 7.8% singly labeled) and 100 mg L-
threonine in 6 mL distilled water, sterilized by passing through a Sartorius membrane (0.22
um), was added to each tlask
After 3 days, the broth was acidiiied to pH 4.8 with 10% HZSO4, then extracted
three times with methyl isobutyl ketone. The organic extracts were combined and
evaporated to dryness. The residue was dissolved in acetonitrile; the solution was washed
twice with n-hexane and evaporated to dryness. The new residue was dissolved in 4 mL
chloroform. Crude virginiamycin (672 mg) was precipitated by the addition of 40 mL rz-
hexane, then filtered. HPLC analysis indicated that the crude material contained
59V
approximately 360 mg virginiamycin M2, 80 mg virginiamycin M1, 16 mg virginiamycin
Mg, and 41 mg virginiamycin S1.
Virginiamycin S; was isolated from the mixture as previously described. One-third
of the antibiotic was hydrolyzed by heating at 105 °C in 1 mL 6 N HC1 for 24 h. N-
Trifluoroacetylamino acid butyl esters were prepared as previously described. Because the
amount of the proline derivative was much greater than that of the 3-hydroxypicolinic acid
derivative, and because separation of the two peaks by GC was unsatisfactory in this case,
the mixture of amino acid derivatives was partially purified by preparative TLC (ethyl
acetate-hexane, 1: 1). The partially purified mixture contained both the 3-hydroxypicolinic
acid and 4-oxo-L-pipecolic acid derivatives, but very little of the proline derivative. This
mixture was separated by GC (10% SP2100, 6 ft x 2 mm ID, 80 °-300 °C at 10 °/min, or
3% OV1, 3 ft x 2 mm ID, 75 °-200 °C at 10 °/min); the peaks due to butyl N·t1·ifluoroacety1-
4-oxo·L·pipecolate and butyl 3-hydroxypicolinate were analyzed by MS. The isotopic
composition of the major fragrnent ion of each derivative was determined on an average of
4-10 spectra by the method of Beimann.l3"
3.9. Synthesis of 4-Oxo-DL-pipecolic Acid Hydrochloride
2-Cyano-4-piperidone Ethylene Ketal (12). To a stirred slurry of N-
chlorosuccinimide (5.15 g, 38.6 mmol) in freshly distilled dry ether (100 rnL) was added
dropwise 4-piperidone ethylene ketal (5.01 g, 35.0 mmol). The mixture was stirred at
room temperature under N2 atmosphere for 17.5 h and then tiltered. The tiltrate wasu
evaporated at reduced pressure to give a colorless oil containing some solid material. An
additional portion of ether was added to the residue, and the mixture was filtered.
To the filtrate was added 18·crown-6 (0.107 g, 0.4 mmol) and potassium
superoxide (4.99 g, 70.0 mmol). The slurry was stirred at room temperature for 24 h and
then filtered. The residue remaining on the filter was washed with a small portion of ether.
60
The combined filtrate and washings were extracted twice with 100 ml 0.5 N HC1.
A solution of potassium cyanide (11.4 g, 0.175 mol) in 100 ml water was cooled
on an ice-water bath, and concentrated HC1 was carefully added to bring the solution to pH
7. The aqueous extract was added to the acidified potassium cyanide solution over 0.5 h,
after which time the pH of the reaction mixture was adjusted to 5 with potassium hydroxide
pellets, and stirring was continued for 7 h, with warming to room temperature. The
mixture was cooled in an ice-water bath, and potassium hydroxide pellets were added to
adjust the solution to pH 9. The mixture was extracted with ethyl acetate (3 x 200 mL).
The combined extracts were dried over MgSO4. Evaporation yielded a pale yellow oil,
which after flash chromatography (CHC13-MeOH, 9:1) afforded 2.75 g of the product
(47% overall yield). IH NMR (CDClg) 5 1.72 (t, 2H, J = 5.6 Hz), 1.78 (NH, br s, 1
H), 1.96 (m, 2H), 2.93 (m, 2H), 3.16 (m, 1H), 4.01 (m, Sl-I); I3C NMR (DMSO) 6
34.9, 37.1, 41.4, 44.4, 63.6, 63.8, 105.4 (OCO), 120.6 (CN); IR (neat) 895, 935,
975, 1025, 1140, 1235, 1290, 1355, 2220, 2885, 2930, 3300 cm‘I; mass spectrum m/z
(relative abundance) 168 (M+) (63), 141 (21), 126 (10), 99 (32), 96 (8), 87 (34), 86
(100).
4-Oxo-DL-pipecolic Acid Ethylene Ketal (13). A mixture of 2-cyano-4·
piperidone ethylene ketal (0.166 g, 0.99 mmol), barium hydroxide octahydrate (0.176 g,
0.56 mmol), and water (2 mL) was heated at 95 °C for 2.5 h. After the addition of more
water (2 mL), small chunks of dry ice were added to the reaction mixture, with the
temperature maintained at 90-95 °C. The mixture was stirred for 15 min, and it was then
filtered. Water (2 mL) was added to the residue, and this was heated for 10 min, then
filtered. The combined filtrates were heated with Norite and filtered. The filtrate was
evaporated to give a colorless crystalline residue (0.14 g, 74%). The product was
recrystallized from 2·propanol with a trace of water and ethyl ether. MP 246 IH NMR
61
(D20) 5 1.78-1.64 (m, 3H), 2.09 (m, IH), 2.93 (m, IH), 3.29 (dt, IH), 3.56 (C·2, dd,
IH), 3.84 (0CH2, m, 4 H); HC NMR (D20, dioxane = 66.5) 6 30.7, 34.8, 41.1,
57.3, 64.6, 104.9 (0C0), 172.8 (COOH); IR (KBr) 1050, 1105, 1185, 1405, 1600,
2950, 3400cm‘I. As the hydrochloridez mp 210-215 °C; IH-NMR (D20) 8 1.72-1.89
(m, 3H), 2.18 (m, IH), 3.01 (dt, IH), 3.35 (m, II-I), 3.85 (OCH2, m, 4H), 3.94 (CH,
dd, IH); HC-NMR (D20, dioxane = 66.5) 6 30.7, 33.9, 41.2, 55.5, 64.7, 104.3
(OCO), 170.7 (COOI-I); IR (KBr) 1050, 1110, 1190, 1380, 1405, 1560, 1760, 2800 (v
br) cm‘I; mass spectrum m/z (relative abundance) 187 (M+-HCI) (0.4), 142 (54), 112
(8), 99 (I2), 87 (100), 86 (II), 56 (56).
4-Oxo-DL-pipecolic Acid Hydrochloride (14). 4-0xo-DL-pipecolic acid
ethylene ketal hydrochloride (0.16 g, 0.72 mmol) was heated in 5 mL 2 N HC1 at 85 °C for
15 h. The mixture was neutralized with 6 N Na0H; then placed on a Dowex 2 (0H‘ form)
column (2 x 14 cm.) The coltunn was washed with water (100 rr1L) to remove the sodium
ions, then eluted with I N aqueous acetic acid. The fractions containing 4-oxo-DL-
pipecolic acid (TLC: 1-propanol-water, 7:3) were combined along with 1 mL 6 N HCI and
evaporated to dryness to afford 38 mg (30%) of yellowish product, which showed one
spot on TLC and co-chromatographed with authentic material (R; = 0.4). IH NMR (D20)
5 1.68-1.88 (m, 3H), 2.21 (m, IH), 3.00 (dt, IH), 3.29 (dt, IH), 3.89 (C·2, dd, IH).
3.10. Synthesis of Virginiamycin S1 Amino Acid Derivatives for HPLC
N-Benzoyl derivatives of virginiamycin S1 amino acids were prepared according to
standard Schotten-Baumann conditions (bcnzoyl chloride in ether, 1 N Na0H) or
according to a procedure adapted from Pirkle (bcnzoyl chloride, TI-IF, propylene
oxide).H5
IN-Benzoyl-L-threonine. L-Threonine (0.46 g, 3.9 mmol) was dissolved in I
62
N NaOH (10 n1L). Ether (10 mL) containing 0.50 mL benzoyl chloride (0.61 g, 4.3
mmol) was added, and the two-phase mixture was vigorously stirred for 24 h at room
temperature. The ether was evaporated, and the aqueous layer was acidified with 6 N HC1,
then extracted with ethyl acetate (2 X 20 mL). The organic extract was dried over MgSO4
and evaporated at reduced pressure. The residue was recrystallized from ethyl acetate to
yield 0.33 g (38%) product, mp 148-149 °C (literature 146-148 °C).13‘
N-Benzoyl-DL·ot-aminobutyric acid. DL-ot—Aminobutyric acid (0.42 g, 4.0
mmol) was dissolved in 1 N NaOH (10 mL). Ether (10 mL) containing 0.5 mL benzoyl
chloride (0.61 g, 4.3 mmol) was added, and the two-phase mixture was vigorously stirred
for 24 h at room temperature. The ether was evaporated, and the aqueous layer was
acidified with 6 N HC1, then extracted with ethyl acetate (2 X 20 mL). The organic extract
was dried over MgSO4 and evaporated at reduced pressure. The residue was recrystallized
from ethyl acetate-hexane to give 0.42 g (51%) product, mp 143-144 °C (literature 145-
146 °c).¤7
N-Benzoyl-L-proline. To a slurry of L-proline (0.50 g, 4.3 mmol) in dry TI-IF
was added 0.50 mL benzoyl chloride (0.61 g, 4.3 mmol) and 0.90 mL propylene oxide
(0.75 g, 13 rnrnol). The mixture was stirred under N2 at room temperature until the solid
proline had disappeared (45 min), then for an additional 45 mir1. The reaction mixture was
filtered with suction to remove any unreacted proline, then evaporated at reduced pressure
to yield a clear oil (1.12 g) that crystallized on standing. The crude product was
recrystallized from ethanol-petroleum ether to give 0.23 g of the product (mp 154-156 °C,
literature 156 °C).m The mother liquor yielded an additional crop (0.41 g, mp 153-157
°C) for a total yield of 68%.
N-Benzoyl-N·methyl-L-phenylalanine. To a slurry of N-methy1-L-
63
phenylalanine (0.125 g, 0.70 mmol) in dry Tl-IF (5 mL) was added 81 uL benzoyl chloride
(98 mg, 0.7 mmol) and 150 1tL propylene oxide (0.125 g, 2.1 mmol). The reaction
mixture was stirred at room temperature in a flask surmounted with a drying tube for 24 h, ‘
after which time no solid remained. Evaporation of the solvent left 0.21 g crude product,
which was subsequently recrystallized from HOAc-water to give 0.108 g (54.5%) of the
derivative, mp 136-138 °C.l39
N-Benzoyl·L-pheny|glycine. L-Phenylglycine (0.45 g, 3.0 mmol) was
dissolved in 1 N NaOH (10 mL). Ether (10 mL) containing 0.40 mL benzoyl chloride (48
g, 3.4 mmol) was added, and the two-phase mixture was vigorously stirred for 24 h at
room temperature. The ether was evaporated, and the aqueous layer was acidified with 6 N
HC1, then extracted with ethyl acetate (2 X 20 mL). The organic extract was dried over
MgSO4 and evaporated at reduced pressure. The residue was recrystallized from ethyl
acetate to provide 0.36 g (47%) product, mp 187-188 °C (literature 192-193°C).1‘°
N-B¢nzoyI-4-oxo-DL-pipecolic Acid Ethylene Ketal (15). To a slurry
of 4-oxopipecolic acid ethylene ketal (13) (0.050 g, 0.27 mmol) in 3 ml dry Tl-IF was
added 60 },LL propylene oxide (0.0498 g, 0.86 mmol) and 40 p.L benzoyl chloride (0.048
g, 0.34 mmol). The mixture was stirred at room temperature for 24 h. After filtration, it
was evaporated under reduced pressure to yield a crude product still containing traces of
benzoyl chloride. The product was puriüed by preparative TLC to yield an oil (0.056 g,
72%), which had partially decomposed on the silica gel TLC plate.
3.11. Separation of Virginiamycin S1 Amino Acids by HPLC
A portion of virginiamycin S1 (up to 50 mg) was hydrolyzed in 1 rr1L 6 N HC1 at
105 °C for 24 h. The mixture was evaporated to dryness, then dissolved in 1 mL 1 N
NaOH. To this was added a solution of 1.1 equivalents benzoyl chloride in 1 mL ether.
64
The two-phase reaction mixture was vigorously stirred at room temperature for 24 h. The
ether layer was evaporated, and the aqueous mixture was acidified with 6 N HC1 and
exuacted with ethyl acetate. The aqueous layer contained 3—hydroxypicolinic acid, which
remained underivatized under the reaction conditions. The organic extract was evaporated
to dryness, redissolved in methanol or acetoniuile , and analyzed by HPLC.
'I'he amino acid derivatives were separated on a C18 colunm by means of a gradient
solvent system. The solvent compositions were as followsz solvent A, water-methanol—
'I'I-IF-formic acid, 85:11.5:2.5:1; solvent B, water-methanol—THF-formic acid, 58:40:1:1.
The solvent programmer was programmed for an delay of 4 min with solvent A, then
a linear gradient (10% per min) until the solvent passing through the system consisted only
of solvent B.
3.12. Synthesis of (2RS, SR)-Lysine-5-d; Dihydrochloride
4,4-Dicarbethoxy-4-phthalimidobutanal (23).117 To a solution of diethyl
phthalimidomalonate (6.11 g, 20 mmol) in benzene (30 mL) was added 0.057 g sodium
methoxide (1.1 mmol), and the resulting lemon-yellow solution was cooled to 0 °C in an
ice-water bath. A solution of üeshly distilled acrolein (1.5 mL, 22 mmol) in benzene (5
mL) was added dropwise to the mixture over 0.5 h, with stirring under N2 atmosphere.
After the addition was complete, the ice-water bath was removed, and stirring was
continued for 1 h. The reaction was quenched by the addition of 4 drops of glacial acetic
acid. Evaporation of the benzene from the mixture at reduced pressure left a turbid straw-
colored oil. Flash chromatography (ethyl acetate—hexane, 3:7) provided 3.89 g (89%
yield) pure aldehyde 23 as a colorless oil: Rf = 0.31 (ethyl acetate—petroleum ether, 6:4);
1H·NMR (CDCI;) 6 1.29 (CH;;, t, 6H, J = 7.1 Hz), 2.72 (m, 21-I), 2.86 (m, 2H), 4.32
(OCH;, q, 2H, J =7.l Hz), 4.33 (OCH;, q, 2 H, J =7.l Hz), 7.81(Ar, m, 4H), 9.71
65
(CHO, t, 1H, J = 0.9 Hz); 13C·NMR (CDC13) ö 13.6, 25.3, 39.0, 62.8, 66.8, 123.4,
134.4, 165.8, 167.2, 200.5 (CHO).
Ethyl 2-carbethoxy-2-phthalimido-5—hydroxypentanoate-5-d, (24).
Sodium borodeuteride (0.767 g, 18.3 mmol) was added in small portions over 3 h to a
vigorously stirred mixture of aldehyde 23 in ether (85 mL) and water (5 mL). The mixture
was stirred for an additional 0.5 h, then for another 15 min following the addition of 35 mL
water. The aqueous layer was separated from the ether layer and subsequently extracted
with additional ether (5 x 50 mL). The combined ether layers were dried over NaSO4 and
evaporated in vacuo. Flash chromatography (ethyl acetate—hexane, 1:1) afforded the
labeled alcohol as a colorless oil (3.43 g, 53%): 1H-NMR (CDCI3) 6 1.29 (CH;, t, 61-I,
J =7.l Hz), 1.69 (m, 2H), 2.58 (m, 2H), 3.63 (CHDOH, br t, 1H), 4.30 (CH;O, q, 2H,
J = 7.1 Hz), 4.31 (CHZO, q, 2H, J = 7.1 Hz), 7.81 (Ar, m, 4l-I); 13C—NMR (CDCI3) 5
13.7, 27.5, 29.5, 61.8 (CHDOH, t, J = 21.4 Hz), 62.5, 67.6, 123.4 (Ar), 131.3 (Ar),
134.3 (Ar), 166.2, 167.3.
4,4-Dicarbethoxy-4-phthalimidobutanal-1-d1 (25). To a mixture of
pyridinium chlorochromate (7.30 g, 3.9 mmol) and sodium acetate (1.39 g, 16.9 mmol) in
dry CH2Cl2 (45 mL) was added labeled alcohol 24 in dry CH2Cl;. An additional 30 mL
CH;Cl2 was used to rinse the tlask that had contained the alcohol. The dark brown
mixture was stirred at room temperature under N2 for 4 h. Ether (100 mL) was added to
the mixture, and stirring was continued for 5 min. The mixture was filtered through a layer
of Florisil (3.5 x 6.5 cm) topped with a 1-cm layer of Hytlo Super Cel. Ether (100 mL)
and CH2Clg (100 mL) were added to the residue remaining in the tlask and filtered. The
filtrate was evaporated to give a pale yellow oil, which was purified by flash
chromatography (ethyl acetate—hexane, 3:7) to provide aldehyde 29 (4.33 g, 73%) as a
colorless oil. IH-NMR (identical to that of the unlabeled aldehyde except for the reduced
u66
intensity of the peak at 9.71 ppm) showed the aldehyde to be approximately 88-90%
labeled. IBC-NMR (CDCIS) 5 199.9 (CDO, t, J = 26 Hz), otherwise identical to the
unlabeled aldehyde 23.
Ethyl (55)-2-carbethoxy-2-phthalimido-5-hydroxypentanoate-5-dl
(26-S). A solution of (1R)-(+)-oz-pinene (Aldrich, 98% optical purity) (3.69 g, 27.1
mmol) in 49 rnL 0.5 M 9-BBN in THF (24.5 mmol) was heated at retlux with stirring
under N2 atmosphere for 4 h. The mixture was cooled and stored under N2 overnight.
Labeled aldehyde (25) was dissolved in 25 mL dry TI—IF and transferred by means of a
cannula to the pinanyl borane mixture. The reaction mixture was stirred under NS for 8 h,
then it was placed on a silica gel column (5.5 x 15 cm) and eluted with ether. The fractions
containing the alcohol were combined, and the solvent was removed in vacuo. The alcohol
was purified further by flash chromatography (ethyl acetate—petroleum ether, 1:1 then 6:4).
The chirally deuterated alcohol (3.78 g) was obtained in 84% yield: IH-NMR and I3C—
NMR spectra were identical to those of alcohol 24.
Ethyl (SS)-2-carbethoxy·2·phthalimido-5-methanesulfonyloxypentan-
oate-5-dl (27-S). To an ice-cold stirred mixture of the (S)-alcohol 26-S (3.78 g, 10.4
mmol) and triethylamine (2.11 g, 20.8 mmol) in 75 mL CH2Cl2 was added dropwise
methanesulfonyl chloride (2.40 g, 20.9 mmol) in 15 mL CHSCIS over 15 min. The ice-
water bath was removed, and stirring was continued 2.5 h. The reaction mixture was
evaporated at reduced pressure, and to the residue was added 100 mL of 6:4 mixture of
ethyl acetate and hexane. The liquid was decanted, and the residue was rinsed again with
portions of the ethyl acetate-hexane mixture (4 x 20 mL). The combined washings were
evaporated, and the residual yellow oil was purified by flash chromatography (ethyl
acetate-hexane, 6:4). The (S)—mesylate 27-S (4.11 g) was obtained in 89% yield: IH-
NMR (CDCIS) 8 1.29 (CHS, t, 6H, J = 7.1 Hz), 1.91 (CHS, m, 2H), 2.60 (m, 2H),
67I
3.01 (CH;S, s, 3H), 4.26 (CHDO, m, 1H), 4.30 (CH;O, q, 2H, J = 7.1 Hz), 4.31l
(CH;O, q, 2H, J =7.l Hz), 7.82 (m, 4H); BC-NMR (CDCI;) 8 13.7, 24.2, 29.3,
37.2, 62.7, 67.1, 67.9 (CI-IDOH, t, J = 58.6 Hz), 123.5 (Ar), 131.2 (Ar), 134.4 (Ar),
165.9, 167.2.
Ethyl (SR)-2-Carbethoxy-2·phthalimido-5-cyanopentanoate-5-d; (28-
R) (S)-Mesylate 27-S (4.11 g, 9.3 mmol) and KCN (1.21 g, 18.6 mmol) in Me2SO
(100 mL) were heated at 50 °C under N2 for 19.5 h. After the mixture had cooled, 100 mL
water was added, and the resulting aqueous mixture was extracted with ethyl acetate (4 x
200 mL). The combined organic layers were then washed with brine (2 x 200 mL), dried
over Na2SO4, and evaporated at reduced pressure. The crude product was purified by
flash chromatography (1:1 ethyl acetate—hexane) to afford a crystalline solid (2.40 g, 69%),
which was recrystallized from ether—hexane to obtain 2.00 g colorless prisms, mp 89 °C:
IH-NMR (CDCI3) 8 1.30 (CH3, L 6H, J = 7.1 Hz), 1.82 (m, 2H), 2.38 (m, 1H), 2.61
(m, 2H), 4.31 (CH;O, q, 2I-L J = 7.1 Hz), 4.32 (CH;O, q, 2H, J = 7.1 Hz), 7.82 (m,
4H); BC-NMR (CDCI3) 8 13.7, 16.8 (C-5, t, J =20.2 Hz), 20.7, 32.2, 62.8, 67.0,
119.0 (CN), 123.5 (Ar), 131.2 (Ar), 134.4 (Ar), 165.9, 167.2; mass spectrum (on an
unlabeled sample) CI, isobutane m/z (relative abundance) (M+1)+ 374 (100), 373 (15);
IR 3020, 2980, 2200, 1800, 1779, 1740, 1385 cm‘1. Elemental analysis was performed _
on an unlabeled sample prepared by the same procedure. Anal. calcd. for C1gH;0N2O,:
C, 61.29%; H, 5.41%; N, 7.52%; found: C, 61.49%; H, 5.36%; N, 7.56%.
(2RS, SR)-Lysine-5-dl Dihydrochloride (29-R ). (R)-Nitrile 28-R
(0.434 g, 1.16 mmol) in 5 mL acetic acid and 2 mL concentrated HC1 was hydrogenated
over PtO2 (0.162 g, 0.71 mmol). After 5.5 h an additional portion of PtO2 (0.162 g, 0.71
mmol) was· added, and the mixture was stirred under H; for an additional 18 h. The
catalyst was removed by filtration, and the filtrate was evaporated to give a yellow oil.
68
Thin-layer chromatography showed no starting material to be present. The crude product
was heated in 5 mL 6 N HC1 at 115 °C for 15 h. An equal portion of water was added, and
the mixture was evaporated at reduced pressure. An additional portion of water was added,
and the mizrture was evaporated to remove any excess acid. The solid residue was
dissolved in 20 ml water, and the mixture was extracted with ether (3 x 20 mL) to remove
the phthalic acid, then treated with activated carbon. After evaporation of the water, the
residue was recrystallized from ethanol—ether to yield 0.063 g (25%) of the labeled lysine
dihydrochloride, which on 'I'LC co-chromatographed with authentic lysine (R; = 0.20, 1-
propanol-concentrated NH40I-I, 7:3). IH NMR (D20, DSS) 6 1.45 (m, 2H), 1.69 (m,
1I-I), 1.89 (m, 2H), 3.00 (d, 2H, J = 7.5 Hz), 3.74 (t, 1H, J = 6.0 Hz) I3C NMR (D20,
p-dioxane = 66.5) 6 21.2, 25.9 (C·5, t, J = 19.5 Hz), 29.5, 38.9 (C-6), 54.4 (C-2),
173.1 (COOH). '
3.13. Synthesis of (2RS, SS)-Lysine-5-d1 Hydrochloride
Ethyl (SR)-2-Carbethoxy-2-phthalimido-S-hydroxypentanoate-5-d1
(26-R). To 52 ml (S)-(-)-Alpine Borane (Aldiich, 81% optical purity) (0.5M solution in
THF, 26 mmol) was added labeled aldehyde 25 (4.72 g, 13.1 mmol) in 25 n1L dry TI—IF.
After the mixture had been stirred at room temperature under N2 for 5.5 h, it was placed on
a silica gel column and eluted with ether. Fractions containing the alcohol were combined
and evaporated at reduced pressure. The crude product was further purifled by flash
chromatography (ethyl acetate—petroleum ether, 4:6, then 6:4) to yield 4.30 g (90%) of the
labeled alcohol. IH and I3C NMR spectra were identical to those of the (SS)-alcohol.
Ethyl (5R)-2-Carbethoxy-2-phthalimido-5-methanesulfonyloxypem
tanoate-5-d1 (27—R). To an ice-cold, stirred mixture of (R)-alcohol 26-R (3.89 g,
10.7 mmol) and triethylamine (2.17 g, 21.4 mmol) in 75 ml CH2Cl2 was added
69
methanesulfonyl chloride (2.45 g, 21.4 mmol) in 20 rnL CH2Cl2 dropwise over 1 h. The
ice·water bath was removed, and stirring was continued for 2 h. The reaction mixture was
evaporated at reduced pressure, and 100 mL ethyl acetate-hexane (6:4) was added to the
residue. The liquid was decanted, and the residue was rinsed with portions of the ethyl
acetate-hexane mixture (4 x 20 mL). The crude product obtained after evaporation of the
combined washings was purified by flash chromatography (ethyl acetate-hexane, 6:4) to
afford 4.34 g (92%) of (R)-mesylate (27-R). IH and IIC NMR spectra were identical to
those of the (SS)-mesylate.
Ethyl (SS)-2-Carbethoxy-2-phthalimido-5-cyanopentanoate-5-dl (28-
S). (R)·Mesy1ate 27-R (4.26 g, 9.6 mmol) and KCN (1.26 g, 19.3 rrunol) in Me;SO
(100 mL) were heated at 45-50 °C under N2 for 23 h. Water (100 mL) was added to the
cooled reaction mixture, and the resulting mixture was extracted with ethyl acetate (4 x 200
mL). The extract was washed with brine (2 x 200 mL), dried over Na;SO.«,, and
evaporated under reduced pressure. The crude product was purified by flash
chromatography (1:1 ethyl acetate—hexane) to afford a crystalline solid (2.69 g, 72%),
which was recrystallized from ether—hexane to yield 2.41 g of (S)-nitrile 28-S as colorless
p1·isms, mp 89 °C. IH and I3C NMR spectra were identical to those of the (SR)-nitrile.
(2RS, SS)-Lysine-5·d1 Hydrochloride (29-S). (S)-Nitrile 28-S (0.75 g,
2.0 nunol) was hydrogenated at an initial pressure of 61 psi over PtO2 (0.225 g) in acetic
acid (10 mL) and concentrated HC1 (94 mL) over 23.5 h. The reaction mixture was filtered
to remove the catalyst, and the filtrate was evaporated under reduced pressure to leave a
yellow oil. TLC (6:4, ethyl acetate—hexane) showed no remaining starting material. The
crude product in 6 N HC1 (20 rr1L) was heated at 115 °C for 20 h. After that time the
reaction mixture was cooled, water (30 mL) was added, and the resulting mixture was
evaporated under reduced pressure. Additional portions of water were added, and the
I70
mixture was evaporated to remove excess acid. The residue was dissolved in water (20
mL), and the aqueous mixture was extracted with ethyl acetate (2 x 20 mL) to remove
phthalic acid. The aqueous layer was treated with activated charcoal; however, it remained
quite yellow., The mixture was placed on a Dowex 50W column (3 x 15 cm). The column
was washed with water (400 mL), then eluted with 2 N NH4OH. The fractions containing
lysine (TLC: 7:3, 1-propanol—concen¤·ated NI-I4OI-I) were combined, brought to pH 4
with HC1, treated with activated charcoal, and evaporated at reduced pressure. The residue
was recrystallized from H20-ethanol-ether to yield 98.6 mg lysine monohydrochloride,
which on TLC co·chromatographed with authentic lysine (R; = 0.19, 1-propanol-
concentrated NI—I4OH). The mother liquor yielded an additional 38.2 mg of slightly less
pure material, for a total yield of 37%. 1H and BC NMR spectra were identical to those of
(2RS,5R)-lysine.
3.14. Incorporation of (2RS, SR)-Lysine-5-d;
The fermentation and preliminary purification of the antibiotic took place at the
laboratories of SmithK1ine—RIT, Rixensart, Belgium. After S. virginiae 5277 was grown
for 3 days in a vegetative medium, ten 250-mL Erlenmeyer flasks each containing 40 mL of
the production medium were inoculated. After growth for 24 h, 250 |.LL of a solution
consisting of 50 mg (2RS, 5R)-lysine-5-d1 and 100 mg L-threonine in 2.5 mL deionized
water, sterilized by passing the solution through a Millipore filter, was added to each flask.
The fermentation continued for an unspecified amount of time (2-3 days), then the broth
was brought to pH 4.7 with HZSO4. The acidified broth was extracted three times with
methyl isobutyl ketone; the combined extracts were centrifuged for 10 min (5000 rpm),
then concentrated in vacuo to 5 mL. n-Hexane (200 mL) was added to the concentrated
extract, and the mixture was allowed to stand ovemight at 4 °C. Only part of the crude
antibiotic precipitated; it was filtered and dried. The supematant was therefore concentrated
71S
to dryness. The residue was dissolved in acetone—hexane; the mixture was extracted with
water. The water phase was lyophilized to yield an additional portion of crude
virginiamycin. A total of 650 mg material containing approximately 42 mg (HPLC)
undefmed virginiamycins was obtained.
The crude material was purified as previously described to yield 7 mg virginiamycin
S1. A portion of the labeled antibiotic was hydrolyzed, and the resulting mixture of amino
acids was derivatized as the N·trifluoroacetyl butyl esters as previously described. The
derivatives were analyzed by GC/MS as previously described. _
3.15. Incorporation of (2RS, SS)-Lysine-5-d1
The fermentation and preliminary purification of the antibiotic took place at the
laboratories of SmithK1ine—RIT, Rixensart, Belgium. S. virginiae 5277 was grown for 2
days in a vegetative medium, then 15 250-mL Erlenmeyer flasks each containing 40 mL of
a production medium were inoculated with the vegetative broth. A solution of 65 mg
(2RS, 5S)-lysine·5-d1 and 100 mg L-threonine in 3 mL distilled water was sterilized by
passing through a 0.45 um filter; 200 p.L of this solution was added to each of the
production flasks after 24 h fermentation. After 48 h, The broth was acidified, then
extracted three times with methyl isobutyl ketone. The combined extracts were
concentrated, diluted with water, then washed three times with rr-hexane. The aqueous
phase was lyophilized to yield 480 mg of a material containing 30 mg virginiamycin M2, 9
mg virginiamycin M1, and 6 mg virginiamycin S1. Virginiamycin S1 was isolated,
hydrolyzed, and prepared for GC/MS analysis as described previously.
3.16. Synthesis of DL-Phen‘ylalanine-3-UC-UN Hydrochloride
Benzoic-carboxy-UC acid. Bromobenzene (46 mL, excess) in dry, freshly
72
distilled TI-IF was added dropwise to a flask containing magnesium turnings (4.57 g, 0.19
mol) that had been activated by heating in the presence of I2 vapors. When the reaction
was complete, the flask was attached to a vacuum line and immersed in liquid N2. When
the mixture had solidified, the system was evacuated. Carbon dioxide—13C was produced
by the addition of concentrated HZSO4 dropwise into a flask containing barium carbonate·
BC (34.6 g, 0.176 mol). The carbon dioxide was condensed into the flask containing the
frozen phenylmagnesium bromide solution, then the flask was slowly brought to room
temperature. The mixture was extracted with ether. The ether layer was extracted with 1.5
N NaOH; the extract was acidified to pH 1 and extracted with ether. The ether layer was
dried over MgS04, filtered, and evaporated to yield 18.7 g benzoic acid (88% yield), mp
120-121 °C. BC NMR (CDCI3) 5 128.5 (C-3, d, 3JcC = 4.4 Hz), (129.4, C-1, not
seen), 130.2 (C-2, d, 2Jgg = 2.6 Hz), 133.8 (C-4, 172.3 (COOH, intense singlet).
Benzyl-a-BC alcohol. Benzoic-carboxy-UC acid (3.20 g, 26.0 mmol) was
dissolved in 150 mL freshly distilled dry ether. Lithium aluminum hydride (2.19 g, 57.6
mmol) was added in small portions over 30 min, during which time the mixture was stin·ed
at room temperature under N2. The reaction mixture was then heated at reflux for 3.5 h.
The flask was cooled in an ice bath, and the reaction was quenched by the dropwise
addition of 2 mL H20, followed by 4 mL 10% aqueous Na0H and 6 mL H20. The
mixture was filtered and the residue rinsed with several small portions of ether. The
combined filtrate and washings were dried over MgS04. Evaporation of the solvent left
3.24 g of the crude product. Vacuum distillation (2.5 torr) afforded 2.58 g of benzyl
alcohol (87.6% yield).
Benzaldehyde-a-BC (31). Benzyl·a-UC alcohol (2.58 g, 24.0 mmol) in 50
mL of dimethyl sulfoxide was heated to 180 °C for 6 h with a stream of air bubbling
through a fine frit into the reaction mixture. To the cooled mixture was added 100 mL
73
H20, and the aqueous mixture was extracted with ether (3 x 100 mL). The combined ether
layers were washed with H20 (2 x 100 mL), dried over MgS04, and evaporated at reduced
pressure to yield 2.13 g benzaldehyde (83%), which was shown to be relatively pure by
TLC (hexane—ether, 3:2), but gave low yields in subsequent reactions. Distillation did not
improve the quality of the product. The crude product was therefore puxitied as its bisulfite
addition product. Crude benzaldehyde from 7.74 g benzyl alcohol was dissolved in 2 mL
ether, treated with saturated aqueous NaHS02 (2 mL), and vigorously stirred for 30 min.
The resulting heavy precipitate was filtered from the mixture, rinsed with a little ether, and
dried. The bisulfite addition product was dissolved in 50 mL saturated aqueous NaHC03,
and the solution was extracted with ether (4 x 50 mL). The combined ether layers were
dried over MgS04 and evaporated to give 1.20 g (16%) pure benzaldehyde (31).
N-Acetylglycine-15N (32). To a stirred solution of glycine-15N (MSD
Isotopes) (1.04 g, 13.7 mmol) in 4 mL H20 was added acetic anhydride (2.9 g, 28 mmol).
The mixture became warm, and a white precipitate formed within 5 min. Stirring was
continued for 30 min, then the mixture was stored at 4 °C ovemight to maximize
crystallization. The product was filtered from the mixture, washed with a small volume of
ice-cold water, and dried to give 1.12 g (69%) product, mp 206·208 °C (literature 207-208
<>C)_141
ot-Acetamidocinnamic-3-13C-15N Acid Azlactone (33). Benzaldehyde-
a-13C (1.20 g, 11.2 mmol), N-acetylglycine-1-’N (0.906 g, 7.68 mmol), anhydrous
sodium acetate (0.470 g, 5.7 mmol), and acetic anhydride (1.8 mL) were heated at reflux
for 2 h under N2. After cooling, the flask was stored at 4 °C ovemight, then water (2 mL)
was added. The solid mass was broken up, filtered, and rinsed with two 1-mL portions of
cold water to yield 1.06 g (73%) product, which was used in the next step without further
puritication.
74
a-Acetamidocinnamic Acid-3-UC-ÜN (34). The crude azlactone 3 3
(1.06 g) in 10 mL acetone and 4 mL H20 was heated at reflux for 6 h. The acetone was
evaporated at reduced pressure, 25 mL H20 was added, and the mixture was brought to a
boil. The hot solution was filtered, and the residue was rinsed with a small portion of
boiling water. The filtrate was boiled with Norit A , and the hot mixture was tiltered. The
filtrate was reduced to 20 mL in vacuo and stored at 4 °C ovemight to allow for the
development of crystals. The white crystalline product was filtered from the mixture and
dried to give 0.977 g (84%) product, mp 193-194 °C (literature 193-194 °C).131
DL-Phenylalanine-3-1-’C·15N Hydrochloride (35). a-Acetamidocinnamic
acid 34 (0.878 g, 4124 mmol) was hydmogenated in glacial acetic acid (10 mL) over 5%
palladium on carbon (0.25 g) at room temperature for 2.5 h. The catalyst was removed by
filtration, and the filtrate was evaporated to dryness. The residue was heated at retlux for
11 h in 20 mL 1 N HC1. The mixture was evaporated to dryness at reduced pressure, and
the residue was dried further in vacuo over Na0H pellets to give 0.645 g (75%) of the
doubly labeled phenylalanine hydrochloride, identical by TLC and co-TLC with an
authentic sample of phenylalanine. IH-NMR: (D20, DSS) ö 2.93-3.16, 3.42-3.65
(CH2, dm, 2H, 1IC;.; = 132 Hz, 3];.;;.; = 3 Hz), 4.36 (CH, m, 1H), 7.32-7.48 (Ar, m,
SH); BC-NMR (D20) 5 35.6 (C-3, intense singlet), 54.1 (C-2, dd, IJCQ = 33 Hz, IJCN
= 6.5 Hz), 127.9, 129.2 (d, JCC = 3.7 Hz), 129.4 (d, JCC = 3.1 Hz), 134.0 (Ar-1, d, UCC
= 43 Hz), 171.5 (COOI—I); mass spectrum m/z (relative abundance) 167 (M+—HC1)
(1.9), 120 (5), 121 (8), 122 (60), 104 (13), 92 (70), 93 (19), 75 (100). The ions 120,
121, and 122 represent the ions (M+-COOH) that are unlabeled, singly labeled, and doubly
labeled, respectively. From the relative abundances of these ions it can be deduced that 35
is 7% unlabeled, 10% singly labeled (with either ßc or ISN), and 83% doubly labeled.
75
s 3.17. Incorporation of DL-Phenylalanine·3J3C-15N Hydrochloride
I The fermentation was carried out as previously described. After 8 h growth, a
filter-sterilized (0.45 um) solution of 290 mg DL-phenyla.1anine·3-l3CJ5N hydrochloride
and 230 mg L-threonine in 10 mL distilled water was added to the production broth (1.5
L). Virginiamycin S1 (7 mg) was isolated as described previously, then analyzed by BC
NMR. It was hydrolyzed and prepared for GC/MS analysis as described previously.
4. CONCLUSION
Some details of the biosynthesis of three unusual amino acid residues in the
antibiotic virginiamycin S1 have been studied. 4-Oxo-L-pipecolic acid was shown to arise
from L-lysine and not from L-aspartic acid or L-methionine as was previously suggested.
Incorporation of demonstrated that its nitrogen originates from the 6-
amino group of lysine.
Like 4-oxo-L~pipecolic acid, 3-hydroxypicolinic acid is biosynthesized from L-
lysine; the 6-nitrogen ofL-lysine is also retained in this amino acid. (2RS, 5R)—lysir1e-5·d1
and (MS, SS)-lysine-5-d1 were synthesized and incorporated into virginiamycin S1. The
deuterium from the 5—(R) isomer was detected in the 3-hydroxypicolinyl residue of the
antibiotic. No incorporation was seen of the 5-(S) isomer. This tinding indicates that the
5—pro(S) hydrogen of lysine is lost at some point during the biogenesis of 3-
hydroxypicolinic acid.
L-Phenylglycine had previously been shown to originate from L-phenylalanine. In
order to study the mechanism of the migration of the amino group to the benzylic position
during the transformation, DL-phenyla1anine~3-1-’C-15N was synthesized and incorporated
into virginiamycin S1. The labeled nitrogen was not incorporated in L-phenylglycine;u
therefore, it can be concluded that the process is an intermolecular one.
In addition, a synthesis of 4-oxo-DL·pipecolic acid was devised. A new derivative
of this amino acid, N-benzoyl-4—oxopipecolic acid ethylene ketal, was developed, which
allowed its separation from the other virginiamycin S1 amino acids by a previously
worked·out procedure.
76
5. REFERENCES
1. Cocito, C. In Antibiotics, Hahn, F. E., Ed.; Springer—Verlag: Berlin, 1983; Vol.6,pp 296-332.
2. Okumura, Y. In Biochemisrry and Generic Regulation of Commercially ImportantAnribiorics; Vining, L. C., Ed.; Addison—Wesley: Reading, MA, 1983; Chapter 6, p147. ,
3. De Somer, P.; Van Dijck, P. Anribior. Chemorher. 1955, 5, 632.
4. Ogata, K.; Matsuura, M.; Irie, H.; Ueno, T.; Tani, Y.; Yamada, H. J. Anribior.1978, 31, 1313.
5. Donovick, R.; Duteher, J. D.; Heuser, L. J.; Pagano, J. F. U. S. Patent 2 990 325,1961; Chem. Abstr. 1962, 56, 3923b.
6. Charney, J.; Fisher, W. P.; Curran, C.; Machlowitz, R. A.; Tytell, A. A. Anribior.Chemother. 1953, 3, 1283.
7. Ball, S.; Boothroyd, B.; Lees, K. A.; Raper, A. H.; Smith, E. L. Biochem. J.1958, 68, 24P.
8. Arai, M.; Karasawa, S.; Yonehara, H.; Umezawa, H. J. Anribior. 1958, 11, 14.
9. Benazet, F.; Cosar, C.; Dubost, N.; Julou, L.; Mancy, D. Semaine Thérapeurique1962, 38, 13.
10. Celmer, W. D.; Sobin, B. A. Anribior. Annual 1955/56 1956, 437.
11. Linder, F.; Walhausser, K. H.; Weidenmüller, H. L. German Patent 1 072 773,1960; Chem. Absrr. 1961, 55, 13764c.
12. Bartz, Q. R.; Standiford, J.; Mold, J. D.; Johannessen, D. W.; Ryder, A.; Maretzki,A.; Haskell, T. H. Anribior. Annual 1954/55 1955, 777.
13. Okumura, Y.; Okumura, K.; Takei, T.; Kouno, K.; Lein, J; Ishikura, T.; Fukagawa,Y. J. Anribior. 1979, 32, 575.
14. Heinemann, B.; Gourevitch, A.; Lein, J.; Johnson, D. L.; Kaplan, M. A.; Vans, D.;Hooper, I. R. Anribior. Annual 1954/55 1955, 728.
15. Hamill, R. L.; Stark, W. M. U. S. Patent 3 932 980, 1975; Chem. Absrr. 1976,84, 87963x.
16. Celmer, W. D.; Cullen, W. P.; Routien, J. B.; Moppet, C. E.; Shibakawa, R.; Tone,J. German Patent 2 516 020, 1975; Chem. Absrr. 1976, 84, 87960u.
17. Martinelli., E.; Zerilli, F.; Volpe, G.; Pagani, H.; Cavalleri, B. J. Anribior. 1979,32, 108.
77
78
18. Brazhnikova, M. G.; Kudinova, M. K.; Potapova, N. P.; Filippova, T. M.;Borowski, E.; Zielinski, J.; Golic, J. Bioorg. Khim. 1975, 1, 1383.
19. Liu, W.-C.; Seiner, V.; Dean. L. D.; Trejo, W. H.; Principe, P. A.; Meyers, E.;Sykes, R. B. J. Antibior. 1981, 34, 1515.
20. Kruglyak, E. B.; Kostantinova, N. V.; Sulavko, L. A. Antibiotiki 1961, 6, 298.
21. Bodanszky, M.; Ondetti, M. A.; Sheehan, J. T. U. S. Patent 3 373 151; Chem.Abstr. 1968, 69, 59563p.
22. Durant, F.; Evrard, G.; Declercq, J.P.; German, G. Cryst. Struct. Commun.1974, 3, 503.
23. Vasquez, D. Biochem. Biophys. Acta 1962, 61, 849.
24. Delpierre, G. R.; Eastwood, F. W.; Gream, G. E.; Kingston, D. G. I.; Sarin, P. S.;Todd, A. R.; Williams, D. H. J. Chem. Soc. C 1966, 1653.
25. Okabe, K. J. Antibiot. Ser. A 1959, 12, 86.
26. Preud'homme, J.; Tarridec, P; Belloc, A. Bull. Soc. Chim. Fr. 1968, 585.
27. Kingston, D. G. I.; Sarin, P. S.; Todd, A. R.; Williams, D. H. J. Chem. Soc. C1966, 1856.
28. Fallona, M. C.; Mayo, P. D.; McMorris, T. C.; Money, T.; Stossel, A. Can. J.Chem. 1964, 42, 371.
29. Bimbaum, G. I.; Hall, S. R. J. Am. Chem. Soc. 1976, 98, 1926.
30. Chamberlin, J. W.; Chen, S. J. Antibiot. 1977, 30, 197.
31. Moppet, C. E.; Whipple, E. B. Abstracts of the Seventecnth InterscientificConference on Antimicrobial Agents and Chemotherapy, 1977, p 243.
32. Vanderhaeghe, H.; Parmentier, G. Bull. Soc. Chim. Belg. 1959, 68, 716.
33. Vanderhaeghe, H.; Parmentier, G. J. Am. Chem. Soc. 1960, 82, 4414.
34. Vanderhaeghe, H.; Janssen, G.; Campemolle, F. Tetrahedron Lett. 1971, 2687.
35. Oberbäumer, I.; Grell, E.; Raschdorf, F.; Richter, W. J. Helv. Chim. Acta 1982,65, 2280.
36. Verwey, W. F.; West, M. K.; Miller, A. K. Antibiot. Chemother. 1958, 8, 500.
37. Watanabe, K. J. Antibiot. Series A 1961, 14, 1.
38. Hobbs, D. C.; Celmer, W. D. Nature 1960, 187, 598.
39. Bodanszki, M.; Ondetti, M. A. Antimicrob. Agents Chemother. 1963 1964, 360.
79
40. Eastwood, F. W.; Snell, B. K.; Todd, A. R. J. Chem Soc. 1960, 2286.
41. Cox, B. R.; Eastwood, F. W.; Snell, B. K.; Todd, A. R. J. Chem. Soc., Chem.Commun. 1970, 1623.
42. Bodanszky, M.; Sheehan, J. T. Antimicrob. Agents Chemother. 1963 1964, 38.
43. Okumura, Y.; Takei, T.; Sakamoto, M.; Ishikura, T.; Fukagawa, Y. J. Antibiot.1979, 32, 1002.
44. Okumura, Y.; Takei, T.; Sakamoto, M.; Ishikura, T.; Fukagawa, Y. J. Antibiot.1979, 32, 584.
45. Chopra, C.; Hook, D. J.; Vining, L. C.; Das, B. C.; Shimizu, S.; Taylor, A.;Wright, J. L. C. J. Antibiot. 1979, 32, 392.
46. Haskell, T. H.; Maretzki, A.; Bartz, Q. R. Antibiot. Annual 1954/55 1955, 784.
47. Sheehan, J. C.; Zachan, H. G.; Lawson, W. B. J. Am. Chem. Soc. 1958, 80,3349.
48. Okumura, Y.; Okamura, K.; Takei, T.; Kouno, K.; Lein, J.; Ishikura, T.;J Fukagawa, Y. J. Antibiot. 1979, 32, 575.
49. Okumura, Y.; Fukagawa, Y.; Okamoto, R.; Ishikura, T. Agri. Biol. Chem. 1982,46, 731.
50. Okumura, Y.; Okamoto, R.; Ishikura, T. Agric. Biol. Chem. 1984, 48, 543.
51. De Somer, P.; Van Dijck, P. Antibiot. Chemother. 1955, 5, 632.
52. Vanderhaeghe, H.; Van Dijck, P.; Parmentier, G.; De Somer, P. Antibiot.Chemother. 1957, 7, 606.
53. Declercq, J. P.; Germain, G.; Van Meerssche, M.; Hull, S. E.; Irwin, M. J. ActaCrystallogr. 1978, B34, 3644.
54. Kiryushkin, A. A.; Burikov, V. M.; Rosinov, B. V. Tetrahedron Lett. 1967,2675.
55. Compemolle, F.; Vanderhaeghe, H.; Janssen, G. Org. Mass Spectrom. 1972, 6,151.
56. Anteunis, M. J. O.; Callens, R. E. A.; Tavemier, D. K. Eur. J. Biochem. 1975,58, 259.
57. Callens, R. E. A.; Anteunis, M. J. O. Biochem. Biophys. Acta 1979, 577, 324.
58. Jannsen, G.; Anne, J.; Vanderhaeghe, H. J. Antibiot. 1977, 30, 141.
59. Kessler, H.; Kühn, M.; Löschner, T. Leibigs Ann. Chem. 1986, 21.
80
60. Kessler, H.; Kühn, M.; Löschner, T. Liebigs Ann. Chem. 1986, 1.
61. Cocito, C. Microbiol. Rev. 1979, 43, 145.
62. Dijck, P.; Vanderhaeghe, H.; De Somer, P. Antibiot. Chemother. 1957, 7,
63. Biot, A. M. In Biotechnology of Industrial Antibiotics; Vandamme, E.J., Ed.;Marcel Dekker: New York, 1984; Chapter 25.
64. Danielson, D. M. J. Anim. Sci. 1972, 35, 187.
65. Olson, L. D.; Rodabaugh, D. E. Am. J. Vet. Res. 1977, 38, 1485.
66. Chassagne, P.; Curveiller, J.; Bocquet, L.; Guibert, C. Thérapie 1963, 18, 969.
67. Fossiez, M. Ann. Pediatr. 1963, 39, 633.
68. Prinquet, R. Presse Med. 1962, 70, 2573.
69. Le Goffic, F. J. Antimicrob. Chemother. 1985, 16 (Suppl. A), 13.
70. Corbet, J. P.; Cot1·e1, C.; Farge, D.; Paris, J. M. French Patent 2 549 063, 1985;i
Chem. Abstr. 1985, 103, 160862f.
71. Corbet, J. P.; Cotrel, C.; Farge, D.; Paris, J. M. French Patent 2 549 064, 1985;Chem. Abstr. 1985, 103, 196421a.
72. Corbet, J. P.; Cotrel, C.; Farge, D.; Paris, J. M. French Patent 2 549 062, 1985;Chem. Abstr. 1985, 103, 160860d.
73. Le Goftic, F.; Capmau, M. L.; Abbé, J.; Cerceau, C.; Dublanchet, A.; Duval, J.Annales de Microbiologie (Institut Pasteur) 1977, 128B, 471.
· 74. Ennis, I—L L. Arch. Biochem. Biophys. 1974, 160, 394.
75. Contreras, A.; Vasquez, D. Eur. J. Biochem. 1977, 74, 549.
76. Parfait, R.; DeBethune, M. P.; Cocito, C. Mol. Gen. Gene:. 1978, 166, 45.
77. Parfait, R.; Cocito, C. Proc. Natl. Acad. Sci. USA 1980, 77, 5492.
78. Chinali, G.; Moureau, P.; Cocito, C. J. Biol. Chem. 1984, 259, 9563.
79. Cocito, C.; Chinali, G. J. Antimicrob. Chemother. 1985, 16 (Suppl. A), 35.
80. Vasquez, D. J. Antimicrob. Chemother. 1985, 16 (Suppl. A), 225.
81. Di Giambattista, M.; Vannuffel, P.; Sunazuka, T.; Jacob, T.; Omura, S.; Cocito, C.J. Antimicrob. Chemother. 1986, 18, 307.
81
82. Di Giambattista, M.; Thielen, A. P. G. M.; Maassen, J. A.; Moeller, W.; Cocito, C.Biochemistry 1986, 25, 3540.
83. Di Giambattista, M.; Ide, G.; Engelborghs, Y.; Cocito, C. J. Biol. Chem. 1984,259, 6334. ,
84. Di Giambattista, M.; Cocito, C. Biochim. Biophys. Acta 1983, 757, 92.
85. Ennis, H. L. Biochemistry 1971, 10, 1265.
86. Grell, E.; Krause, G.; Lewitzki, E.; Mager, G.; Ruf, H. In Prog. Bioorg. Chem.Mol. Biol., Proc. Intl. Symp. Front. Bioorg. Chem. Mol. Biol.; Ovchinnikov, Y.A.,Ed; Elsevier: Amsterdam, 1984; pp 239.
87. Grell, E.; Oberbäumer, I.; Ruf, H.; Zingsheim, H. P. In Biochemistrjy ofMembraneTgansport; Semenza, G.; Carafoli, E., Eds.; Springer—Verlag: Heidelberg, 1977; p1 7.
88. Kurahashi, K. In Antibiotics; Corcoran, J. W., Ed.; Springer—Verlag: Berlin, 1981;Vol. 4, pp 325-352.
89. Kamal, F.; Katz, E. J. Antibiot. 1976, 29, 944.
90. Molinero, A. A. Master's Thesis, Virginia Polytechnic Institute and State University,1982.
91. Hook, D. J.; Vining, L. C. J. Chem. Soc., Chem. Commun. 1973, 185.
92. Hook, D. J.; Vining, L. C. Can. J. Biochem. 1973, 51, 1630.
93. Ogawara, H.; Maeda, K.; Umezawa, H. Biochemistry 1968, 7, 3296.
94. Nagato, N.; Okumura, Y.; Okamoto, R.; Ishikura, T. Agric. Biol. Chem. 1984,48, 3135.
95. Rodwell, V. W. In Metabolic Pathways, 3rd ed.; Greenberg, D. M., Ed.; AcademicPress: New York, 1969; Vol. 3, Chapter 15 (Part H), and references therein.
96. Clark-Lewis, J. W.; Mortimer, P. I. J. Chem. Soc. 1961, 189; and referencestherein.
97. Gupta, R. M.; Spencer, I. A. Phytochemistry 1970, 9, 2329.
98. Grobbelaar, N.; Steward, F. C. J. Am. Chem. Soc. 1953, 75, 4341.
99. Rothstein, M.; Miller, L. L. J. Biol. Chem. 1954, 211, 851.
100. Meister, A.; Padhakrishnan, A. N.; Buckley, S. D. J. Biol. Chem. 1957, 229, 789.
101. Fowden, L. J. Exp. Botany 1960, 11, 302.
102. Meyer, J. J. M.; Grobbellaar, N. Phytochemistry 1986, 25, 1469.
82
103. Boulanger, P.; Osteux, R.; Bertrand, J. Biochim. Biophys. Acta 1958, 29, 534.
104. Shoji, J.; Hinoo, H. J. Antibiot. 1975, 28, 60.
105. liggéis, M. B. Ph.D. Dissertation, Virginia Polytechnic Institute and State University,
106. C1ark·Lewis, J. W.; Mortimer, P. 1. J.Chem. Soc. 1969, 189.
107. Jolles, G.; Poiget, G.; Robert, J.; Terlain, B.; Thomas, J.-P. Bull. Soc. Chim. Fr.1965, 2252.
108. Halin, F.; Slosse, P.; Hootelé, C. Tetrahedron 1985, 41, 2891.
109. lgggson, T. A.; Desaty, D.; Brewer, D.; Vining, L. C. Can. J. Biochem. 1967, 45
110. Helbling, A. M.; Viscontini, M. Helv. Chim. Acta 1976, 2284.
111. Birch, A. J.; Simpson, T. J. J.Chem. Soc. Perkin I 1979, 816.
112. Luclmer, M. Secondary Metabolism in Microorganisms, Plants, and Animals, 2nded.; Springer—Verlag: Berlin, 1984.
113. Personal communication from Dr. Yasushi Okumura.
114. Rodwell, V. W. In Metabolic Pathways, 3rd ed.; Greenberg, D. M., Ed.; AcademicPress: New York, 1969; Vol.3, Chapter 16 (Part H).
115. Levy, G. C.; Lichter, R. L. Nitrogen-15 Nuclear Magnetic Resonance Spectroscopy;John Wiley: New York, 1979; Chapter 4.
116. Townsend, C. A.; Ho, M.; Mao, S. J. Chem. Soc., Chem. Commun. 1986, 638.
117. Ho, M. Ph.D. Dissertation, The Johns Hopkins University, 1983.
118. Kwart, H.; Nickle, J. H. J. Am. Chem. Soc. 1976, 98, 2881.
119. Kwart, H.; George, T. J. J. Org. Chem. 1979, 44, 162.
120. Midland, M. M.; Greer, S.; Tramontano, A.; Zderic, S. A. J. Am. Chem. Soc.1979, 101, 2352.
121. Sallach, H. J.; Fahien, L. A. In Metabolic Pathways, 3rd ed.; Greenberg, D. M.,Ed.; Academic Press: New York, 1969; Vol. 3, Chapter 14, and references therein.
122. Leete, E. Can. J. Chem. 1987, 65, 226.
123. Townsend, C. A.; Brown, A. M. J. Am. Chem. Soc. 1981, 103, 2873.
124. Townsend, C. A.; Brown, A. M. J. Am. Chem. Soc. 1983, 105, 913.
83u
125. Hammond, S. J.; Williamson, M. P.; Williams, D. H.; Boeck, L. D.; Marconi, G.G. J. Chem. Soc., Chem. Commun. 1982, 344.
126. McGahren, W. J.; Martin, J. H.; Morton, G. O.; Hargreaves, R. T.; Leese, R. A.;Lowell, F. M.; Ellestad, G. A.; O'Brien, E. O.; Holker, J. S. E. J. Am. Chem. Soc.1980, 102, 1671.
127. Parry, R. J.; Kurylo—Borowska, Z. J. Am. Chem. Soc. 1980, 102, 836.
128. Thiruvengadam, T. K.; Gould, S. J.; Aberhart, D. J.; Lin, H.-J. J. Am. Chem.Soc. 1983, 105, 5470.
129. Aberhart, D. J.; Gould, S. J.; Lin, H.-L.; Thiruvengadam, T. K.; Weiller, B. H. J.Am. Chem. Soc. 1983, 105, 5461.
130. Yoshitake, A.; Makari, Y.; Kawahara, K.; Endo, M. J. Labelled Comp. 1973, 9,537.
131. Herbst, R. M.; Shemin, D. In Organic Syntheses; Blatt, A. H., Ed.; John Wiley:New York, 1943; Coll. Vol. 2, p 1.
132. Gillespie, H. B.; Snyder, H. R. In Organic Syntheses; Blatt, A. H., Ed.; JohnWiley: New York, 1943; Coll. Vol. 2, p 489.
133. Personal communication from Dr. André Biot, SmithK1ine—RlT.
134. Beimann, K. Mass Spectrometry: Organic Chemical Applications; McGraw—I—Iill:New York, 1962.
135. Pirkle, W. H.; Hyun, M. H. J. Org. Chem. 1984, 49, 3043.
136. Elliot, D. F. J. Chem. Soc. 1950, 62.
137. Mozingo, R.; Wolf, D. E.; Harris, S. A.; Folkers, K. J. Am. Chem. Soc. 1943,65, 1013.
138. Abderhalden, E.; Heyns, K. Chem. Ber. 1934, 67, 530.
139. Although N-benzoyl-N-methyl-L-phenylalanine is a known compound (Jones, G.Eur. Pat. Appl. 19,411, 1980; Chem. Abstr. 1981, 95, 62696b), we were unable tolocate a reference to its melting point. Molinero reported in his notes a melting pointof 138-140 °C for the same compound. The racemic compound has a melting pointof 121 °C (Deulofeu, V. Chem. Ber. 1934, 67, 1542).
140. Balog, A.; Breasu, D.; Voinescu, V.; Herman, M.; Vargha, E.; Ramontian, E. Rev.Roum. Chim. 1973, 18, 123.
141. Herbst, R. M.; Shemin, D. In Organic Syntheses; Blatt, A.H., Ed.; John Wiley:New York, 1943; Coll. Vol. 2, p 11.
APPENDIX
84
LIST OF SPECTRA
1. IH-NMR spcctrum of butyl 3-hydroxypicolinatc
2. IH-NMR spcctrum of 2-cyano-4-pipcridonc cthylcnc kctal (12)
3. BC-NMI{ spcctrum of 2-cyano·4»pipcridonc cthylcnc kctal (12)
4. 1H·NMR spcctrum of 4-oxo-DL—pipcco1ic acid cthylcnc kctal (13)
5 . BC-NMR spcctrum of4-oxo-DL—pipcco1ic acid cthylcnc kctal (13)
6. 1H·NMR spcctrum of 4—oxo-DL-pipccolic acid cthylcnc kctal hydrochloridc (13)
7. BC-NMR spcctrum of 4-oxo·DL-pipccolic acid cthylcnc kctal hydrochloridc (13)
8 . 1H-NMR spcctrum of 4-oxo-DL-pipccolic acid hydrochloxidc (14)
9. IH-NMR spcctrum of 4,4»dica1·bcthoxy-4-phthaliuiidobutanal (23)
10. BC-NMR spcctrum of 4,4—dica1‘bcthoxy-4-phtlialinüdobutarial (23) ·
11. IH-NMR spcctrum of cthyl 2-carbcthoxy-2·pht11a1i111ido-5-hydroxypc11tauoatc—5·d1
(24)12. BC-NMR spcctrum ofcthyl 2-carbcthoxy-2-phthalimido-5·hydxoxypc11ta11oatc-5-dl
(24)
13. IH-NMR spcctrum of 4,4·dicarbcthoxy-4-phthaliuiidobutanal-5-611 (25)
14. BC-NMR spcctrum of 4,4-dicarbcthoxy-4—phtha1in1idobutana1-5-d1 (25)
15. 1H-NMR spcctrum of cthyl 2-carbcthoxy-2-pht1ia1i111ido·5-mctl1ancsu1fonyloxypcr1tan~
oatc-5-dl (27)
16. BC-NMR spcctrum ofcthyl 2-carbcthoxy·2·phtl1a1irrLido-5-mct11ancsu1fony1oxypcn-
umoatc-5-dl (27)
17. 1H-NMR spcctrum of cthyl 2-carbcthoxy·2-phtha1imido·5-cyanopcntanoatc-5-dl (28)
18. BC-NMR spcctrum of cthyl 2-carbcthoxy-2-phtha1imido·5-cyanopcntanoatc-5-61)
(28)
' 85
86
19. 1H·NMR spcctrum of (2RS, 5R)-lysinc—5d1 dihydrochloridc (29)
20. 13C·NMR spcctrum of (2RS, 5R)-lysinc-5d] dihydrochloddc (29)
21. BC-NMR spcctrum of virginiamycin S1 standard
22. BC-NMR spcctrum of virginiamycin S1 from lysinc-6-13C-6-UN incorpomtion
23. 1H·NMR spcctrum of bcnzoic-carboxyJ3C acid
24. BC-NMR spcctrum ofbcnzoic-carboxy-BC acid
25. BC-NMR spcctrum of bcnzyl-oz-BC alcohol
26. IH-NMR spcctrum of DL-phcnylalaxxiric-3-13C-15N hydrochloridc (35)
27. 13C—NMÜR spcctrum of DL-phcnylalariinc-3J3C-15N hydrochloridc (35)
28. BC-NMR spcctrum of virginiamycin S1 from phcnylalaninc-3J3CJ5N incorporation
87
2 2.2252 2 2 2 2 2 2 2 22. gf; { Q '; — é E E E 2E? ?=1
E E = 2 E E1 2 2 2- 2
12 7 Ā22 1
1
1.311;:4111;.14 141 1 4 4 42 . 2- 22-:32..-2 222-%22 .2;.2:.-2 ¤.
é: ä: E E2:''- ;:???E:'22:äIiä-fiiiä; l*E*E—2 E E E *5 Z¢E2··éI·‘!: E-~:E '2 : " E E E
.2 1 é ~2112;.22.’2‘
$2212*21:12 i 2 2 2-
2 2 2
2 2 · 2 : 2 2 1. 22 · - 1 2 i. Q Z 2 1 2 2 2 2 2 Q 2 2 2 2 2 2.
2.; 2‘
2 2 2 2 2 2 2 2 2 1 2 2 2 2E 2 1 ä ä E 7 ä 2 F — é = .2 §=§=’=2=€21i E 2-1- li 1 2 1 z -52 Ai E F 2 -2 2 i·iä— Q
-2.. -22. 2 2 2 2 2 2 I ';2°é_ 2 2 2 2 2.i.z *7E ‘ ¥ · . 2 = i >·_- §1=2 121-=2 2- 2-=€ 2 2 21 2 2 é 2 2 :2 ät?"°“ 1*1212.21 2j:ä§?·222— 1 2 Z i 2 4 E ä 2 E 1.
.§ 2 i' 21 1 § E ? é — ii Q 2 Q ä-"
__;..22 -$2 4 . .. E
‘2 __ 2 . ä 2 E
2‘
2 2 2 12 2 2 2 1 2 *52 2 2 2- 2 2 1 2 _¤
1 2 1-2 2 2 212r 22212 2=$-22512-22 .2 1221;:-.2--2. = 2 2 2 2 2 = - 2 2 12·—”. E ¥ ~ · Z Y E EÄEQLQi E 2 2 .2 E E 2% E.2 2 é
2 2--2- 2 z-1 *5222 2 2 2 2 1 2 2: 2 2 2 2 2 1 2 - - 2 :-212 v2 1 2 : 2 2 1 · 2 Q-2 E 2 EI 2 . = VE . 1 E ä EK
4 22 2 2 2 jl 2 · 2 2 2 . · 1 E ::2-2 2 2 2 jl 2--: 2 1 2 2 2 1 E Z äf
2 2 2 2 2 E 2 · 2 2 1 1 22äE 2 jl 2 E 2 2 2 2 1 2 _ 2 1 2 2 2-
;‘
2 2 2 2 . il · . 2 2 - 2 2 2 1 2 2-2 2l 2-2 2 2 1 2 jl 2 2 z 1 2 2 . 2 2
2: 2 g jl g Z 2 2i2 2 2 2 2 EE 2 2 z. 6 I1‘ 1 1 ai -4*1
2 I
._
..2 {2 é 2
1§_2 2 Q. I :1.--22 :.2 :2 2 .2 2 2
Ä-E -- 2-2-
—· 2 ; 2
’“E 'Ä ' ;
“
88
QV
·“ . QVQ, Q
;-Z _ÜQ—-——QV
i..ij_li%_*?¥22Ä;§ä=§—;EQQ2YQ 2 2 ° Q ä 2 Z Z Z
Q · - äE
Q =_ Ü l Q E 2 TZ E 2 N
Q = Q ·’ Q V
Q 2 SQ 2
*‘———--_ -
‘————— ‘ 2 6 2 ;2 "
; l 2 *+%------2 -------_ 2 Q g T
Z·---____;; . Q Q — 6 *"Q*·····~-—--2- -_ ; _ _ Q E E
iQ Q
ä
Ü r 2 Q ¤ 2 2 ·‘Q‘Q Q·—·—-———-__; = Ü * 2
. Z 2 2 - E · I = · —·——- 2 - 5
:·-2 -2-
po
E Ä 2Ü 22;-I
; ; g . V
Q
’*'·——-_._ I
Q Q
Q
""Q·· _ § z-_Z:Ä·.i.2-Z-
QQ ·—i-2_Z__j"“
;———-i_-—·_l V
IQ V Ü :'*§ - -1.-_ ;
Q
—¢2-_________ Z‘ "“——— 2-___ 2 Q E
ÄÜZ9
2Z Z g IE
—— QQ"·—··——— :-— 2 Q 9
2i N
+—·...-.-;-:—"“‘—··——-
°""“Q—lÜ _T « Q G-
ÜÜÜ Ü ——*·*
Ü E ÜQ
Q
Q = é w
·E—‘x·Q ;__Q__
_—* _ _
-
2~- * 2 Q Q Ü 2 Z
~--·‘ *”Q‘* ·—·- Q 2 Q Q
Q.._
_>— Ü_—-Ü---YA
Y -’*——T_____··
3Q Ü
QÜ
—l
..-_"°QQ·····——~--°Q‘QQ·1—·--- — 2--
Q · ""
-.;—··——V - --5- 1*V Q 1
Z2-;-l‘*·
°""—Q Ü
Ü Ü Ü Ü Ü__ÜÜÜÜ_—_ÜÜÜ Ü ÜÜ Ü
2
Z
_i = Q I E
QQ?··——·Ü-—·V-
· Q Q-
?
+—-______;“"fQ—1-22-- ____ ;_
Q 2; §_V ___ M; -
2 I V
;—._;___ Q _V 2
SQQ
2---; T_"‘·——-
-L--:Ü ~ ° Q
_ 89
Om
T .7"‘
-FQ „S'b£ O
2':·*
:°' ·' cs
3$4 S
Q)G2” >•E6
g Qj oJ E
U
s S¤ .E-‘“
ea.I‘
YA G· G
*:2.of¤¤‘ 9NhuG
‘ E‘ 0¤· ENUäm
8 ä+ J ¥vi - UQ ; M’
Il
9. -: F;
Q O
90
6**L
1[As
M•-n\/
E_, *52:::: ' !'x”
foS.2>.E0
‘°+. IE’ " ues
N .260
99V 6><‘ 9
i Q-
9 a-‘6
ESn.*50__ S-
Eärg '. U}
«_‘Y
f ¥SIll
4
N
äcon n n 9 9 V_________
91
Q Q. Q Q QQ QQ Q Q Q- 2V · I 6
°“ 6QE
Q QQij
2 I ’ Q2
5 Q oB -*2
·°‘· J
E Q Q
ä Q E>i.:E
QQ Q Q IE0’ G
Q .22 E
Q QQ ä.9Q.
„ I
QQ Q
__ I"Q Q~ XQ . Q- Q
Q Q
.. QQ E- — n-
EQ3.
; vz- Q
ä¥U
S 2
vi
M x-’ I
92
‘ ’ _? ' " A‘ $2·
9 "ÜM " Ü ~·V - V--- 9
Q __ ,7 7 7 .. . -_ H
.29‘ O» · LU
7 ____ >•' E
· ‘ EVu- 9
’ V S7, » 2
V >•_ E
9
9>· ‘ NV .2E
· 99
.. .9_ S-n_}
7 m · QÜ ’QN9 ··· ‘?Q
Ms-’ _ G
V ES
- - InLn99
Y Y? ~. %‘ S
¥’ SI4
~
93
5-
•—_
E
o0
' 1;.2 66. $2
„ '‘; Q
4)
1-ä “ 2.¥’«
.:3 Q
U VT E
S E E, '
E06)
_ Z0C2
. Ei80
.,.... n· QE cu
2"5QQu.2-Q-—iQ— éhtSc'és-Q
E
Q2
2 2Z’övsid
I-
94
‘
"‘?*""——*——·——·+-~—-—·-—;-„..%_____;;__;%;Ä § · ¥ ä ¤·
»;„2»···=·--———+-q-é26»;«:„;;;;;¥a.=
3 ; E li · ; _ Ü _ ‘„—;V ¢ 2 6 , Q j E Ä ll E 2 ‘
21 ‘ ; „ { ; ä 1 2 2 1 2 § A
‘ ä 1
i · ‘ · + 2 ; ; 2 2 2 ; „ 2E ¥ 2
Q . 2 ä ¢ ° V 2 2 A ; - ; ; Q g Li-
Q-? 2 A ? :‘ = ¢ = 2 ; 2 E Q 2
' . i
··<—-—~—»-—--...- _ ~ E‘ 2
’E E E
’°2‘‘$ E
a . ; - ; A ·-————__...._;;......'“°"1—*—+‘—“—·— j ? ä r-
2 3;-;- :"··— ·———————..;._.+.é.;é Ü q
‘; ä ?¥‘ L 2
E4 ;
_ i¢§§i é .=
; ··· ·—-———-—--M;. · _- » · ; „ Z_ ; · 2 5 E v
;— Q .‘ ——--——-_.p.
; . ä
2 ___ ______r—-.‘i~—i-r Q ; Q ¤
· ·--—··——·-——--—--.. . 1. = ·äß s -=
,-4 ~——--- °
·· --—— ‘ ~; E :„.’°·=@·‘@ °°‘ 2 ; ;
+··+———· ———‘ ; · Äg·<;1;g¤er2 _g
g_ ‘ ‘“"‘··". ' ' ’ E ir? '5
~ E, ·--—- -— ........__________;__ __·1•_·——-‘ Q · ··· -—» _ I - · ;- g . Q
J‘ ·· ?Ii§‘%E€¢%=·.E·
E
i5.;;-=·2 -— % _ 1 Y--- Q‘i‘5"z '
·——~ - ~ -~ = .„;_.___‘
. T E =
—-~————„-- W „.-, .___·—-—· · ; ; ¤ 1 <=
ä 25 ; ä's 2
X?‘*
·-__
„.•—
‘ °'°‘*“ I ‘ ; °1 a - i »+————>--M-M-..-.„.-..;____
__‘ ; ;„E i -
2 ;·?*°i ä ä
-> =
··—·- -- ~ 1 =— A--1---. .. ..
—“ A S Ü = 3
··· ·-— - ¤
"’
T‘i’——--·—————--—„....__________
"“" "·—‘—~'~—-—— _ 5 5 -
-E.„-;____
···————-———.——-—
_·-;-__
—+———-—..-QÄÖ'° ___ M"; ··—·——«—————--~— ‘... V„_ ; E j "
·- ····· ·—i'
”"‘*—————·-,..;....;.-—__ 2___ _ " ‘·• ” I
Ä‘—
Ä °
· ··· —-——-———- .»_‘_A________________ · V ——— . g = ¤¢
“—°_——'1‘*—·-————-—-'’ Q ‘ {ww- » -
I '—“"'“— , ? - . 5
"··—·—··-——-~ ._________ _____;____ . V i = g. = 2 Q —
1-:·‘_
"‘*
i ~ Ü j_g j *€··· ·YÜ.‘
·+——+ L L;
95
—-LV——·
2 ‘ ;» ; 1 °’*T‘°""“—·—————···—2 ‘ A——'·—’ ;1§i;Vi€ä ··_;
2 2 2 2 3“T°’T‘“‘T“‘i"*i·—·—v—§ E ä iii? l =
g;22%;· ·-g2 2 E ä 2 i 2
V—°"’"—*"*"‘“‘l·—; V‘ 2 V
’V i · °i I j
E A ;VV; ·;; g V 2 2 2 2 ¢ ‘V — ’ ‘
V E _ 2 ; Q ; _ i.€’;;?:?i-ä: Ü
6 E ·; 2 5 · · . = l V ä ' _. Ä‘ Z ’
E i?ÜÄÄ;.*‘§=;v;g_:g·;;;;:
;’ ' ' '
” ' - ·-_ Al &,_
ä - ; ·:*’“‘+‘—*"‘ . ä E E VE ·¢% N°” Er:
.-‘ E ; ——-·-..__.. § g é 2 V E Z I —
H2-222 ~ _ V‘ 2
2 ä = z Q ; § g 2‘»
cs
r S+·—······——--·····+——- ‘ —«-— i
‘ ä; Q ‘ _-mm Ä--——V‘ A--__-___- ——--«-——- V ‘ · V7 ‘ "
5.*--;E1Z.ZT !"_'
*———· ··—··-———·————--2 · - ;.- V ...-’‘ ¥ ¥ = - 6 2 -=
L..2. _ . . V V · ; . , . .. V, ,_
> ””—°'"T'“"i'···—+-——
—-·¥ ———--...... ' Zg
·‘ eg
E - V 7**;*** V I 2'Y I -— f‘El2‘ >•· ··——·---——-— —-—·——+5 ‘ -3—— ——~ —»—— 2 Q * T 1 ? i ä
——— «V—-—----VV-„—_.;..l-i - ,.-.-°‘_ Q ä Ä ·¥ ? Q 5 ·=
M ‘·—· ····‘·—···—··+——+—++—-——— — --—- Q 1 V 2 ä -2
i 2 ¤-— ä :Z’jTjj"’"*—*——······ — ‘ i V ‘ i §‘
g···—+·-—+— ll =a
g 2 g :_'_l___‘Q____'°_:'___'?""""‘“""‘ ··*—···—·—~—· - é 4V .. 2 2 SI
—·—— ·»—-—--——-——- ——-—i-1;----...-. af =„_,
ä ä · —··· -— —·~-- .„---_.V. _ _ Q: c
—- g _ 2
i %’ '...- Q 2 ‘ V g ¤
;+—-.— ‘? ___ ——·· ’ g E „= "’
f * T —E——————— ——————-——— ” ‘ ; ‘ ; ; V ¤ ¤¤% 5 · ‘··· ····— ··— · Q I E2
‘a
’ _—_— " “"l‘“"**—‘—”·*·l———— V ° * _ I Z
_ V -—-————-——· V ...._..... .V _____V__ _ V V ; [ H A
; — =” '
—”_”‘—‘"“"“‘——I—"‘··—·————— E * nui f- l "
V · V"‘* ——'··—·— —-————V ‘
ä V Iz E
_—--_——-H " ' ’ ‘ ‘""*—· ‘·——·---—---——— ———--. .
t-______i_j Ö l
•;____ ‘ V_____i” ‘°“‘‘‘‘‘"*"*i—···——·· —· j 2 V . § ' E °‘—·ff·—*—l—· ~·— V V
...1 _ ‘“Ü”“” Ü‘°‘‘'''"’ ‘ ° ‘ “""‘ t‘<+*—‘··é·;··—: V-? I- 2 E E l ;
L —~M_M; ·—·——--··· · · —-——-— V-——--V—7V..—?....._.-.t___ ,___ - V ;··, ; V ; _; I
' ""'""‘ —‘···-—- - - V .· V, . .
>
_ __ ___ N-I1
Q
____ El, _ Mw u A4— VV V . -V . VV
96
i Y 5 1 5 5 1
--
2 5 Q A
5 V 5 5 Q g . es* Q Ü E Ö 5 "
QQ5Ä
Q Ä . Ä Q_" 5***- 5
5 2 I iÄ
·¤5 . = 1 5 5 5°g 25ä&-§5--..5
5 5l- ,_
5 '5S5
;‘
SQ 1 ‘ 5 - 5
@5..-.2Qi E5
2 I; ae
· 2__
55 Q 5 E —— 5 % S I 1
· 2Q = 5 j
5 5 g 5 -25 l 5-1 —--Ä 1 2 °? :5 I 1 .2
Y.V E Ü- Y
52 5 5
E Q.
2--%5 55
V5 V V
5 .. V
Q 4;Q-
I WÄ 2 . V
5 Z'°55555**5*55 I Ü E
_..·· 5 55__55__ 5P 5 5 Z
2 55 .
"‘55 P *·-_55____5__ ___ ¤g_
5-...5
Ä'-‘
5 QQ
5
5555 2 -;255555S --5 --.---
5 Q
5- 555555 f -Ä x, 2 i 2 Ä {
i€§i‘;‘°5i5i *5i5i***.5555**- 2 i IlS~ -—55--5
5 5 5
i
‘—
97
; ‘··?T._;;__
A
. 6
--
_
—
—
—_
ie;-.=: 5
;.·Ä.ÄIZZEÄÄ‘·;·
6
*5;
1
-5.
—..__V
°‘·____·
_
5.____
—— 55
3
'-Q
5
EÄ i 5 *
‘—-.‘ =
‘ e
55‘
‘·—-_ ··55"—
EQ ¥ 5
€;.I *
Ä"·—Ä
ä5?1=.‘Ä
E
:
V;
T5
.,
—
,
-_
_
._-:
E56.:
é
z
:_
_
Ü°
6 ;„_
·—5~
1=
_ Ä
·A-_Ä“-..—-5,
' 5-_Ä’ _
.’—— ‘ .
9.
97+7 QQ
""" Q1 "
1
‘ 5_
_____
_
_
'-._
_Ä .
"-__
Q
‘·
I
-;3_5.__—-‘;;·»-._Q—55.
·--_ 5;
;·
'éé Q‘··——"——Q"—% ·· 1
2
_—".;_ "
'L: __";
YA' __
_ÄA'·_ .Ä—
I 'ZI
——ÄÄ ” Ä- ‘ ‘QÄ—“Ä__ “ Q _’“—ÄÄ”
__-"·—
9.
'Ä
es
-5 __Ä
·5_·° ;
--._Q“‘—1
__
=
;Q
'?6··~ ·"+——-¤ °‘
Ä _'
Ä"
--
__ ”'«
IÄ
Ä1'QÄ°_Q;
*-Q
I‘;
‘ —“
--__‘—~-_
Q‘——.__
°‘_°"·-_
._'··
I -6-
.V
~__
·______
__ ——,__Ä
0Q
‘.6
3---}-
-__-
—”~—-_—__;.
_Ä_
I
_
._
-_— _
_v—-A_
—-
6—·-Q;'
i‘~4-1__5l_;.jj—-_
*5.- T g 1 _ *5
Q,_
~_Tx_Q_
98
IIiZ_
Z
ZZZ·‘5 F Y LY Y
· Y
2 _ QY__Z · V; ‘ Y g g F E Q · _ Q QZ···Y:#;:Q?§§§f‘%ii:,§__ Qfj 1 v
im
‘+___· Y
_;
Y--·-_Z -
__
_4;”‘——-__.
Z _ Y ; E I‘ 6
an
1—
Y §‘!“2_Y,_·§Y·:;=;6.QLJ)
Z
Z · 5Y ·i
••
"·L.__QQl—“'5
.Y Y 5
I 5E
g
.+_-—Y
Y-·_ Y 6 E
g
Z Z Z ZZ-
Z Z Z ’**·¤
—+ -— ·
- E · 6·-·
Y —
ZZ"·——
Z
F--Z
F ·.
E
ZY Z YZZZZZ
‘1
ZZ
5 6
E
1_
‘· 6 6
—.._
I.”“·——._
-
Y ,
Y Yä·i2.&2Q11;jfZ
¤¤
*—_
‘·--—————;..
·-l 1 Y·Y Y:
Y
-;;_}
2 2*-Y-
>-
~—_·‘”—
ZY-—-.._1 YLS I
FYZE:6. . : -=
—-__——_I"———..._
°°—;;—;
Y L T 5 Y
9Q}; é
__5
LQ,
E;
——-._"·—Y-._’*—YL.1
Z‘*·-~___
Y · QY 6
Z2~Y?‘=.
"'
E
V-—Q—
'«,__ Q-
" V
.
§;
1
—— äY5 1 5 L
Ü4--
*1-{Q Y--1
--—Y —_
Q { Y
'=
Z—·L · 2 ’ 5 Z ; Z 6 Q ¤-
Ä·IiZ Z Ä
.
1Vw
ZQ"—.;"
I ÄQ E ;_Q
;·
YZI;
ZZ
"‘-_
.._Z“·—-._·
Q’Y':&·[•.
Ö
ZI—. T—-;1—·—-“‘. i i 6 g.6. g5;;Qg Q,
E.;.._
--_
__1‘ Y.‘·---
_I----
Y _; .6 _ 2 6.6 ; a.¤
Z
Q66 YZ
L
Y 25.
L;——L
2 = Y +-2I
Ä”——·-_~
_l~‘-—-_
"‘..._'
Z I1—I___Z_ZZ*-
_"·-_Z Z Y
;·
:
Y 6YY
Y
_—Z Z
ZZ
Q YY1
2 Y 2 . L I . c
.;*---61
Y‘———-.+_......
Ä Y f E*16 Y ,
5 ; 2 ; I · — E
1*-._-YYY1-—Y__
Y I Q Z j _ '7 =
1
6 2 6 g I*— 1:
Y ..
Q—ZY·___Z
··-L,__Z--5-___'·-
_
---—;—‘j~-.____
Y
HT--.
1_
Q
5.
·
-
Ü
·- .1
Q
Q_j I2
Z
6I
L
Z-
1Y‘‘ - L.°Y
Y Z i·
_:
1 ···
*Ä} é— Ä
I _ I-I
99
ä?1§=·€2.1_.'Q°Z5_1·1
-
;;.1‘€2‘
;_F*2;_
;..
--;-212;-.21;-6
Q—
‘ I ? I ¥ 2 I ‘ Q
1 ---9 '———— T ‘——--I '~
?”'T*--+9-- _ ·
*--—---;__I 6 ·¤
-—
I
—*I—IIIAI~*•—
I EI °-
2;; I
4-
j*****9.2
‘*—-.___; =_ I 2
ä
2 - T'- -4;
1 ---__ _ ; , 1; E
2 E
‘· 2 .I ;"*·--_ ·
-I"T*-~-.;_Q_4
1I Q.
E
---9
I I-·-—--E Ä E
Q —- --.9 1 - -66--...‘ 2 SE
2 . 5
Q 2 Q 5Z:
*IIII_
= 1 1 — -‘
l ~ -=
I ‘*—III'° ·2¥·é 9
__ 2
I1;*1;*·-
--91 --49 · é Q ·¥ ;°
°—‘— IIIQ ;·—-—
I;I °
Qä-2 (
; · ——-—— --1--*-9.9
"*·—9_
_"T
éf E; -¤
2=6‘·°°I—
°“';——.-.-4-*
°·--?%I
1····-—--
-- Ü
2 -
;;·i22‘: - Q II—AI—_"i'“'.___
T T';
—--;_
Iu :
---9
--g E
II_
III ä'
I
"·:4
--_Q_ ...;_; T- —..____
;·—---.__—-___I
__Ü
——
"'
III
I
__Q" · . _ 1‘ ----______
·—..___—_
_I
Z
In- '
—-—;——I
I
- -.
{ -
=
I 4;.Q';‘‘; 2
2 1
”’“;·-——-.9_ 6-•
T ·I ; ‘ ;··2
'·-{II j*·$--4
9__lI _·
1 Q---9
1 _E
T
.
II I QT; I2-;1j_ II-Ä Ü ’‘‘‘‘ 9
_Q
1()()
.
' -' ·l 7
'Q Q
I' Q · Q ·
___—
X Q
5§22i§
ä _ 2 2 f*·
Ü' -
Q Ü 2 Ü 2 Ü Ü
’ 2
-———-_2 _2h+._ :
:,7
ÜÜ
2 5g
,_
,_____Q 2 4--—-- Ü Q Q Ü: E
+-22..44_
ä4444
·h-_%___
g‘++___
‘ Ü 25 · 5
gQ
· Ü g5
Ü1 Ü Q
Ü
T-_’—
Q Q
1;Ü~.,_Q_- .1 Q =
Ä ; Q E Q” Q ·
,_Q Q 3
.,•
Ü — Q {”T
_
—— 5 ZQ 2
- ·ä
Q EÜ 5
Y3
gQ Q Q Q Q QQQ
YÜQ Q Q Q Ü?
Q Q Q Q Q QÜ
hg
—;ÜÜ Ü Ü Ü Q ÜÜ ÜÜÜ J E
Q-1
ih hr 4
Qi-—
—'*_‘- .___
Ü "·_
-? Ü
*3
__....i‘
...__.Q-—·—..-.
h--— . * Q"
9
Ü--
Ü- ~—Ü' .._ _ Q Q-
Q)
QQQ--
2 ·· -2 E 2 E ÜÜ 4&
—2ÜÜ
—- Ü‘“’·—— Ü EU
Ü -·—·—2-_; _
____
Ü E Ü"'
ÜÜÜÜ22E.44-i·5—2-;;~22-__;
2*2--··;Ü„iigé—__·5-——._
Ü 2 Q Ülé 4
2
ÄÜÜ-—~—
ÜÜ
"‘-—-.„Q___
„_____ "-
+——__·—22-__L
Ü Q · Q
—L
"+ E2
-——~2„.Q___ ·~
-.____ ä g 2 .;§
4
-22-Ü2 -§- 2
.
Ü Ü—{._QJ--
2
,:
T
Q
-+‘Tf E
2
5,
T
Z-
‘——.._._
- 2 2 Q _ QT·--V:2;;:-,.
=¤
.
Vj--;
---4-
++22 · 2 Q
g
F--2 3 1 = V
.52;«
+2 2·—
2---+-
2 2 2 2 1 2
Q1--;—· 3
‘-—Q T Q - Q 2 Q 2 222*2-2
5
Tf
T T Q 2*22 ’°
°
‘T’“----_
T__T’——--_
.;
= = Q { E ; 3 ·;_
2
>•
Q””;T T ··
2 · ·
2 2 · 2 25
T'
‘·—+;_·+-—
T "T V 2 Q
•·;
---
2 T · E Q
2
; { .? „
'fff-__
T ‘
g
2 1 Q
E
TT
2 * Q2Ä
°2 2
T
Q jf---QS
——-.1__—-._ — i g 2 ä §
7,
-‘ ·2 {
2 =
E · Q {Qi Q
~’‘2·· --_
" ~ -~+_ 2- 22--; . ,
12- Q-
T
-.1
.
‘·-_ _
"·
_}
"‘-. V
j- ___.
F*·-
_;;;_j_°‘2—-—-{
‘ 2 ><
V
2
2 _g
T
+_{__-_T"^—__-A "2—.
··—f‘
QT ~
:
++
V T 21..
T-f-+__·
- *---2
_—'·..AT .
_‘..+
_"—_--__
..
jQ
2-
V{
Iu
_‘—-___;
-——'-=.i———;”-
T””'·__; T _—‘—--_T-T’·-~-
_—._
4*--—— 2 =
Z4:]
”j·——— ·l-T_Tf?·é ’22·
21-:
11 -'=
„_
‘·—-..‘—· f---__ .
2 ·-——-
’Q·—._=
—— es
·—
Q
I
QI
T' -_
;_j
“,+..—T'··-3___{Y
ij
°‘~. _
I:
1- ‘2··2~T
_- TI E
;1· TT
-2--3-;-+-2 2··—---
2--—— .1_2 2
I..
--
__
Z
—--
u;-Iä
TT V;
T_-_
— *12= i
„
..
__-72 ;;~
E
__102
A
Y**Y—°
I Y ~F1 6 .
'Z1’1=.11: -
6 1Q+--____
Ü 6 ~ 6 1 F ' F 6
S - Q—-—Q.-_
1 1 6 1 6.; I I 6zz.
Q; Q
6 6 éYÄ.
Y ' I · Q 2 1 6 Q Q Q Y §‘ Q F 6
Q
6 F 6 .Y 6 6 . ; QYYIYY
6;:?%i-YQY§’Q§2:‘i*£6éY•·•
I.Y
2 6 Q Q ? Q
"°°——·-·———--._I . 6
¤-
”YY
6>»
"S"1··6——..Sa~,_____S ’ F
‘Q 6 E . Q i j 2.
I1. ä
6 Y 1S T. 'S.
Q Q‘1———;2 Q Q
Ü! Q
·—·»———-;Q_ 1
6 Q f j 2 YS 6 ._
Q
1 1 —· Q6 Q 6 2 é..Q.‘:§: 6..2
I
j -=
~
'°"—{~-__ . F j 1 i 6,=»· E
S-ä
—·· -
_
YY YI
° Q YY j6§§E?Q§
•-
6
6 S 6
LÖYI. S
1- S 6Y
S 1 ··——;S 6
S 6 Q Q 166-:6::; -=
- —-,,. ..
—’—1·——.:Y’”*—·——YY 6 · 6 1 X Q 2 E-
..
_I 1 1 —-—-- QQQQ Q3 ? Q Q .I N
YY QQ Q 6 E“ ·—·_Ii‘11--#——-_Q
YY"· _ I z ’ Q g ä. A
-I°‘SS-1--—~-——_
.—{.--__“”—; S ä 6 l ae
‘ -1---..;
6SSSS {_ Y I ’ Q Q . E 2
‘ YY
-1-—____%__— —— 2 6 g
Q ' ' .I ‘*——---.__
—1i__
'·——---______---.- 6
·Q6 ,6:,
YY
S11---- Y Q ; .91
I '*
.
——————-_._IYY Y --—--..__
i——‘i_...—Y"-·———--
Y
>·•
YY 6..'*——.1;I;·-~—-Q-
. "1 Q Q r 1 l Q -=
···—-.Q-Q_—~—·——.-„____;
S1161 -——-.é__ I
Ya
—Y“11—·-——-.
”“————;__’j.
’ YY 1~—
QY°“11—·———«..„Q
T‘S1‘————...__ .Ü”Y'*-YY”———— ?
E°
QY "‘“‘ ~-IQ
YY ' 6Q _ .
„ E
·—;;..;___—-—..;.;j-—+.-Q.i
.Q *5
-·—-—-.-_____;—___
————4__Q___
. I 6 01 $16
·
Z 2 _ 6 S 2 1 U
1 · ~ Q- .. Q—— .1;;
2
6 6. Q Q 6 •-«
1Ü3
11
·
é1=l?5¢SE5A.-
'
f · 1 Ä : S 5 ; ES 1 ä
5Ä
’—+-- 5T
-5
Ä
——
S A"*
Q;.7Q—·7-
—-——7 -
‘·
711———_j111··—-__
+——_
-°“———-_——· 1 E .
1 ä T I. *'
1
ÄÄ——···— ÄÄ
Ä
——
s:
—ÄÄÄ'”?1--_§Q—‘——¥—7-——
—”Ä1—1'1—+_
Ä;Ä'7—Q“111·—
Q'‘‘‘1--;
ÄÄ Ä ·
EÄÄ, 1:
QÄÄ
Ü Ä S Z "g_
1
__ 1 @ A f1 5 ? { A
—_
"Q>·
1 —-— 1 A
A _—- 1A 5 ~ v
}—_;_711*
7---._Q°'· -___—·.«___·
A1_7Q- ä E j 5 = éE
77
1 { 2
1Ä ÄTTÄÄÄ"7-
1· E
__;L·'1—-1
1 __°1—
1 A__1
7-¤ .1___-_
..___
1--- L
E
_
_ _
Ä1 ._
Ä_' -—_
—·-___1*-.
·---_ÄÄ’“1—·-.-_
5§ _ I:
AAA Q
-.-_____ AA 11
1; A
1_ rn
_ÄÄÄ
Q .
_ (7
·ÄÄÄ7—
¤¤ ==
*7
'·—-.ÄÄ
7 —-A,-
-_—_— Ä Ä E
C-;
TÄÄÜÄ7-
71ÄÜ;7—~7Ü°111·
-1-ÄÄ 1
Z——»
ÄÄ’1·7Ä
-_·‘·
—_LQQ.
__
Q- ·-,__Q_.—_Q
ÄÄ“ -. Q·
_ÄÄ ‘· .
LQ-"·»___—_i
E!
.;‘
.;
Z
*1 -5.
ÄÄ
-*7
Z-__
ÄÄ-Ä"11·—Ü
-
Ä- s.--11
Ä ÄÄ7
ÄÄ¥111ÄÄ1ÄÄ*#1
777äg 1 E
-~--.i_
.1ÄÄ 1 4;;1·-·
Ä'i;1é·—l_·A"--—-_1
1‘·—;lT1
.1 J:.
___-AA.._ A. Q
A
AAA..1
.-1
1·
·~
3·
._
·Z1
1 AÄA
Z
._ Q
, Q·· . _
-—___—
-..I
T·- ·-
Ä 71 {_— E
1·Ä '1
—7.Ti;¢
ni
‘:3.
3-
—=
*_
*-25
2-—--
•'
*2;
A
-EE=EEEE-
--—
ggN
Eägigä
*2:
Z
ln
:.
EEFEE 9
F»
-'22
--.12
‘=__;
Z
FFF°F
9
2__
F
¤'
--2
Q
*— II
öl
--2
2:*
I
5;
:—
1hI
gaäägiääc
__éä é-
.=
.:
.,N
*1.
-1:
2 2:
.:3Ü2
-¤
5:* —- S’2’522§S§.=22ZZ --2-
,.:3;;;.525252Q
I';
2.l-
n
-.;.2
N
'ij; 55%
1
2%
„:
gg- --
;_ _:
Q
., 2E. I- 22.lE
6
G
2 E.} ÄFiÄ
' =-l
Z
i':-‘= .
‘ 232: . 2 : ¤* . . 2
2 2--r 2 2
2*222 2 2 2 2 2
N
2 2.
2* 2 ä 6 YI
--22*
2_ ·-.; = E I
_ ZI 2 2
•-
2 : =2
I
9
i' - Ü 2" 2 2 F = 2 Z 2F
Sg:
Z -F
F 22
2Z F 2 E .
F -Z
-2 1. ;_; _2 ·
= $*::;**2*-3_
6 2 · ÜI I -Z
2*E
lF2' F Z I 2 F I F 2 -F -=
2 2
2 2 2 Z —- ; 2,....
1-
2 6 2 -2222 Z2??Yf?ii
¤-
2 2·‘ 6
2 2 6 I I2
.:
5-* -2 2 .26 . 1.- 2 ·2 — I 2 1
‘°
2 ~ :2,; _ · · Q · . Z
:..2Y . ¤ 2 :
. 2
-=
,, QJ ; .
2 E
2 ·-
j ;2
FFYZEi,_._ * 2 2 , ;. I g
E *i FE-’ 2 E
Ä2 ;. j i
I FF·_;;
; Q.; ; 6 2 6
2 2--i
"‘
2 —?Y22‘2 2 -22- -2Z-i.é2?2222"Z22222
2 .2F 2 2
FFFF
FFF 2 Z
‘· ·.
ÜFFZ
-2 .2
"‘
Z2 z
6;- ·;; .,j2-;
1 6—.._;j§2.—·,-..;.;;;-5-...2;;;,-
g
-—-;
Z Z..63
‘ 2 1::-5*;;;:
EE· ————§§‘§§:‘;—‘;E"—°°°E=”—°-E=;1—””—°:°°°°::EE°°°°°ä·;;“:_:§§i—·——ää§:E?:”ä LEE"-§E°E:—EEi—EE-——-·.E~—-··~~~—-•—·~:--—-—-—·;;;;:::·—·——···—•; ....
EEEEEEEEEEEEEE:E‘I=*"—?:E*E'*‘EE——==EEEEE;;;E;;;g;;ggEEEEEELEQEEE'“EEEEEE=E==EEE==_E;Eg‘ä.E;g;Egg-gggggg
E""'°;.E—E-=Z’;;F—"‘;§E;EE*€E.'“-;EE§§j§·§§·§EEEEEEEE;:E—E=EEEEEE=—=-——E'=—EEE=—·E§g§§5;+;E_;g;;ggg;EEEEEEEEEEEEE—=EE"=='=EE==;.;gEE;—E.—=EEEEgg-;-EE2EEEEEEEE-·'EEEEE:;—=-*I‘==EE=—*—=EE=E—E—E—=-g—g————
—L.Eg=E§g§ä§ääEE§äEE—EE§E:EEEEEEEEEEE;—EEEEE:E*—;.;;gg;..;;E;g;äääääääääääéäéääéEE;-EEEEEEEEE »—‘—=:‘;::—T?—1ZT—2?E?T£ElE·—·:E.EEE.EE.—;EEEEEE—.—:§§:.*·..=;§;*%—;g;.;E;_§§_§gi Ü2:;;
Öl:::.-:;:::1*::-.‘T‘“’—”=:“'=—’=_-“====——
5EEEEET-TE' —EEi=‘=EEEE——._gT__ä;;§§§§§§_;§;; ggggg ....gEEE**=:* · EEE‘E——— ——·L——===°'-:L‘.-—='_·-i;—·°:ÜEE§EEE££EEé§E 1.EEEEEE
’ E§EEggggggggg;gg c
EEEEEE . =‘——*“..--==.....gggg;;gggggg—E2E;=;ggg cEEEEE—-— ==E==*'°··—*;E——EEEEEEEE-EEEE Z-'E;·TjE__E'~EEE;-'EE UEEEEEEE 'E"==E—LiE""EE==E==EE'jEE.ä::‘¤"äE;‘;.g.;=‘gägg >•==EEE—E E · ='7E""‘..E1;?.EE-E-EI-—L;;.Z—..:~·lF’.:r::-_;.§.·:;;g;.;; ggEEEEE =—————·—-—ä=EEEE;—E;;;;EtEE>__—;;g;;gg;gg EEE7EEEEE.. ..—E';==EE3—*—=—&·g===EEr**:;EE—;—·;-g:- gggggg-ggg; „cEEÄ-EEEEE -- EEEEE~—EEa·„*=-:§—§5·**§E·—E.=-=— "_fQ’EE§§3§?;E§ ggEEEEE EEEEEEEE * E=EEgE?„*§EE~;2;== .EE====„=„=„=g„„„——„——Egggg-2
~EE 'Q"‘“""""""""'“ÖÄÜ"E=""I.‘E5E=E gn
.=E'EE.EE5***··E*- EE*E—=£EE*g.§§§§gE——g·— ‘_ _EE—- 2:;-5 :====='='““==EE='E***g§g;§_E=== 2;;. Etäääääz-éEE‘:.;‘*.EE§ Q
€g§EE= _ L E§I=E§EE§§ggg;g;EE EEEEEEEEEli ::2:: •-
EEEEEEE "=EEE§EEEEEE§EEEEEEEEE2EE uggg :*x; I
;:—=·E&·E-E=°EE==*é’§E= -.._-:_—-?=§=;__E䧧 E; gggEgEE= „
gg v;-2: Q
gggggg== · E-— _§§EEE§gggg_ggggggg ggggggg §g„—ggggggg; gg
EE "5EEEEEEE _ -2-. -2 ..... 2 . EE E. °" ggggj=
=L *-ZT *5o‘EEl=‘·= nr :*2*n*:2=:======~=—--&~;$=‘= . =——*··**—EEEE' Q-
Z-’ .......E;E..... Ei'.? EEEgg;§g-· M
.1 Z. .'‘'''T' ‘ZI'-2 L 2-2- Z ZV2* sgsaagzagea--2-22;-6 z222.2-22..T? 1- J Jil-- Z? .2:.L-. 2E;
E L. ·E· ** *2:.*2:-..$°*‘äE§é§§i=?€2=22駗-E;..— 3 .- 2-22 ä.2=-2 * gg
_°'
E.."‘g=.—-:..~"‘.... ··=-= EE;-=E;— °‘
""—2?°§*.21¤¥E= ..€E‘2rgäwlélév2.2222 esggry 22-22 .,..2: L=- ggE=_ggEgggg2ggg
* ** ” ::7:.. :*3::.. :.:,:7 **:3.. 2;..:,. ::2::1 E? E:
Ü2i‘Ä”;"";"’; "*fir.§"1T'i.l.f'iIE?i':?j*°iQ,;;;:·Ei'1Lf’ET!j. gg;2 .Tj;gg:„_;;t.g3*_;;;ä;:;j.g;j.. 3:;;;::*2: * :2* *:.*2*: 2 2:: n.:r;{’€r* .::::.2 ...:.:.2 __
‘· EE' !°' 'V " ‘‘ . .. .. *'**‘= E
= 2 2..s· 2 1 E·‘·*··¢·z·t*r·.;*·¢<=*—'*“E*~-··-2 2‘""?*’“= "‘6E
2 ‘ §‘ 'i’2*‘·%%*-‘=‘ 2- " LL EE= L 22-+-2
‘- ~ —.·-··· . ·; · 2- S2 ¢;‘2i"Ti" F‘--- ·-äE~E:£E'·%Ei.;E&iL&I.i..:.2 ;.26§222ia;-2;; 2 .2 ;- ’E %—;gL........."= —— .:.:*2* *' E·····
·‘·* "’”""‘ ’ °" '” 2 ’ """" '
—+
ggg;5,
E-
-¤·:IEEEE.-g. .. -,
q"EEEEIZ “'
___
*52;1 :‘
1:;::;
7
>,.="''°3-*
Y1zggääa
"°
2,gääsat-:;:3; ·· *-
**gggaagf éääié
3 ééäééi-:;:.2;-
.1;;;:*Egg;
=- 1.;;._*
E1
:13;-.:-*-.* .22323-:- 8I ;;cg‘°lg „;.1 .11.
zg
¢·=- -11-1 ‘€
ZQi :1::;: if 7-__
__—— *,*7 -1-;:1:; 2
:· :· -5 - F; :·= Ef:-_ ; :.;:21%.2%:%: *—* ¢‘=="*-"'—'“* gi gs=ä itZEL=§F=Eä Zj :—
.. i.i7:*"" :· E :..1: :* ¤=‘ F E 1?—::......-:1:;::.-:... .--...--—--·;;;_;;; _···· ;..·.1 :=- --2;;-;;::v :---1; N-1 *1*:.I - I ZI- I-
*·gg1.
;. -3:****5:;äläfi
*31j. —·.1:...
.I II.1 ;· -;1 :· -1*::--+- §¢"§,1—1€E.1.· 1:::- ,~ rg " Z _; ,, ;1. .1;.---
107
_Y___ 2E
1 ; 4 ; . · ii§^‘ E
—~——'_V“T°_° Z 4 Q 3 ; _ = Q
- _ 4 . ° Q 2 . i L :1..:;; __ .: E“'““°i‘°‘*”"’*‘Q— L Q Q L
3 : · . Q Q§’Ü’L’jj°'=
;·· Q 2 - Q Q.: i
2 Ü Q LL; S FQ jjS. 1
E ·· ‘¥=‘222
2—*‘*+··—·——. 1 _ Q S_’‘°"‘““’**Q‘3Q*—‘————————————.· ..—.;," T Q
L QQ Q '¤‘ *3*- QQ S··—···+-——-2 ·--~+———--— 2 3 222 2 ; '·—" 4 Q
‘ — S SQ S S E E Q
" ’ ‘ SQ ‘—¢—·—%=—++———--—~—.—.. · S § r' 2 2
-;-2*** ···——;+·————ÜQ Q .—·———.Qi Q Q g·
...QQQQ‘·ääääää ——-?-- E"
Ü i—— 1j; 4r* 2 „.S} =Kg 2 E
··———--1--__rw--
g :· -4 · *2 =
___—-—- —~-——---..-...-.-.. .„__ _______1_______ _ 4 __
E_ , z
gQ 1:
" ' ‘ ‘*—·· ——-——~—-··——-.—..._.____ _ { ' .
‘Q. aa 4
l -——.-
—”—"-?"""* *‘**———·—<————- ""————~___ %
la · ¤Q--—...Ä..Ä.4.
ÄQ;Ä'“'°""’L“"’“L‘L‘ °° Q···*———·———— — E
. . . .. . 2. L 1__„; "“”""Q**—·Q — - Z-——————~ 22—~-Q-- ÄÄ”.ܧ..Ö
‘° "”“—* ” .—··· Q}
-3;..;; LLL4 4 QQ-1
1 N—""‘°°Q°"""QQ*‘QQ3——···———·——————— — Üujw I
QL
_ ___ °° "" QQ *3*; ·~·—·· - ~—-- _ ‘-_3
.4.4 L L 'Q"“Ü'Ü°'ÖÜ7Ä"'ÜÄ7TÄ;*”“”Z"-
~———=Q*?‘*‘<Y¢1”——§2·¥ QQ
T-~—-—«-—~»—-S--—‘-=Ä-;4
. 2 ,3___''''L
108
A QQ Q 2 g FäQ
EA; iQ · Q2?-11
A·2;€Q§
= Q ’f ä Q ä
Q Q Q ; i Q ä Q Qui?
1FA? gQ ä A jfi
E Q Z TE i F 1 2 5 A A ff 2
Q ‘€ 2 1 äÄ1
Q- Q _ Ü Q Q Q = Q 2
Q Q2-
1
H
1.
AE
Ä Q FAA1A 2 F 1 Q Q Ü E_.
Q" F F ä·‘f j 2 . A Q A Q
'1
A
Q A1
Q Q
Ai Q ä
WQ?
2AA’ AQQQQ ää E
*-+-+1Q 2 F1.: i ? äh c
QQ Q Q QQQQQQ
A 1 2 2 §1;fi 2*2 1
!” .5
22 = 2 ,_ 3
Q · 1-;·@
7'E
-- 1 i 5.;
“; z Q~_
Qj_· Q 2 A
A‘
Q A5 3
1.A1AA
‘¤
Q 2 i S U
1 A 52;
U $2_;
Z Ag
2 Q QN
A2 1
2Q N
109
7_‘A N
fi? ‘ "" E._-_... . . - · - Q
—— . " ' ‘ Q...- -...... . (Q. .
~
::.1 - L . M M - R—"’ ’ *0
‘
M -— ¤1* _g
MM M ;I7—T°4 ¢A ;.é-ii; N
·:— :i———;r ... -1----
"5- · E
=A - !-MMM M N ~· =B-
N
S> s
2 L
:=T:g:2=‘
EE-:
-- .,.1:;-::;.·-ii
2; -7+4ii
-«EE;.i*:1.r:·“i7*:..2*; xff 2~;.'*EE 'II EI.}? ILE _c*‘*· Q'·°’ if *373 __
'GF-2*2* 2- 2* 2
1 ;_ 2 ;_Q.;*2 1.* 2 Q2--? · .=.·=·2--2 *;.2 22 Z- 2-* 2 M
-_:—...... 2-:-- ; q_
-_ 2 2 . VL- _;--; . .. 22 J- ·»-~·-~ - ·
__________7;;.£-'QEQYQQÄIQ ,-2gi-; -2 u
Q;2 22:2-- .2-2.. 2. -· 22-S32?·"—2* 2* i
‘.' Z'. 2 .2 ‘-°Z-LQLQY-1;},--2-„-2 2 2 2 ;„
L7.2-2 2 2 ;2.24+*·
2E 2E22-2-1 * ‘ I. Y ° ZE- -2:- - .222--·:.. 2:: EE :2;-2222-22:2%-2:; -2* 4;-2-122 .2- 222::-- .-. _¤
Z Q? Y? =*2-- .: *2: g-.. · 2 2 2 *2 :-22.:2+'· *2 *2 *2 ‘'‘‘ 2- · · 1 :.*:::;.:2:. -§ 2-
’Z== 2 2:22 = ** 2---=*-
. -2. . 2 E--.· 2 j 2 2 ° g .jj:I‘?2f 22 2.2.-.
*2 2. 22.2.2 ää 2-.= 2 2:2 2- =* *2 2 I¤*¤·2= 2**2 2 · : .2 * 22 4;
2 "· ·‘
2 * 2 V,
**12 22...22
·= **-.2 2 2 2E · 2** ·· 3- 2 ..1..1 Z F U U ’.: F 22 . = -42 72--. L2- i·: : 2
+-4--;* pi·— Q EE 2; .
L:L’= "2L 2 1 2.2.-.2
22;...-.--22
?" 2 . 2- 52*2 2 -4*+ -·g * = 2r E - j 2-.+.2 2 =
‘*Ä I
iig
Y_....-;,,,;*;*122:;* 2-
.. == E;
==‘1
#66 #21;;.6- E. Ewggägifg
-=71==5-*:;:
-2=;=116
Q'E
EF1 .. 1-..-..;-.. ....- .... . .. . ......-.-- .1.-... ,.
Ö1
.. :•J- . .1 .. xr:2-
QYY 1
‘_
2- ...2-2 Y; 2.-2 2Ezäiäi 2
.1.-E 1
,,2ÜÜI5ä— 2F E 3 -- 2112 2 2 FE" $12. ¤_
2 *1:12 F ÜF *2 2 =Ü .-112 21213 12.5;-$22--11 cn
Ü‘E —EÜ 1 ‘‘‘‘ 1 .:111 1 —·———-————11* ---11 1;11jZ1E-——====— X11: E 1 * . 1 Z 22 . 2. 11 1 6 2 = -- ; 2 _;
35- Ä*Ü *Ü
-Ü
2 2 E—-Ü1 TÜ F _ = FZEÜ
Z. 1 -1 Ü ____ 21:- 2 Ü?—F 1 1 - t
‘2 1
1*1 . ___ Fi
2 11 fw; -1;:Y
FÜ é 1 1- 2_- __ 2 23Z 2- I :=
2 - j Y 2 1-61 ä 11 = .. 6 1 2 -.2: 17 -2.65 N1 2 2 2 1 1 · Y ._ L Y
¢‘
E — 2 .1211: 2 - ;_ S 1 Y 2 12+ 2 Z .1
Ei E 1 1 = 2 - 2 _ 1 2.1112 1; .7%. 21 %.2 Y' 2. ° 2- Z 2 2Ü1 Ü =* . = F 1
E 2 - 1 E2 2 * F - - ..2 - 2 Z--6; :1--2 211.:212 §211* iEF—§.1 6 2 . ä 1.s=:§=121 -— s-—2-.F '1 €“
2-2 - 2 . 6 .. 2. .2— L. é-1 12---2 E ;§ i' .2. 2. 2__1
2 * Ü Z 1* 1: ä 31-1_1 E i 2 2 2 El; ~S+4
E E1% 1 12 --2 12 2 :12- .. Y___ % : E.
.€·:2- 2-.1- 6 1 .. -5 2‘· *‘72 :22- .-2-22 *" 1i _ --...-= ;-
--.. ;
112 . ITGQ-‘ä
•~4.:ooäE
· Z1 vs11 ‘ «> '1
U’Ö ,'—1 ** (vg;i* - "'1'?1, 6 o ·11 ‘ .E_ 6 :1EE>.:— 2
. ?·•-I¤1*’
·n °'5E:n*8cu¤m
E¥E6_ öl
1.~
\1.
113
, — 0
. 6 . .ä„- .„ ‘- - 9
.; ., S ’ “.3
MMT
-S.:i"‘ ‘
é 8„ - ~ • ,;I Q EQ 2
.‘2
ä 6i E1
,.EES 26 >„5
• ää‘ S
¤
—— E*i" 3 :S 5
-ZQS E¥U
*j
S• . • ° N
1 14
l¢F5
G
2:N·1.
°Q1.GG•= .EQZ-6~abM'¤
. J1éEGEE
2 E„GG.
EGJ:cb".EG>.E
°.*2
co EGBE>a.G
E:L:G_ GG.va
Z¢I~ E
ZcbenFl
oéN
