Organic Chemistry of Natural Products

菅 裕明 (Hiroaki Suga, Bioorganic Chemistry Lab.)

Robert Campbell（Biomolecular Chemistry Lab.）
佐竹 真幸 (Masayuki Satake, Natural Product Chemistry Lab.)
後藤 佑樹 (Yuki Goto, Bioorganic Chemistry Lab.)
加藤 敬行 (Takayuki Katoh, Bioorganic Chemistry Lab.)
Tentative lecture schedule:
date No. instructor subject
9/26 1 Goto overview of natural products
10/3 2 Goto terpenes
10/10 3 Goto terpenes
10/17 4 Katoh amino acids and peptides
10/24 5 Katoh amino acids and peptides
10/31 6 Satake polyketides
11/14 7 Satake polyketides
11/21 8 Satake polyketides
11/28 9 Satake chikimic acid pathway and its products
12/5 10 Satake chikimic acid pathway and its products
12/12 11 Campbell topic: protein engineering for diagnostics and therapeutics
12/19 12 Campbell topic: protein engineering for diagnostics and therapeutics
12/26 13 Suga topic: discovery and application of pseudo-natural peptides
1/9 14 Suga topic: quorum sensing controlled by auto inducers
12/8 11 Satake TBA
12/15 12 Satake TBA
12/22 13 Satake TBA
1/5 14 Satake TBA

Grading:
Final exam + attendance

Suggested reference textbooks:

医薬品天然物化学（南江堂）、海老塚豊 監訳
マクマリー生化学反応機構（東京化学同人）、長野哲雄 監訳
Medicinal Natural Products: A Biosynthetic Approach, Paul M. Dewick
The Organic Chemistry Of Biological Pathways, John E. McMurry & Tadhg P. Begley
 What is natural product?
What is natural product?
It is a chemical compound produced by a living organism—that is, found
in nature

In a narrow sense, natural products often means secondary metabolites that

have intriguing bioactivities.
essential and universal
What are secondary metabolites? ( primary metabolites)
They are organic compounds generated in organisms (metabolites) that are
not essential in the reproduction, development, or normal growth.

They often provide evolutionary advantages for the producing organisms.

e.g.) "chemical weapon" agents against prey, predators, and competing organisms
spider venom frog toxins anti-bacterials
 What is natural product?
What is natural product?
It is a chemical compound produced by a living organism—that is, found
in nature

In a narrow sense, natural products often means secondary metabolites that

have intriguing bioactivities.
Humans use secondary metabolites discovered from nature
as medicines, agents, and flavorings.
e.g.)

Paclitaxel Artemisinin Avermectin

anti-cancer drug anti-malarial drug anti-parasitic worm drug
isolated from a plant isolated from a plant produced by a bacterium
(Taxus brevifolia) (Artemisia annua) (Streptomyces avermitilis)

Physiology/Medicine, 2015 Physiology/Medicine, 2015

 What is natural products chemistry?

What is natural products chemistry?

A field of organic chemistry that deals with “natural products”.

It is related to various chemistry fields such as medicinal chemistry,

synthetic chemistry, and agricultural chemistry.

Japanese scientists have historically shown large presence.

Typical flow in natural products chemistry

Discovery of organisms that show interesting bioactivities

Isolation of objective bioactive compounds from the organisms

Structural determination

Elucidation of the mode of action

Total synthesis

Derivatization of natural products aiming at development of novel drugs

 Typical flow in natural products chemistry
Discovery of organisms that show interesting bioactivities

Willow bark can be used for Morays sometimes cause

A fungus kills microbials
suppression of toothache food poisoning

Isolation of objective bioactive compounds from the organisms

Structural determination
Elucidation of the mode of action
Total synthesis

Derivatization of natural products aiming at development of novel drugs

 Typical flow in natural products chemistry
Discovery of organisms that show interesting bioactivities

Isolation of objective bioactive compounds from the organisms

meat of morays
culture medium of fungi willow bark (e.g. 850 morays, 4 tons!)

(0.3 mg of final compound)

Purifying the natural samples by means of various separation methods,
and hunting the fractions that shows the objective bioactivity

extraction liq.-liq. separation chromatography crystallization

Structural determination
Elucidation of the mode of action
Total synthesis

Derivatization of natural products aiming at development of novel drugs

 Typical flow in natural products chemistry
Discovery of organisms that show interesting bioactivities

Isolation of objective bioactive compounds from the organisms

Structural determination

Determination of the structure of the bioactive compound

by means of elementary analysis, crystal structure analysis, and
comprehensive spectrum analyses (e.g., MS, MS-MS, IR, NMR, etc.)

H
N S OH
O N O
O
OH OH
O

benzylpenicillin salicylic acid ciguatoxin

Total synthesis Elucidation of the mode of action

Derivatization of natural products aiming at development of novel drugs

Typical flow in natural products chemistry
Discovery of organisms that show interesting bioactivities

Isolation of objective bioactive compounds from the organisms

Structural determination

Total synthesis (de novo chemical synthesis of natural products)

available reagents

O HOOC
O BnO
I O
tBuO2C O
BnO OH BnO
O
O OPMB O O
SPh
OBn MP MOMO
O
SPh

Science, 294, 5548 (2001)

Elucidation of the mode of action

Derivatization of natural products aiming at development of novel drugs

Typical flow in natural products chemistry
Discovery of organisms that show interesting bioactivities

Isolation of objective bioactive compounds from the organisms

Structural determination

Total synthesis (de novo chemical synthesis of natural products)

Purpose/motivation: Confirmation of proposed structures

Mass supply of useful natural products

Derivatization of natural products

Artistic synthesis of sophisticated structures

Elucidation of the mode of action

Derivatization of natural products aiming at development of novel drugs

 Typical flow in natural products chemistry
Discovery of organisms that show interesting bioactivities

Isolation of objective bioactive compounds from the organisms

Structural determination

Total synthesis

Elucidation of the mode of action

transpeptidase

Penicillin

Derivatization of natural products aiming at development of novel drugs

Typical flow in natural products chemistry
Discovery of organisms that show interesting bioactivities

Isolation of objective bioactive compounds from the organisms

Structural determination
Elucidation of the mode of action
Total synthesis

Derivatization of natural products aiming at development of novel drugs

H
N S
O N
O
OH
O
O
benzylpenicillin OH
(penicillin G) O
O

O
OH
OH
salicylic acid acetylsalicylic acid
NH2 O
H
N S H (aspirin)
N S
O N
O O O N
OH O
O OH
O
Ampicillin methicillin
Typical flow in natural products chemistry

Discovery of organisms that show interesting bioactivities

Isolation of objective bioactive compounds from the organisms

Structural determination

Elucidation of the mode of action

Total synthesis

Derivatization of natural products aiming at development of novel drugs

 Communications to the Editor J . Am. Chem. SOC..Vol. 107, No. 16, 1985 4797
An example of drug development based on natural products
1985: Uemura et al.

homogenization, extraction,
liq.-liq. separation, chromatography
Figure 1. Computer-generated perspective drawing of the final X-ray model of the p-bromophenacyl ester 2.

and Tsukitani resulted in the identification of okadaic acid2 as

a cytotoxic constituent of this animal. However, our interestDiscovery in and identification of novel compounds
3162
the same animal focused on the fact that sponge J . extracts
Am. Chem. SOC.1992, 114,
exhibited ' 3162-3164
O
"/ T :0
remarkable in vivo antitumor a c t i ~ i t y .Bioassay
~ against B- 16 with potent cytotoxicity
radical
melanoma5 which,
cells after
guided recombination
the isolation with the flavin semiquinone,
of halichondrins in low yield, features, coupled with this fact, encouraged synthetic efforts
600leads
kg of
to sea
covalent sponge
modification (Halichondria
of FAD (Scheme I, okadai)
route
including norhalichondrin A (1):4 amorphous solid [ ( 5 X lo")% B). Since toward this（named halichondrins）
class of natural product^.^,^ In this paper, we report
the reactions
yield fromfromof carbon
wet Miura
animals];radicals, particularly
Peninsula
[aID -47.8' (MeOH,cyclopropylcarbinyl
c 1.13). the first total synthesis
OR Pure BAppl.
of halichondrin andOR Chem. 58 (5):B,701–710
norhalichondrin
and homoallylic species, with molecular oxygen are well docu- which has, we believe, potential to meet the demand.
mented,14 the inactivation derailment may be envisioned as
trapping the acyclic radical 5 with O2to form a transient peroxy
radical 6 which, upon reduction by one-electron transfer from the
active-site-bound flavin semiquinone, gives rise to a peroxy anion
7. Since the partition ratio of this inactivation is approximately R= COC6H4Br

3, reacting with oxygen instead of coupling with the flavin co-

Figure 2. Exciton chirality of the 12,13-bis(p-bromobenzoate)system in
enzyme is clearly a more facile process for the ring-opened radical 3.
intermediate 5. As depicted in Scheme 11, this reroute is cul-
minated by an intramolecular epoxidationls converting 7 via a
1,2-dioxolanylcarbiny1 anion 8 to the observed turnover product and 0.1102 for the enantiomer. The computer-generated per-
9.16 The mechanistic insights derived from this study provide spective drawings8 of a molecule of 2 is shown in Figure 1, in-
highly convincing evidence sustaining our early notion that in- cluding its absolute configuration.
norhalichondrin A
activation of GAD by MCPA-CoA is likely to proceed through halichondrin B
Application of the nonempirical dibenzoate chirality methodg
a radical mechanism. These results may also be extrapolated to was also consistent with the X-ray crystallographic result.
suggest isolation
that GAD is capable yield: 35 mg
of mediating one-electron oxida-
The SIMS spectrum showed the highest mass peak at m/z 1127
cytotoxicity
Treatment against B-16
of 1 with p-bromobenzoyl melanoma
chloride cells
in pyridine at 80
+
tion-reduction.
( M 0.00005%
1) correspondingof tothe original
the final molecular formula CS9H82021. "C yielded product
sample! IC 3 with the y-lactone [IR (CHC13) 1775 cm-'1;
50 = 0.093 ng/mL (= 80 pM)
3 contains two p-bromobenzoyl groups judging from the IH N M R
The IR spectrum (KBr)
Acknowledgment. Weindicated
thank Dr.theVikram
presence of hydroxyls
Roongta for his(3450
as-
cm-I), lactone spectrum [C&, 6 6.9-7.8(8 H, m), 5.89 (1 H, s, H13)]. This
sistance with N larger than five-membered
M R measurements. This ring
workorwas estersupported
(1740cm-'),by reaction seems to proceed by reasonable C13-Cl2 acyl
3 : homohallchondrin B migra-
a and carboxylate
National ofMode
(1580
Institutes cm-I).
Health Theof lH
grantaction:
and 13C
(GM NInhibition
40541). M RH.-w.L.
spectra alsoof celltion,1°
were division via targeting tubulin
followed by ordinary acylation of the secondary alcohol.
not sothe
thanks fruitful for structural
National Institutes elucidation
of Health for because of their
a Research com-
Career J. Biol. Chem. 266 system
(24): 15882–9
The bis(p-bromobenzoate) 3 possessing the 1,2-dibenzoate
 3162 J . Am. Chem. SOC.1992, 114, 3162-3164
An example
radical of drug
5 which, after recombination development
with the flavin semiquinone, based
features, coupled on natural
with this products
fact, encouraged synthetic efforts
leads to covalent modification of FAD (Scheme I, route B). Since toward this class of natural product^.^,^ In this paper, we report
the reactions of carbon radicals, particularly cyclopropylcarbinyl the first total synthesis of halichondrin B and norhalichondrin B,
and homoallylic species, with molecular oxygen are well docu- which has, we believe, potential to meet the demand.
1992: Kishi et al.
mented,14 the inactivation derailment may be envisioned as
trapping the acyclic radical 5 with O2to form a transient peroxy
Total
radical 6synthesis of halichondrin
which, upon reduction B from the
by one-electron transfer
active-site-bound flavin semiquinone, gives rise to a peroxy anion
Since the partition ratio
7.commercially overof160this inactivation is approximately
steps chemical reactions
3, reacting with
available oxygen instead of coupling with the flavin co-
enzyme is clearly a more facile process for the ring-opened radical
reagents
intermediate 5. As depicted in Scheme 11, this reroute is cul-

Additional Reading:
minated by an intramolecular epoxidationls converting 7 via a
1,2-dioxolanylcarbiny1 anion 8 to the observed turnover product
9.16 The mechanistic insights derived from this study provide halichondrin B
highly convincing evidence sustaining our early notion that in-
2004: Eisai Halichondrin B and E7389
J. Am. Chem. Soc. 114 (8): 3162–3164.
Co., Ltd.
activation of GAD by MCPA-CoA is likely to proceed through
a radical mechanism. These results may also be extrapolated to
suggest that GAD is capable of mediating one-electron oxida-
Simplification of halichondorins - development of E7389
tion-reduction.
Acknowledgment. We thank Dr. Vikram Roongta for his as-
sistance with N M R measurements. This work was supported by 3 : homohallchondrin B
a National Institutes of Health grant (GM 40541). H.-w.L. also
thanks the National Institutes of Health for a Research Career
Development Award (GM 00559).
Scheme I outlines the synthesis of the right half of the hali-
(14) (a) Nakamura, E.; Inubushi, T.; Aoki, S.;Machii, D. J . Am. Chem.
SOC.1991, 113, 8980. (b) Beckwith, A. L. J. Tetrahedron 1981, 37, 3073. chondrin Bs. We planned to form the C-21-C-22 bondS via a
(c) Porter, N. A. In Free Radicals in Biology; Pryor, W. A., Ed.; Academic Horner-Emmons reaction, followed by conjugate reduction. We
Press: New York, 1980; Vol. IV, p 261. (d) Beckwith, A. L. J.; Ingold, K. were concerned with double-bond isomerization from the C- 19
U. In Rearrangements in Ground and Excited States; de Mayo, P., Ed.; exocyclic to the C- 19-C-20 endocyclic position in this process.
Academic Press: New York, 1980. This transformation was accomplished via the preparation of the
( I 5) (a)•Halichondrin B is a
Clark, C.; Hermans, P.; Meth-Cohn, highly
0.;Moore, cytotoxic
C.; Taljaard, H. (antimitotic) marine
2/3 molecular natural
weight and product.
chiral centers
C.; van Vuuren, G. J . Chem. Soc., Chem. Commun. 1986,1378. (b) Rontani, aldehyde from the primary alcohol 36.7by Des-Martin oxidation:
Total synthesis: Kishi, 1992
J.-F. Tetrahedron
SOC.1991, 113, 9403.
Lett. 1991,6551. (c) Dowd, P.; Ham, S. W. J. Am. Chem. Horner-Emmonsimproved
reaction under
in carefully controlled conditions,
vivo stability
and the conjugate reduction of the resulting enone by the Stryker
(16) An• alternate
Recent diverted
mechanism total
is also conceivablesynthesis has ledreagent:
involving an intramo- to simplified analogisomerization.
E7389 with
without double-bond Hydride reduction
lecular cyclization of 6 to produce a 1,2-dioxolanylcarbinyIradical (Feldman, FDA approval in 2010!
of the resulting saturated with
ketoneketone
yielded approximately a 1:l
K. S.;Simpson, similar
R. E. J .antimitotic activity.
Am. Chem. SOC.1989, 111, 4878)Also,
and thenreplacement
a of lactone
Drug development using artificial compounds

Drug development based on natural products chemistry

Discover the “seeds” of objective drugs from nature,
and brush them up to develop desirable drugs

Drug development using artificial compounds

de novo development of drug seeds
       using synthetic chemical approaches
 Drug development using artificial compounds
High-throughput screening (HTS) of chemical libraries
Chemical library HTS

collection of
various compounds
diversity = 105∼106

Structure-activity relationship
(SAR) study
drug candidates “hit” compounds
(lead compounds) (drug seeds)
derivatization
 Drug development using artificial compounds
High-throughput screening (HTS) ofR chemical
EVIEWS
libraries
Chemical library
compounds linking to be achieved in an efficient manner.
after SAR study
Figure 3 | Schematic representation of a low-quality
image of thescreening
HTS hit. The high-throughput interaction
(HTS) hit is large andof
makes surface contact with the receptor without forming
the final
high-quality interactionsdrug candidates
in key pockets. The affinity is spread
throughout the entire molecule and, in the absence of
collection of the medicinal chemist does not know
structural information,
variousExperience
compounds
which areas of the molecule to focus on during hit optimization.
shows that optimization of these kinds of hits is
very difficult. This is in contrast to the schematic in , in
1051∼10
FIG. 4
diversity
which=fragment 6
is much smaller, makes high-quality contacts
with the receptor and has relatively weak affinity. It has been
shown that such fragments can often be built up into attractive
leads with the aid of structural information (for example, TABLE 2)
Structure-activity relationship
Table 2, entry 7. NMR screening (that is, structure–activity
relationships (SAR) by NMR7) was used to identify a (SAR) study
drug candidates series of triazine-containing compounds that bind in the
mM range to ErmAM methyl transferase24 (for example,
(lead compounds) TABLE 2, entry 7, NMR Kd = 1 mM), an enzyme target for
ameliorating antibiotic resistance. Optimization of this
initial lead using parallel synthesis led to inhibitors in the
low µΜ range (for example, table entry Ki = 7.5 µΜ).
NMR and X-ray structures show that these non-nucleo-
side compounds bind to the S-adenosylmethionine-
binding site on the Erm protein and that there is scope for
the incorporation of additional binding interactions.
Table 1 | Comparison of fragment-based approaches and high-throughput screening
Fragment-based approaches
R
Lead identification by fragment linking
HTS Lead identification by fragment link
TABLES 3,4 show examples in which two fragments have TABLES 3,4 show examples in which two
been identified that bind in separate binding sites that been identified that bind in separate bi
are close enough to each other to be chemically linked are close enough to each other to be ch
(FIG. 5). For this to be an efficient lead-identification (FIG. 5). For this to be an efficient lead
approach, one needs to both identify the initial frag- approach, one needs to both identify t
ments and also have a process that allows the appropriate ments and also have a process that allows
initial hit
linking to be achieved in an efficient man
The potency increase achievable from optimally The potency increase achievable f
linking two fragments is often assumed to benefit from compound linking two fragments is often assumed
an approximate additivity effect, such that the free energy an approximate additivity effect, such tha
of binding of the joined molecule is approximately equal
Figure 3 | Schematic representation of a low-quality
of binding of the joined molecule is appr
to the sum of the free energies of binding ofHTS image of the interaction of
the hit.
frag-The high-throughput screening (HTS) hit is large and to the sum of the free energies of bind
ments (that is, two millimolar fragments when makes joined
surface contact with the receptor without forming ments (that is, two millimolar fragmen
the initial hit compounds
together lead to a micromolar inhibitor)5. Suchhigh-quality
additivity interactions in key pockets. The affinity is spread together lead to a micromolar inhibitor)5
requires that the contribution from the linker isthroughout
negligiblethe entire molecule and, in the absence of requires that the contribution from the lin
and that the loss in rigid-body entropy on binding structural
of allinformation, the medicinal chemist does not know and that the loss in rigid-body entropy o
which areas of the molecule to focus on during hit optimization.
components to the enzyme is very small. Recently, an
Experience shows that optimization of these kinds of hits is
components to the enzyme is very sma
analysis of the experimental energetics associated with This is in contrast to the schematic in FIG. 4, in
very difficult. analysis of the experimental energetics
optimally linked fragments has suggested thatwhich the rigid-
fragment 1 is much smaller, makes high-quality contacts optimally linked fragments has suggeste
body entropy loss on protein binding constitutes witha barrier
the receptor and has relatively weak affinity. It has been body entropy loss on protein binding con
of around three orders of magnitude to theshown bindingthat such fragments can often be built up into attractive of around three orders of magnitude
leads withof
affinity, and that this barrier is essentially independent the aid of structural information (for example, TABLE 2) affinity, and that this barrier is essentially
molecular mass25. The analysis implies that there should molecular mass25. The analysis implies th
be a super-additivity effect when two fragments are be a super-additivity effect when two
linked in an optimal fashion. Such super-additivityTable 2, entryis 7. NMR screening (that is, structure–activity linked in an optimal fashion. Such sup
observed for entries 2 and 7 in TABLE 3. relationships (SAR) by NMR7) was used to identify a observed for entries 2 and 7 in TABLE 3.
“hit” compounds
series of triazine-containing compounds that bind in the
Table 3, entry 1. SAR by NMR7 was used to mM identify
range Table 3, entry 1. SAR by NMR was us
a to ErmAM methyl transferase24 (for example, 7
potent 49 nM inhibitor of the FK506-binding TABLE 2, entry 7, NMR Kd = 1 mM), an enzyme target for
protein (drug
potent 49 nMseeds)
inhibitor of the FK506-b
(FKBP) binding domain by linking twoameliorating weaker antibiotic resistance. Optimization of this
(FKBP) binding domain by linkin
inhibitors (2 µM and 100 µM). NMR screening initial
of alead derivatization inhibitors (2 µM and 100 µM). NMR sc
set using parallel synthesis led to inhibitors in the
of 1,000 fragments — including pipecolinic acid µΜ range (for example, table entry Ki = 7.5 µΜ).
low deriv- of 1,000 fragments — including pipecol
atives, a class of compounds known to bind toNMR FKBPand — X-ray structures show that these non-nucleo- atives, a class of compounds known to b
identified the pipecolinic acid (Kd = 2 µM) sideandcompounds
the bind to the S-adenosylmethionine- identified the pipecolinic acid (Kd =
diphenyl amide (Kd = 100 µM). Use of 15N-13binding C-filtered site on the Erm protein and that there is scope for diphenyl amide (Kd = 100 µM). Use of 1
protein–ligand NUCLEAR OVERHAUSER EFFECT (NOE) data
the incorporation of additional binding interactions. protein–ligand NUCLEAR OVERHAUSER EFFE
Table 1 | Comparison of fragment-based approaches and high-throughput screeni
High-throughput screening (HTS) Fragment-based approaches High-throughput screening (HTS)
 Drug development using artificial compounds
Fragment-based drug discovery (FBDD) REVI

small compounds monitoring theprotease

caspase-3, a cysteine interaction
34 “extended”
. Mass spectrometry was these library fragments were coupled to known re
(fragments)
REVIEWS REVIEWS
by NMR
used to and/or
identify thecrystallography fragment
library members that had formed a library
cysteine-binding elements to generate potent re
disulphide bond to the tethered thiol. Subsequently, molecules.
Lead identification by fragment linking
DNA gyrase TABLES 3,4 show examples in which two fragments have

caspase-3, a cysteine protease34. Mass spectrometry was these library fragments were coupled to known reversible
been identified that bind in separate binding sites that
Table 2 | Lead identification by fragment evolution* are close enough to each other to be chemically linked
used to identify the library members that had formed a this tocysteine-binding
(FIG. 5). For elements to generate potent reversible
be an efficient lead-identification
approach, one needs to both identify the initial frag-
Entry Target/method Fragment Evolved
disulphide bond to the tethered thiol. Subsequently, fragment
ments and also have a process that allows the appropriate Lead
molecules.
1 DNA gyrase1/ H
linking to be achieved in an efficient manner.
H
The potency increase achievable from optimally H
VS and SBD N linking twoN fragments is often assumed to benefit from N
Table 2 | Lead identification by fragment evolution* N N N
REVIEWS
an approximate additivity effect, such that the free energy
Figure 3 | Schematic representation of a low-quality
of binding of the joined molecule is O
approximately equal
Entry Target/method Fragment HTS hit. The high-throughput screening Evolved
(HTS) hit is largefragment
and to the sum of the free energiesOof binding Leadof the frag- O

1
diversity = 10 ∼10 2
DNA gyrase1/
4 makes surface contact with the receptor without forming ments (that is,Stwo millimolar fragments when
together lead to a micromolar inhibitor)5. Such additivity
joined O S
Hhigh-quality interactions in key pockets. The affinity is spreadH H
VS and SBD Nthroughout the entire molecule and, in the absence of
structuralK = 10themM (bychemist
NMR) N requires that the contribution from the linker is negligible
MNEC = 8 µg per ml 34MNECN= 30 ng per ml
4
. Mass spectrometry was these library fragments were coupled to known reversible caspase-3, a cysteine protease . Mass spectrometry was
N areas
which
information,
d
MNEC >250 µg per ml
of the molecule
medicinal
to focus on
does not know
during hit N
optimization.
and that the loss in rigid-body entropy on binding of all
components to the enzyme is very small. Recently, an
N these lib
mbers that had formed a cysteine-binding elements to generate potent reversible used to identify the library members that had formed a
Experience shows that optimization of these kinds of hits is
analysis of
O
Othe experimental energetics associated with O
O cysteine
2 Thymidylate very difficult. This is in contrast to the schematic in FIG. 4, in CO2H
red thiol. Subsequently, synthase /19 disulphide bond to the tethered thiol. Subsequently,
which fragment 1 is much smaller, makes high-quality contacts
molecules. S
optimally linked fragments has suggested
body entropy loss on protein binding constitutes
that the rigid-
O S molecu
with the receptor and has relatively weak affinity. It has been CO2Ha barrier
tethering shown that such fragments can often be built up into attractive
leads with the aidO
redesigned
of around three orders of magnitude to the binding O NH H
and SBD K = 10 mM (by NMR) MNEC = 8
of structural information (for example, TABLE 2)
µg per ml
affinity, and that this barrier is essentially MNEC =
independent 30 ofng per ml
O S
n*
d
MNEC >250 µgO per S Table
ml 2 | Lead identification O
molecular massby 25
. Thechemical library
fragment
NH implies
analysis evolution*
H thatRthere
E V should
IEWS
O
CO2H
be a super-additivity effect when two fragments are Figure 3 | Schematic representation of a low-quality
O S CO2H N
2 drug candidates
Thymidylate
Evolved fragment 19
Table 2, entry 7.N
Entry Target/method
NMR screening (that is, structure–activity
Lead relationships (SAR) by NMR ) was used to identify a observed
7
Fragment
linked in an optimal fashion. Such super-additivity
for entries 2 andO TABLE 3.
7 infragment
is
Evolved O contact
makes surface
fragment
CO 2H
HTS hit. The high-throughput screening (HTS) hit is large and
with the receptor without forming
synthase / O N
Lead identification by linking
bind in the1/
(lead compounds)
H tethering mM range to ErmAM
1
series of triazine-containing
H methyl
compoundsDNA
transferase
thatgyrase
24
(for example,
TABLES 3,4 show examples
Table 3, entry O 1.
CO2H in which two fragments have
SAR by NMR H 7 was usedOto identify a
high-quality interactions in key pockets. The affinity is spread
throughout the
NH HH entire molecule and, in the absence of
NHinformation,
N
O OH VS and
N Kd = 1 mM), an enzyme target for SBD been identified that bind in N separate binding sites that structural N the medicinal chemist does not know
and SBD O S
TABLE 2, entry 7, NMR
O
potent
NH
49 nM
are close inhibitor
enough
H
of the
to each FK506-binding
other
N toObe chemically
S protein
linked which areasCO HN
of2the molecule to focus on during hit optimization.
N N (FIG. 5)OH
ameliorating antibiotic resistance. Optimization of this (FKBP) binding domain by linking two
. For this to be an efficient lead-identification weaker Experience shows that optimization of these kinds of hits is
initial lead using parallel synthesis ledOto inhibitors in the inhibitors (2COµM and 100 µM). NMR screening of a set O
very difficult. This is in contrast to the schematic inO
O S O approach, one 2Hneeds to both identifyNthe initial frag- O FIG. 4, in
O N low µΜ range (for example, table entry O Ki = 7.5 µΜ). of ments
1,000 fragments — including pipecolinic acid deriv- which fragment CO 2H smaller, makes high-quality contacts
1 is much
and also have a process that
O a class of compounds known to bind to FKBP allows the appropriate
O—
NMRO and X-ray structures Sshow that theseNnon-nucleo- IC atives, with the receptor and has S relatively weak affinity. It has been
S IC O
50
= 1.1 mM
side compounds bind to the S-adenosylmethionine- 50
= 24
linking
identified
toµM be achieved in an efficient manner.
the pipecolinic acid (Kd = 2 µM) and the
IC50shown
= 330 nMfragments can often be built up into attractive
that such
O The potency increase achievable from optimally leads with the aid of structural information (for example, TABLE 2)
binding site on the Erm protein and that there is scope for diphenyl
linkingamide K(Kdd ==100
10 mM
isµM). Use (by NMR)
of 15N- 13
NH from
to C-filtered MNEC = 8 µg per ml
MNEC = 8 µg per3 ml p38 kinase2MNEC
/ =OH 30 ng per ml
the incorporation of additional binding interactions. protein–ligand
two fragments often assumed
NUCLEAR OVERHAUSER EFFECT (NOE) data
benefit N
NMR OH
an approximate MNECadditivity>250 µg
effect, such thatper ml
the free
S joined molecule is approximately equal
energy
Figure 3 | Schematic representation of a low-quality
of binding of the CH
Table 2, entry 7. NMR 3
screening (that is, structure–activity
Table 1 | Comparison2
HTS hit. The high-throughput of Thymidylate
fragment-based
screening
CO2H (HTS) hit is approaches
large and toand
the sumhigh-throughput
of the free energies screening CO2H of the frag-
of binding
CO2H relationships (SAR) by NMR7) was used to identify a
makes surface contact
Fragment-based with the receptor without forming
approaches 19 ments (that is,
High-throughput Ntwo millimolar
screening (HTS)fragments when joined N
series of triazine-containing compounds that bind in the
 Natural products vs artificial compounds

Pros and cons of drug development approaches based on natural products

approach based on natural products approach using artificial compounds

Strong bioactivity obtained Limitation in diversity of

during the process of evolution available chemical libraries

Initial hit compounds are Optimizations of hit compounds

  
often drug-ready molecules are generally required

Highly rely on the compounds Rational design strategies of

produced in nature novel compounds are possible

Supply by chemical synthesis The drugs can be readily

  
is sometimes challenging. supplied by chemically synthesis
of
a
)
e.
Large presence of natural products in drug development
y,
a-
Figure 1. Source of small molecule drugs, 1981–2006: major
an categories, N ) 983 (in percentages). Codes are as in ref 1. Major
he categories are as follows: “N”, natural product; “ND”, derived from a
natural product and usually a semisynthetic modification; “S”, totally
es synthetic drug often found by random screening/modification of an
e, existing agent; “S*”, made by total synthesis, but the pharmacophore
ed natural products
is/was from a natural product. natural products
a- mimics
an
natural products
derivatives
1),
al

ve 58% of the approved drugs

re are natural products-related.
m
on
in sources of small molecule drugs, 1981–2006 (N = 983)
be Figure 2. Sources of small molecule drugs, 1981–2006: all categories,
ic J. Med. Chem., 51, 2589–2599 (2008)
N ) 983 (in percentages). Codes are as in ref 1. Major categories are
ts as follows: “N”, natural product; “ND”, derived from a natural product
0s and usually a semisynthetic modification; “S”, totally synthetic drug
ti- often found by random screening/modification of an existing agent;
of “S*”, made by total synthesis, but the pharmacophore is/was from a
natural product. The subcategory is as follows: “NM”, natural product
 Power of natural products –1
五大抗生物質
Antibiotics 五大抗生物質

Antibiotics are compounds that kill bacteria or suppress bacteria growth

Many natural products exhibit antibiotic activities.

ペニシリン
ストレプトマイシン
五大抗生物質 ペニシリン
ストレプトマイシン

クロラムフェニコール
ペニシリン
penicillin chloramphenicol
クロラムフェニコール tetracycline
ストレプトマイシン テトラサイクリン
テトラサイクリン
抗生物質には、語尾に「-mycin」（マイシン）がつく エリスロ
抗生物質には、語尾に「-mycin」（マイシン）がつく
ものが多い エリスロマイシン
ものが多い
O
H
クロラムフェニコール N S
テトラサイクリン
O O N
O
抗生物質には、語尾に「-mycin」（マイシン）がつく エリスロマイシン OH
O
ものが多い
vancomycin methicillin
 Power of natural products –1
Development of antibiotics changed the world.
changes in causes of death http://www2.ttcn.ne.jp/honkawa/2080.html

pneumonia

gastroenteritis cancers

tuberculosis
heart diseases

cerebrovascular
disease

benzylpenicillin release in 1942 emergence of MRSA in 1980s

 Power of natural products –2
Immunosuppressive drugs

Immunosuppressive drugs are compounds that prevent the immune system.

They drastically increased success rate of organ transplantation.

Number of transplantation in US
kidney
cyclosporin heart
liver
discovered by Sandoz (now Novartis)
approved in 1983

tacrolimus
discovered by Fujisawa (now Astellas)
approved in 1994

http://www.medi-net.or.jp/tcnet/tc_1/1_2.html
 Power of natural products –3
Statins

They reduce cholesterol in blood and are used for hyperlipidemia treatment.

This class of drugs resulted in many blockbusters.

Mevastatin Mevalotin
natural product natural product derivative
Daiichi-Sankyo/BMS
4,746 million $ sales in 2003
(約5000億円)
 Major categories of natural products

alkaloids (highly modified amino acids) steroids (terpenes with a specific ring system)

cocaine caffeine cholesterol testosterone

terpenoids (oligomerized isoprenes) glycoside (highly modified sugars)

menthol retinol salicin kanamycin

 Major categories of natural products

polyphenol polyketide (oligomerized malonic acid)

catechin sesamin erythromycin tetracycline

peptides (amino acid oligomers)

cyclosporin vancomycin