A Synthetic Route to β-Hydroxytyrosine-Derived Tetramic Acids: Total Synthesis of the Fungal Metabolite F-14329

Article abstract of DOI:10.1002/chem.201701259

3-Acyltetramic acids derived from β-hydroxytyrosine are synthetically challenging. The first route to this structural motif, based upon a condensation between a Meldrum's acid conjugate bearing the acyl side chain, and a β-hydroxytyrosinate, N-protected by an ortho-nitrobenzyl group is presented. This group enables the Dieckmann cyclization of the resulting N-(β-ketoacyl)amino ester, after which it can be removed photolytically without compromising the delicate 3′-hydroxy group. This strategy was applied to the first total synthesis of the fungal metabolite F-14329 (1).

Full text of DOI:10.1002/chem.201701259

A Journal of
Accepted Article
Title: A Synthetic Route to β-Hydroxytyrosine Derived Tetramic Acids:
Total Synthesis of the Fungal Metabolite F-14329
Authors: Sebastian Bruckner, Robert G Haase, and Rainer Schobert
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201701259
Link to VoR: http://dx.doi.org/10.1002/chem.201701259
Supported by

10.1002/chem.201701259
Chemistry - A European Journal
COMMUNICATION
A Synthetic Route to !-Hydroxytyrosine Derived Tetramic Acids:
Total Synthesis of the Fungal Metabolite F-14329
Sebastian Bruckner,† Robert G. Haase,† and Rainer Schobert*
!"
2
!
Abstract: 3-Acyltetramic acids derived from !-hydroxytyrosine
are synthetically challenging. We present the first route to this
0
&
"
#
0I
0+
*+
structural motif, based upon
a condensation between a
$%
$%
&'('")'*+'(',!-
&'('")'*+'(',"-
&'('!")'*+'(',!- 456789:;<;8%'>',#-
./012*3''',!-
456789:;<;8%'=',"-
J
!
2+
Meldrum’s acid conjugate bearing the acyl side chain, and a !-
hydroxytyrosinate, N-protected by an ortho-nitrobenzyl group.
This group enables the Dieckmann cyclization of the resulting N-
(!-ketoacyl)amino ester, after which it can be removed
photolytically without compromising the delicate 3’-hydroxy
group. This strategy was applied to the first total synthesis of the
fungal metabolite F-14329 (1).
"!
"!
"!
"!
"!
H+
1+
!"
!
!"
$%
"
"
"
"
#
J
$%
$%
J
!
$;:;C6B;898%'>',%-
!"
!
!"
$%
#
#
#
Tetramic acids are distinguished by a range of structural
varieties, biological sources, and types of bioactivity. Numerous
total syntheses and biosynthetic variations were reported.[1-3] 3-
Acyltetramic acids derived from !-hydroxytyrosine are typically
produced by fungi. The first derivative to be isolated by
Hamburger et al. in 2003 was militarinone B (2), a metabolite of
the fungus Paecilomyces militaris (Fig. 1).[4] Four years later
Sankyo Co. Ltd. patented a biotechnological method for the
$%
$%
!
D9:E@9A:6<%89:'>',&-
!"
!
!"
$%
$%
$%
!
FB9C9C%8%::;8'G','-
!"
!
!"
production of F-14329 (1),
a
metabolite of
a
fungus
$%
Chaunopycnis sp., and they recommended its use "for the
prophylactic and therapeutic treatment of obesity, diabetes,
hypertension and ischemic heart disease".[5] In the years to
follow, more compounds of this class were found, most of them
while studying the biosynthesis of pyridone alkaloids.[6-8] For
instance, tolypocladenol B (3) was isolated from the fungus
Tolypocladium cylindrosporum by Lou et al.[9] Capon et al. re-
discovered F-14329 (1) together with closely related
chaunolidines A (4) and B (5) in extracts of Chaunopycnis sp.,
found in the inner tissue of a pilmonate false limpet Siphonaria
sp.[10] Prototenellin D (6) was identified as an intermediate in the
biosynthesis of the 2-pyridone tenellin by the insect pathogenic
fungus Beauveria bassiana.[7] Epicoccarine B (7) was isolated
from a fungus Epicoccum sp. dwelling on the fruiting body of the
tree fungus Pholiota squarrosa.[6] The structures and
configurations of these compounds were determined by
spectroscopy, X-ray crystallography and circular dichroism.
Despite their broad spectrum of bioactivities, no synthetic route
to !-hydroxytyrosine derived 3-acyltetramic acids has been
reported, so far, probably owing to the incompatibility of their
benzylic hydroxy group with the usual protocols of tetramic acid
cyclization and acylation. We now developed a synthesis of F-
14329 (1) which is flexible enough to open access to this
important class of compounds.
$%
$%
$%
!
?@;A9AA6B;8%'>',$-
!"
Figure 1. Structures of !-hydroxytyrosine derived tetramic acids.
Scheme 1 delineates the retrosynthetic approach. A late stage
Dieckmann cyclization of an aptly protected N-!-ketoamide 8,
followed by mild deprotection, was to afford target compound 1.
Precursor 8 should be accessible by N-acylation of a protected
!-hydroxytyrosine 9 with a suitable !-ketoacyl derivative 10 of
the required side chain, which could be a thioester, or a
Meldrum’s acid derivative. Amino acid 9 was to be built up,
according to Schöllkopf, from bislactim ether 11 and p-
methoxybenzaldehyde 12. The side chain precursor 10 should
be available from an ",#-dimethyloctenoic acid 13, the
stereogenic centers of which could be established by two
consecutive Evans alkylations starting from hexenoic acid 16
and proceeding via chiral imide 14 and triflate 15.
["]
S. Bruckner M. Sc., R. G. Haase M. Sc., Prof. Dr. R. Schobert
Organic Chemistry Laboratory, University Bayreuth
Universitaetsstr. 30
95447 Bayreuth (Germany)
E-mail: Rainer.Schobert@uni-bayreuth.de
These authors contributed equally
[†]
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/chem.201701259
Chemistry - A European Journal
COMMUNICATION
"'()
The major isomer 20 was cleaved with LiBH4 at 0 °C to afford
(S,E)-2-methylhex-4-enol (21) which was treated with triflic
anhydride at !78 °C to afford the unstable triflate 15. (R)-4-
Benzyl-3-propionyl-2-oxazolidinone (14) was deprotonated with
a small excess of LDA and the lithium enolate was quenched
with freshly prepared triflate 15 to give imide 22. Its cleavage
with lithium benzylate followed by saponification of the resulting
ester left (S,S,E)-2,4-dimethyloct-6-enoic acid (13).
):;<&8,7=&2=
$:;>+&26?@..
#"$%&
%&
%&
!
!
%&
'()"
'($
"
"
#
):;@2A5@=&
$:;8,7=&2=
"*
"
"*
#"$%&
!*$
/
0
%&
%&
!%
%&
As we intended to generate the 3-acyltetramic acid moiety of
1 by a Dieckmann cyclization, we converted acid 13 into two
different !-ketoacyl derivatives 23 and 24, suitable for an N-
acylation of an aptly substituted !-hydroxytyrosinate (Scheme 3).
!-Ketothioester 23 was obtained by reacting acid 13 with 1,1'-
carbonyldiimidazole according to Moody et al.,[11] to give the
corresponding imidazolylacyl derivative which was treated with
the lithium salt of tert-butyl thioacetate. For the synthesis of
derivative 24, acid 13 was activated with EDC#HCl and then
reacted with Meldrum's acid.
*"
$
1234556789
#*"
"%&
!
*"$#
%&
%&
"
%&
/
!
!&
+',
BC@.D
@56A5@=+7.
"%&
"-.
"
!
!!
!"
"E9
"
/
%&
%&
!
!
%&
-.
!'
!(
()*+,)
#"
BC@.D
!$%&
"#
"#
@56A5@=+7.
"#
!'
"#
"#
*"$#
!"
!
%&
!)
-)
!
#$
Scheme 1. Retrosynthetic approach to F-14329 (1).
"#
"#
!
!
"#
For the synthesis of the 3-acyl side chain precursor 13, but-3-en-
2-ol (17) was reacted with triethyl orthoacetate in an E-selective
Johnson-Claisen rearrangement to afford ethyl hex-4-enoate
(18) (Scheme 2). Its saponification gave carboxylic acid 16
which was converted first into a mixed anhydride with pivalic
acid, and then into an Evans imide 19. Its deprotonation with
NaHMDS at –78 °C and quenching of the resulting enolate with
Scheme 3. Syntheses of 23 and 24. Reagents and conditions: a) CDI, THF,
RT, 16 h. b) tert-Butylthioacetate, LDA, THF, –78 °C, 1 h; 46% over 2 steps. c)
EDC#HCl, DMAP, CH2Cl2, RT, 30 min; then Meldrum’s acid, 24 h; 99%.
The stereogenic centers in the !-hydroxytyrosinate were
introduced as described by Boger et al.[12] (Scheme 4). Aldehyde
12 was reacted with the (R)-configured Schöllkopf bislactim
ether 11 to furnish the separable diastereomers 25 and 26 which
were hydrolyzed to the optically pure !-hydroxytyrosinates 27
and 29.[13]
iodomethane left
diastereoisomers.
a
separable mixture of two product
"
"
"'
+,
-,
!"#$
3,
"
*
%&
$%&
!
$+ $%&
%&
%&
IC
&')*$>:K+?(:+$
H
BC
!
;C
!'./!/0/12
!(./!/0/'
'!
!"
() !#
!
!
JC
!"#
)*$
!"#
'(
"
"
$%&
$%&
4,
&,
7,
''
>BC$(D!%0>?E@(#0F >BC%(D!&#>?G@(
-&.'#',/&0"!#01/'-2032#45'647#'.28
$+
!"
%&
"
*
%&
%&
%&
$%
()
$!./!/0/'
!)./!/0/56
$+
6,
,(
:$;%&
:$;%&
"
"
F
'"#$
8,
!+;
!+;
%&
"
*
%&
9$
3(
)*$
%&
%&
%&
%&
!"<090=0)*#>?;@(
!$<090=0+#>AA@(
!(#>?L@(
$$
!&
()
Scheme 2. Synthesis of 13. Reagents and conditions: a) MeC(OEt)3, EtCO2H,
reflux, 1 h; 73%. b) KOH, EtOH, H2O, RT, 20 min; 72%. c) PivCl, NEt3, THF,
0 °C, 1 h; then (S)-benzyloxazolidinone, LiCl, RT, 1 h; 89%. d) NaHMDS, MeI,
THF, –78 °C, 1 h; 84%. e) LiBH4, Et2O, MeOH, 0 °C, 35 min; 84%. f) Tf2O,
pyridine, CH2Cl2, –78 °C, 90 min. g) (R)-4-Benzyl-3-propionyl-2-oxazolidinone,
LDA, THF, –78 °C, 30 min; then 15, 4 h; 49% over 2 steps. h) BnOLi, THF,
0 °C, 2.5 h; then KOH, MeOH, H2O, RT, 3 d; 76% over 2 steps.
Scheme 4. Synthesis of !-hydroxytyrosinates 27 and 29. Reagents and
conditions: a) n-BuLi, THF, –78 °C, 15 min; then 12, 1.5 h. b) 0.25M HCl, THF,
MeCN, RT, 22 h. c) 10% Pd/C, H2, MeOH, RT, 1 h.
The feasibility of an N-(!-keto)acylation of unprotected !-
hydroxytyrosinates by a silver mediated aminolysis of thioester
23 according to the general protocol by Ley et al[14] was explored
This article is protected by copyright. All rights reserved.

10.1002/chem.201701259
Chemistry - A European Journal
COMMUNICATION
with methyl !-hydroxytyrosinate (28), prepared by debenzylation
of the waste diastereomer 27 (Schemes 3, 4). !-Ketoamide 8’
was obtained as a crude product not amenable to further
purification. An attempted cyclization of 8’ under conditions
described in the literature for simple unprotected N-acyl
tyrosinates,[15] failed to give 31, a diastereomer of F-14329 and
chaunolidine A, but afforded the retro-aldol product 30 instead.
afforded the protected F-14329 35. However, the DMB group of
35 could not be removed to give 36 without decomposition,
neither under acidic, nor hydrogenolytic, nor oxidative conditions.
This finding is in accordance with previously reported problems
to deprotect the nitrogen of a tetramic acid bearing a saturated
3-acyl side chain,[18] while N-DMB residues of tetramic acids with
unsaturated 3-acyl side chains were successfully removed by
trifluoroacetic acid.[17] So, in
a new approach, the ortho-
/0
nitrobenzyl (oNb) group was chosen for N-protection. Like the
DMB group, it should favor a cis amide, yet may eventually be
removed under neutral conditions. It was employed by Tatsu et
al. as a photocleavable amide backbone protecting group.[19]
Aminoester 32 was O-silylated with TBSOTf and N-protected
with ortho-nitrobenzaldehyde / NaBH3CN to give fully protected
aminoester 34 in 51% yield over four steps (Scheme 6). While
its silver mediated acylation with thioester 23 failed, the reaction
with Meldrum‘s acid conjugate 24 under conditions as described
in a patented general protocol[20] for the acylation of secondary
amines afforded !-ketoamide 8 (PG1 = TBS, PG2 = oNb). Its
Dieckmann cyclization without prior purification gave the triply
protected F-14329 derivative 37 in 54% over two steps.
1/23%
),
3%
3%
&!
.
3%
0/
0
/
/
!"#
!"
-,
/0
$%&$'()*+'*
0/
/
#
10/
0.
#$
/0
0/
/
#%
Scheme 5. Failed cyclization of unprotected N-(!-ketoacyl)hydroxytyrosinate
8’. Reagents and conditions: a) AgCO2CF3, MS 4 Å, NEt3, 23, THF, 0 °C, 1.5 h,
light exclusion; 31% crude. b) KOtBu, THF, RT, 5 min.
2<-=
>2),4
To
favor
Dieckmann
cyclization
over
retro-aldol
3.+,-
2<-=
<-=2
F0B'G0
40B'H0
fragmentation, we introduced an N-(2,4-dimethoxybenzyl) (DMB)
group, previously shown by Schlessinger et al. to render !-
ketoamide anions more nucleophilic.[17] Moreover, nitrogen
substituents larger than hydrogen are thought to promote the
Dieckmann cyclization by changing the amide conformation from
trans to cis, thus allowing for a sterically more favorable
nucleophilic attack at the ester carbonyl (Fig. 2). This
assumption was first uttered by Suzuki et al.[18] in their synthesis
!!
2<-=
>2),4
E0B'?0
&'
3.)
.2
>2),4
!&
3.!3?
<-=2
2.
!(
2
2.
(
7
2
"-@=
&!
,4
)
!!
(
3
A0B'C0
of the macrolactam macrocidin A, where they noticed
a
,4
,4
conformational change in X-ray structures of acyclic precursors.
2
2
2
2.
<-=2
7
*'(7
2
(
2<-=
2
!"&''()'*'+,-
!#&''()'*'.
!)&''()'*'!3?
!%&''()'*'.
&(
!(
!%
!"#"$"%
2.
/0
D0B'C0
2
,4
10
.
3
,4
,4
Figure 2. Predominant conformations of N-(!-ketoacyl)hydroxytyrosinates with
and without directing group (DG)
2
.2
56789):''';$0
2.
Scheme 6. Synthesis of protected F-14329 35, failed deprotection, and
synthesis of F-14329 (1). Reagents and conditions: a) TBSOTf, NEt3, CH2Cl2,
–10 °C $ 4 °C, 16 h; 82%. b) 10% Pd/C, H2, MeOH, RT, 15 h; 99%. c) 2,4-
dimethoxybenzaldehyde, MeOH, AcOH, RT, 30 min; then NaBH3CN, 3 h; 51%.
d) TBSCl, imidazole, CH2Cl2, RT, 19 h, 68%. e) TBSOTf, NEt3, CH2Cl2, –10 °C
$ 4 °C, 22 h; 82%. f) o-nitrobenzaldehyde, MgSO4, MeOH, AcOH, RT, 30
min; then NaBH3CN, 3 h; 76%. g) AgCO2CF3, MS 4 Å, NEt3, 23, THF, 0 °C,
1.5 h, light exclusion, 51% crude. h) 24, MS 3 Å, dioxane, reflux, 2.5 h, 57%
crude. i) NaOMe, MeOH, RT, 10 min; 50% (35) / 54% (37) over 2 steps. j)
10% TFA in CH2Cl2, RT. k) Pd(OH)2, NH4HCO2, MeOH, reflux. l) CAN, MeCN,
H2O, RT. m) DDQ, CHCl3, CH2Cl2, H2O, RT. n) h% 366 nm 4 W, MeCN, H2O,
RT, 1 d; 72%. o) TBAF, AcOH, THF, 0 °C $ RT, 38 h; 81%.
The introduction of an N-DMB group was possible only after
protection of the !-hydroxy group. O-silylation and debenzylation
of 29 gave aminoester 32 (Scheme 6). Its N-DMB protection and
subsequent N-!-ketoacylation with thioester 23 gave a !-
ketoamide (not shown) that decomposed like 8’ when treated
with NaOMe. Apparently, in the presence of a !-hydroxy or !-
silyloxy group, the phenolic OH-group of tyrosine-derived N-!-
ketoamides needs to be protected, too, prior to the Dieckmann
cyclization, in contrast to tyrosinates lacking the !-hydroxy
function.[16] Hence, phenol 32 was first N-protected with 2,4-
dimethoxybenzaldehyde / NaBH3CN and then O-silylated with
TBSCl to give the fully protected aminoester 33. Its Ley acylation
with !-ketothioester 23 and subsequent basic cyclization of the
crude intermediate !-ketamide 8 (PG1 = TBS, PG2 = DMB)
This article is protected by copyright. All rights reserved.

10.1002/chem.201701259
Chemistry - A European Journal
COMMUNICATION
[9]
X. Li, L. Li, R. Zhu, W. Li, W. Chang, L. Zhang, X. Wang, Z. Zhao, H.
The oNb group was removed by irradiating a solution of
compound 37 in acetonitrile–water with a 4 W lamp emitting light
of wavelength 366 nm for 1 d. The resulting bis silyl ether 38
was desilylated using TBAF in THF, with acetic acid as a buffer,
to yield F-14329 (1) in 81% after purification by MPLC, identical
to the natural product as to NMR spectra and specific optical
rotation.[21] A conveivable mechanism for the photolysis of the
oNb group, based on the studies of Il’ichev et al.[22] is shown in
Scheme 7. Light-induced tautomerization of 37 could generate
an activated intermediate 39 that undergoes cyclization with re-
aromatization to intermediate 40, which fragments to tetramic
acid 38 and o-nitrosobenzaldehyde.
Lou, J. Nat. Prod. 2015, 78, 2155-2160.
[10] Z. Shang, L. Li, B. P. Espósito, A. A. Salim, Z. G. Khalil, M. Quezada, P.
V. Bernhardt, R. J. Capon, Org. Biomol. Chem. 2015, 13, 7795-7802.
[11] N. A. Butt, C. J. Moody, Org. Lett. 2011, 13, 2224-2227.
[12] D. L. Boger, J. Zhou, R. M. Borzilleri, S. Nukui, S. L. Castle, J. Org.
Chem. 1997, 62, 2054-2069.
[13] for alternative, less effective methods see: D. Crich, A. Banerjee, J. Org.
Chem. 2006, 71, 7106-7109 and references therein.
[14] S. V. Ley, S. C. Smith, P. R. Woodward, Tetrahedron 1992, 48, 1145-
1174.
[15] N. Riache, C. Bailly, A. Deville, L. Dubost, B. Nay, Eur. J. Org. Chem.
2010, 5402-5408.
[16] L. Kong, M. Rao, J. Ou, J. Yin, W. Lu, M. Liu, X. Pang, S. Gao, Org.
Biomol. Chem. 2014, 12, 7591-7597.
[17] R. H. Schlessinger, G. R. Bebernitz, P. Lin, J. Am. Chem. Soc. 1985,
107, 1777-1778.
!
!'
!
!'
())*+,
[18] T. Yoshinari, K. Ohmori, M. G. Schrems, A. Pfaltz, K. Suzuki, Angew.
Chem. Int. Ed. 2010, 49, 881-885.
!
5
'
-./%0'1!2*&"
%
%
&
%
&
6
!
'
[19] Y. Tatsu, T. Nishigaki, A. Darszon, N. Yumoto, FEBS Lett. 2002, 525,
20-24.
!
"#$!
!
34
!"#$
!&
!"#$
!"
[20] A. Aldo, K. Aulinger-Fuchs, A. Gotschlich, B. Kramer, M. Lang, W.
Saeb, U. Sinks, A. Wuzik, U.S. Pat. Appl. Publ., US 20040235914 A1,
2004.
789:;*8+<=>.<
;?=;@,.48A?;8@+
B*!6+8;4@C@
D.+A?7<.:E<.
[21] T. Fukuda, Y. Sudoh, Y. Tsuchiya, T. Okuda, N. Matsuura, A. Motojima,
T. Oikawa, Y. Igarashi, J. Antibiot. 2015, 68, 399-402.
[22] Y. V. Il’ichev, M. A. Schwörer, J. Wirz, J. Am. Chem. Soc. 2004, 126,
4581-4595.
!
!
!'
&
!'
'!
5
!
%
%
'
%
%
&
6
!
!
!
!
34
34
!"#$
!#
!"#$
$%
Scheme 7. Conceivable mechanism of the photolytic deprotection of 37.
In summary, a synthetic route to !-hydroxytyrosine derived
3-acyltetramic acids was devised. It features the use of ortho-
nitrobenzyl as an N-protecting and directing group, readily and
selectively removable at a late stage without compromising the
delicate !-hydroxy group and the configurational integrity of
stereocenters. The fungal metabolite F-14329 (1) was prepared
by this approach in 14 steps and 3.9% overall yield.
Keywords: natural products • F-14329 • chaunolidines • tetramic acids • total
synthesis
[1]
[2]
X. Mo, Q. Li, J. Ju, RSC Adv. 2014, 4, 50566-50593.
a) R. Schobert, A. Schlenk, Bioorg. Med. Chem. 2008, 16, 4203-4221;
b) M. Petermichl, R. Schobert, Synlett 2017, DOI:10.1055/s-0036-
1588402.
[3]
a) D. Boettger, C. Hertweck, ChemBioChem. 2013, 14, 28-42; b) F.
Hemmerling, F. Hahn, Beilstein J. Org. Chem. 2016, 12, 1512-1550; c)
C. Gui, Q. Li, X. Mo, X. Qin, J. Ma, J. Ju, Org. Lett. 2015, 17, 628-631.
K. Schmidt, U. Riese, Z. Li, M. Hamburger, J. Nat. Prod. 2003, 66, 378-
383.
[4]
[5]
[6]
[7]
T. Nakada, M. Nakajima, H. Kobayashi, M. Takahashi, I. Tanaka, Jpn.
Kokai Tokkyo Koho, JP 2007153840, A 20070621, 2007.
H. V. K. Wangun, C. Hertweck, Org. Biomol. Chem., 2007, 5, 1702-
1705.
L. M. Halo, M. N. Heneghan, A. A. Yakasai, S. Zhongshu, K. Williams,
A. M. Bailey, R. J. Cox, C. M. Lazarus, T. J. Simpson, J. Am. Chem.
Soc., 2008, 130, 17988-17996.
[8]
Z. Wasil, K. A. K. Pahirulzaman, C. Butts, T. J. Simpson, C. M. lazarus,
R. J. Cox, Chem. Sci. 2013, 4, 3845-3856.
This article is protected by copyright. All rights reserved.

10.1002/chem.201701259
Chemistry - A European Journal
COMMUNICATION
COMMUNICATION
S. Bruckner,† R. G. Haase,† R. Schobert*
Page No. – Page No.
A Synthetic Route to !-
Hydroxytyrosine Derived Tetramic
Acids: Total Synthesis of the Fungal
Metabolite F-14329
OH makes all the difference: The fungal metabolite F-14329 was synthesized as
the first member of natural !-hydroxytyrosine derived tetramic acids. The proneness
of the !-hydroxy group to elimination and retro aldol fragmentation necessitated
reagents rarely used in tetramate chemistry, such as a Meldrum’s acid derivative for
the introduction of the 3-acyl side chain and a photocleavable ortho-nitrobenzyl
protecting and directing group on the amine.
This article is protected by copyright. All rights reserved.