New strategy in synthetic biology: from enzyme inhibition and natural products synthesis to PET...

www.tcr.wiley-vch.de 119

New Strategy in Synthetic Biology: from Enzyme Inhibition and Natural Products Synthesis to PET Imaging by 6π-Azaelectrocyclization

KATSUNORI TANAKA,*1 KOICHI FUKASE,1 AND SHIGEO KATSUMURA2

1Department of Chemistry, Graduate School of Science, Osaka University, 1-1 Machikaneyama, Toyonaka, Osaka, 560-0043, Japan2Department of Chemistry and Open Research Center on Organic Tool Molecules, School of Science and Technology, Kwansei Gakuin University, Gakuen 2-1, Sanda, Hyogo 669-1337, Japan

Received 13 October 2009In memory of Carlo Rosini

ABSTRACT: While elucidating the inhibitory mechanism of a hydrolytic enzyme by aldehyde-containing natural products, we discovered a reaction involving a rapid 6π-azaelectrocyclization of azatrienes generated from aldehyde with lysine residues. The electrocyclic reaction of the 1-azatriene system, a cyclization precursor, exhibited a substituent effect. Structure-reactivity studies showed that azaelectrocyclization, which usually proceeds in low yield at high temperatures, produced a quantita-tive yield in less than 5 min at room temperature. Asymmetric chiral piperidine synthesis and a one-pot library synthesis of pyridines on solid-supports were applied to synthesize pyridine/indole alkaloid-type natural products. Additionally, we developed lysine-based labeling of biomolecules based on the rapid 6π-azaelectrocyclization. Both DOTA as a metal chelating agent (either for MRI, PET, or other radiopharmaceutical purposes, e.g., SPECT with gamma emitters) as well as fl uorescent groups were introduced effi ciently and selectively into lysine residues within 10 min at concentrations as low as 10−8 m. The DOTA-labeled somatostatin and glycoproteins were then radiometallated with 68Ga to observe the receptor-mediated accumulation of somatostatin in pan-creatic tissue. Furthermore, microPET visualized the oligosaccharide dependent circulatory residence of glycoproteins for the fi rst time. © 2010 The Japan Chemical Journal Forum and Wiley Periodicals, Inc. Chem Rec 10: 119–139; 2010: Published online in Wiley InterScience (www.interscience.wiley.com) DOI 10.1002/tcr.200900026

Key words: synthetic (chemical) biology; 6π-azaelectrocyclization; enzyme inhibition; lysine; natural products synthesis; chiral piperidine; pyridine; PET (positron emission tomography); glycoprotein

The Chemical Record, Vol. 10, 119–139 (2010) © 2010 The Japan Chemical Journal Forum and Wiley Periodicals, Inc.

T H E C H E M I C A L

R E C O R D

Introduction

In the last few decades, organic chemists have contributed to the understanding of complex bioprocesses at the molecular level using organic reactions and/or interactions between organic molecules and either enzymes, receptors, DNA, or even cells by applying synthetic chemistry. Additionally, syn-thetic chemists strive to elucidate how biomolecules carry out their chemo-, regio-, and stereo-specifi c reactions under mild

conditions to effi ciently establish, retain, and protect biosys-tems. Hence, they have tried to realize such effi cient chemical transformations in a fl ask by developing small organic mole-cules that mimic biomolecules.

� Correspondence to: Katsunori Tanaka; e-mail: [email protected]


T H E C H E M I C A L R E C O R D

120 www.tcr.wiley-vch.de

� Shigeo Katsumura received his B.S. (1968) from Kwansei Gakuin University and PhD (1973) from Osaka City University under the direction of Prof. Takeo Sakan. In 1973, he joined the Faculty of Science, Osaka City University as an Assistant Professor, and was promoted to Lecturer in 1986 (Prof. Sachihiko Isoe). During that time, he spent two years (1977–1979) in the group of Prof. George Buchi at Massachusetts Institute of Technology (MIT) as a research associate. In 1989, he moved to Kwansei Gakuin University as an Associate Professor, and began his indepen-dent research career. In 1990, he was promoted to full Professor. His research interests include the development of new methods useful for natural product synthesis and the synthesis of biologically attractive natural products, especially terpenoids, carotenoids, alkaloids, and sphingolipids. �

� Koichi Fukase received his PhD in 1987 from the Department of Chemistry, Graduate School of Science, Osaka University under the direction of Prof. Tetsuo Shiba, and continued his research as a postdoctoral fellow in the same group. He became a Research Associate (1988–1996) at the Department of Chemistry, Faculty of Science, Osaka University (Prof. Shoichi Kusumoto). From 1994 to 1995, he joined the group of Prof. W. Clark Still at Columbia University as a postdoctoral fellow. He was promoted to Assistant Professor (1996), Associate Professor (1998) (Prof. Shoichi Kusumoto), and became a full Professor in 2004. Research projects within his group focus on 1) chemistry in innate immunity, 2) effi cient and selective methods for glycosylation and oligosac-charide synthesis, 3) new labeling methods for PET imaging, and 4) combinatorial chemistry, solid-phase synthesis, and high-throughput synthesis using affi nity separation. �

� Katsunori Tanaka received his BS (1996) and PhD (2002) from Kwansei Gakuin University in Japan under the direction of Professor Shigeo Katsumura. After a post-doc with Professors Koji Nakanishi and Nina Berova at Columbia University, New York (2002–2005), he joined Professor Koichi Fukase’s group at Osaka University as an Assistant Professor (2005-present). His research interests include exploring new methods for total synthesis, confi gurational analysis, biological evaluation, molecular imaging, and molecular recognition of natural products. �

In this regard, we have been investigating a new strategy at the chemistry/biology interface, which is summarized as the concept in the “triangle” as shown in Figure 1. Biosystems at the molecular level consist of numerous organic reactions. Thus, it is possible to completely miss unknown reactions and/or their reactivities. Thus, we envisioned a new strategy for useful chemical transformations, including applications to

natural products synthesis. In addition, such synthetic investi-gations could lead to the accumulation of many reactivity profi les, which may provide feedback to better understand biosystems. In the last decade, we have focused on concerted 6π-azaelectrocyclization, which is a “traditional and old”, but not “widely utilized” reaction. We have successfully circulated the “triangle” in Fig. 1. Herein we provide a personal account

N e w S t r a t e g y i n S y n t h e t i c B i o l o g y

www.tcr.wiley-vch.de 121The Chemical Record, Vol. 10, 119–139 (2010) © 2010 The Japan Chemical Journal Forum and Wiley Periodicals, Inc.

on how we discovered the new reactivity profi le of azaelectro-cyclization, how we applied this discovery to natural products synthesis, and how we eventually used feedback from the reac-tion to visualize in vivo dynamics of biomolecules as well as to establish methods for clinical and pharmaceutical applications; we offer a unique category in chemical and synthetic biology.

Discovery of Smooth 6π-Azaelectrocyclization: Inhibitory Mechanism of Bovine Pancreatic Phospholipase A2 by Unsaturated Aldehyde Terpenoids

Phospholipase A2 (PLA2) specifi cally catalyzes the hydrolysis of the ester linkage at the sn-2 position of glycerophospholipids, important membrane constituents.1 Normally, unsaturated fatty acids, which can be represented by arachidonic acid, are stored by an ester linkage at the sn-2 position of the glycero-phospholipid. The rate determining step to produce eico-sanoids such as prostaglandins, leukotrienes and thromboxanes through biosynthetic pathways is the release of arachidonic

acid from this position. This reaction is known as the “arachi-donic acid cascade”.2 A characteristic of PLA2 enzymes is that they exhibit a much higher activity (about 10,000 times greater activity) toward aggregated substrates than monodispersed substrates. It is postulated that the surface region of a PLA2 enzyme, which is distinct from the catalytic site, is responsible for the interaction of the enzyme with micelles of the substrate, and that this interaction enhances the catalytic ability at the active site. In particular, bovine pancreatic PLA2 shows a much stronger hydrolytic activity toward anionic micelles of the sub-strates than non-ionic micelles of the substrates. Specifi cally, based on mutagenesis experiments, Lys56 with a positive charge has been postulated to be part of the enzyme’s interfa-cial recognition site.3

We investigated the inhibitory mechanism by a marine natural product, manoalide (Figure 2),4 one of the strongest PLA2 inhibitors.5 We found that the manoalide irreversibly reacts with lysine(s) in the interfacial recognition site of bovine pancreatic PLA2. However, the detailed reaction mechanism, and thus the inhibitory mechanism at the molecular level, could not be elucidated. Nevertheless, during the course of this study, we found that (E)-3-methoxycarbonyl-2,4,6-trienal 1 (Fig. 2) shows a more powerful inhibitory activity to bovine PLA2 than manoalide.6 Hence, our research focus shifted to the inhibitory mechanism, as a model for that of manoalide.

After establishing general synthetic methods for 1 and its derivatives 2–9 (Fig. 2), namely, titanium-catalyzed hydro-magnesiation on the hydroxymethyl acetylenes,6 oxygenation of conjugated furan derivatives by singlet oxygen,7 and isom-erization of their products, the necessary functional groups for the strong inhibition of PLA2 were examined through a struc-ture/activity relationship (SAR) study. Bovine pancreatic PLA2

Feedback Discovery

New Strategy for Natural Products Synthesis

Development of New Reaction

Elucidation of Biosystems

Application

Fig. 1. Our synthetic biology triangle: schematic presentation of our research strategy.

MeO2CCHO

CO2Me

CHO

CO2Me

CHO

CHO

CO2Me

CHO

CO2Me

CHO

CHO

CO2Me

4 (++), [6 Lys]

CHO

OH

O

OH

O

8 (-), [0 Lys] 9 (-), [0 Lys]

3 (++), [2 Lys]

; Amount of the modified lysine residues

6 (-), [0 Lys] 7 (-), [0 Lys]

CO2Me

CHO

5 (+++), [2 Lys]

1 (+++), [6 Lys]

(+) ; Relative inhibitory activity.

seco-manoalide (SM)

CO2Me

CHO

after 90 min incubation2 (-), [0 Lys]

Fig. 2. Inhibitory activity and number of modifi ed lysine residues.




was reacted with 1–9 using the appropriate time intervals, and the residual PLA2 activity toward anionic mixed micelles of 1,2-dilauroyl-sn-glycero-3-phosphocholine with cholic acid was measured (Fig. 2). Although all the derivatives containing the (E)-3-methoxycarbonyl-2,4,6-trienal structure of 1, namely 3–5, showed powerful inhibitory activities, the following derivatives did not lead to signifi cant inactivation of PLA2: (Z)-stereoisomer 2 and 6, which lacked a methoxycarbonyl group, 7 in which the two carbonyl groups of 1 were exchanged, and 8 and 9, which lacked a C4- or C6-alkenyl group. Thus, the (E)-3-methoxycarbonyl-2,4,6-trienal system in molecule 1 is essential for the inactivation of bovine pancreatic PLA2.

Similar to the case of manoalide where the aldehyde moiety is essential for the PLA2 inactivation, it was hypothe-sized that an irreversible reaction of 1 with lysine residues would also cause inactivation of this enzyme. As expected, amino acid analysis indicated that six of the eleven lysine resi-dues were irreversibly modifi ed by 1 and 4, while two lysines were lost by 3 and 5. These derivatives did not react with other amino acid residues such as Leu, Ala, Ile, and Arg. Therefore, covalent bond formation with the lysine of bovine PLA2 might lead to the inactivation of hydrolytic activity for the anionic micelles of glycerophospholipids. Furthermore, derivatives 2 and 6–9, which are inactive for enzymatic inhibition, did not modify any amino acid residues in PLA2. Thus, a primary amine was used to model the lysine residue of PLA2 and to examine the reaction mechanism between (E)-3-methoxycar-bonyl-2,4,6-trienal compounds and the lysine residues.

The reaction of 1 with n-propylamine in 1,4-dioxane quantitatively yielded 1,2-dihydropyridine derivative 1DHP within 5 min at room temperature (Scheme 1). Additionally, reactions with 3–5 produced the corresponding dihydro-pyridine derivatives (3DHP–5DHP) with n-propylamine. The reaction proceeded via a 6π-azaelectrocyclization of the inter-

mediary Schiff base, 1-azatriene 1N. On the other hand, (Z)-stereoisomer 2 and 6–9, which did not exhibit signifi cant inhibitory activities, gave only the corresponding Schiff bases within 60 min at room temperature. Obviously, the (E)-3-methoxycarbonyl-2,4,6-trienal system is essential for a smooth 6π-azaelectrocyclization of the intermediary 1-azatriene as well as to signifi cantly inhibit PLA2. Namely, 1 and 3–5 reacted with lysines to give 1,2-dihydropyridine derivatives, and led to enzyme inhibition.

In 1989, Okamura and co-workers found that the reac-tion between 13-tert-butyl-13-cis retinal and n-butylamine for 1 h at room temperature provided the corresponding 1,2-dihydropyridine instead of the Schiff base (vide infra).8 Through structure-reactivity studies, they examined how steric, electronic, and conformational factors affected the reactivity of the disrotatory 6π-electrocyclization. The tert-butyl group preferred the s-cis conformer, resulting in the rapid azaelectrocyclization arising from the derived product-like cyclic conformation. On the other hand, our results were the fi rst observation where both the C4-carbonyl and C6-alkenyl groups in 1-azatriene systems signifi cantly contributed to the acceleration of 6π-azaelectrocyclization.

The irreversible modifi cation of lysine by 1 was also con-fi rmed by MALDI-TOF-MS (Figure 3). The ion peaks from irreversibly modifi ed PLA2 by “one”- to “seven”-molecules of 1 could be detected after 15 min in the reaction of 1 with PLA2 at 40 °C. Similarly, the mass spectra provided dihydropyridine signals of modifi ed PLA2 from the reactions with 3–5. The model reactions with primary amines clarifi ed the new inacti-vation mechanism of PLA2 by these trienal terpenoids.

As previously mentioned, especially for anionic micellar phospholipids, the presence of the interfacial recognition site, which consists of cationic Lys56, has been widely accepted for the hydrolytic mechanism of glycerophospholipids by bovine

N

CO2Me

N

CO2Me

N

CO2Me

CO2Me

R

CO2Me

RR

R

CO2Me

CO2MeR

CO2Me

CHO

N

CO2Me

N

CO2Me

5DHP

(R= -CH=N-C3H7)

7N

3DHP 4DHP

8N2N 9N

PrNH2

1 1DHP

6N

1Nacceleration

accelerationRapid

(b)

(c)

(a)

6 -azaelectrocyclization

Scheme 1. Model reaction with propylamine. (a) Reaction mechanism of dihydropyridine formation of 1 with propylamine via azaelectrocyclization. (b) Dihydropyridine products. (c) Schiff base formation.



Lys116, which are located on the outer face of the protein, Lys56 was responsible for the modifi cation. Therefore, irre-versible 1,2-dihydropyridine formation might eliminate the positive charge at Lys56 in the interfacial recognition site of this enzyme and disturb the bindings between anionic micelles and PLA2, leading to enzyme inhibition. In our studies to elucidate the inhibitory mechanism of the bovine PLA2 by unsaturated aldehyde terpenoids, we discovered remarkable substituent effects, which signifi cantly accelerated azaelectro-cyclization (Figure 4).

Renaissance of 6π-Azaelectrocyclization: Remarkable Acceleration by Substituent Effects

The thermal cyclization of 1-azatrienes to 1,2-dihydropyridines, known as 6π-azaelectrocyclization,9 is analogous to the well-known ring closure of 1,3,5-hexatrienes to 1,3-cyclohexadienes.10–12 This process likely involves a classical concerted electrocyclization, which proceeds in a disrotatory mode. For example, cis-dienone oximes formed pyridines at 160 °C in about 25% yield through azaelectrocyclization followed by dehydration of intermediary 1,2-dihydropyridine derivatives (Scheme 2a).13 Additionally, O-acyl hydroxamic acid

Fig. 3. Formation of dihydropyridines with lysines of PLA2 through azaelec-trocyclization. a) Intact PLA2. b) Modifi ed PLA2 by 1 for 15 min at 40 °C.

Fig. 4. Inhibitory mechanism of PLA2 by aldehyde terpenoids.

pancreatic PLA2.3 From separate experiments on the enzymatic digestion/MALDI-TOF-MS peptide mapping of the modifi ed PLA2 by derivative 5, we identifi ed the critical Lys residues for the PLA2 inactivation by 5.6b In addition to Lys113 and




derivatives gave N-acyl-1,2-dihydropyridine derivatives through electrocyclization of N-acylazatrienes, which were produced by heating hydroxamic acid at 650 °C under fl ash vacuum pyrolysis (Scheme 2b).14 Although numerous examples of 6π-azaelectrocyclization of 1-azatrienes have been reported, the reactivity study as well as applicability to natural products synthesis are very limited in the literature probably because the cyclization process requires high temperature and the prepara-tion methods for 3-cis-1-azatrienes, precursors of the electrocy-clization, are not well established.

As noted previously, Okamura and co-workers have reported that Schiff base 10, which was obtained from 13-tert-butyl-13-cis retinal and n-butylamine, produced 1,2-dihydropyridine at a relatively fast rate, namely at 23 °C with a half-life of 11 min (Scheme 2c).8 Their explanation for this rate is based on conformational analysis. The predominant 12,13-s-cis arrangement of the 1-azatriene (Schiff base), i.e., a “product-like” cyclic conformation led to a smooth electro-cyclic reaction. Furthermore, using 4-tert-butyl-1-azatriene templates for azaelectrocyclization, they found that introduc-ing either an electron donor or an acceptor group at the 1-azatriene terminus moderately, but not remarkably, accelerated the reaction (Scheme 2c).15

In our studies on the enzyme inhibition of unsaturated aldehydes, we found that the presence of both the C4-carbonyl and C6-alkenyl groups in 1-azatriene remarkably accelerated

azaelectrocyclization. Nevertheless, we encountered some limitations in applying this fi nding to a wide range of compounds such as natural alkaloid synthesis. Our investiga-tions focused on the azatriene system, containing a 2,6,6-trimethylcyclohexene substituent, was a characteristic system owing to its steric bulkiness and fairly twisted nature around the C6-C7 single bond of the azatriene. Therefore, the sub-stituent effects could be rationalized in more general azatriene systems such as 11–14 from a viewpoint of molecular orbital interactions.16

Below we describe an effi cient and general synthesis of 3-cis-1-azatriene derivatives. Based on cyclization studies, we concluded that the combination of the C4-carbonyl and the C6-conjugating groups in 1-azatrienes contributed to the strong orbital interactions between the HOMO of the olefi n part and the LUMO of the azadiene part under the reverse electron demand mode. Thus, we developed our smooth azaelectrocyclization into a new strategy for natural products synthesis.

Synthesis of 3-cis-1-Azatriene Derivatives and Their Reactivities toward Azaelectrocyclization

Although our previous research established the stereoselective synthesis of (E)-3-alkoxycarbonyl-2,4,6-trienal compounds by hydrometalation of ethynyl groups as the key step,6 we knew

N

OCO2Me

CO2MeN

CO2MeN

CO2Me

NHO

NHO

Ph Ph

N

Ph

650 oC

160 oC

25%

18 h

13

13

CHO

Nn-Bu

Nn-But-Bu

N

n-Bu 12, 13-s-cis

12

12

12

13

10

n-BuNH2

0 oC, rt, 1h

(a)

(b)

(c)

Scheme 2. Previous examples of azaelectrocyclization of 1-azatrienes.



that this protocol was not applicable to a wide range of deriva-tives. Thus, we developed a more general route using Pd(0)-catalyzed Stille coupling between (E)-vinylstannane VS and (Z)-vinyl halide VH (Scheme 3).16 The reaction between stan-nane 15 and bromide 17 proceeded smoothly in the presence of 5 mol% tetrakis(triphenylphosphine)palladium(0) and two equivalents of lithium chloride in dimethyl formamide at 115 °C to give the corresponding alcohol in 71% yield while retaining the stereochemistry. Oxidation of the alcohol with manganese dioxide provided 11 in 80% yield. Similarly, cou-pling between 16 and 17, between 15 and 18, and between 16 and 18, produced the corresponding alcohols in 71, 91, and 77% yields, while oxidation led to the corresponding aldehydes and α-pyrane, 12P, 13, and 14 in 64, 67, and 59% yields, respectively. The equilibrium favored pyrane 12P with a small amount of aldehyde 12 by electrocyclization.

The azaelectrocyclization process was directly monitored by 1H NMR at 23 °C by the treatment of 11–14 with 1.1

equivalents of n-propylamine in [D6]benzene (Scheme 3 and Figure 5a). Similar to the case with 1 (see Scheme 1), (E)-3-carbonyl-2,4,6-trienal 11 smoothly gave 1,2-dihydropyridine 11DHP through 1-azatriene 11N within 5 min. In contrast, trienal 14, which lacked both C4-carbonyl and the C6-alkenyl groups, smoothly produced the corresponding Schiff base (azatriene 14N), but it slowly cyclized to dihydropyridine 14DHP. Azatriene 13N, which was derived from 13 but lacked a carbonyl group, was almost inactive toward cyclization within 7 h. The equilibrium mixture between 12, 12P, and 12N slowly provided dihydropyridine 12DHP over 10 h. Thus, the results observed in the retinoid system could be applied to more general linear compounds, namely, the presence of a carbonyl group at C4 and an alkenyl group at C6 accelerated 6π-azaelectrocyclization.

To further understand the acceleration effects, p-substituted phenyl derivatives (19N–21N), the carboxylic acid, amide derivatives (22N and 23N), and N-hydroxy derivatives

SnBu3RO

SnBu3RO

I

OH

Br

CO2Et

OH

CO2Et

X

ROCO2Et

X

RO

X

RO

X

RO

H

H H

H

R1

O

RO

H

SnBu3RO

Hal R1

OH

CO2Et

ORO

12P + 12

14

17

16 + 17

15 + 18

15 + 17

16 + 18

a, b

15

16

18

12P

+

(R = TBDMS)

11 : X = O11N : X = NPr

12 : X = O12N : X = NPr

13 : X = O13N : X = NPr

14 : X = O14N : X = NPr

(R1 = -CO2Et, Me)

11-14

11

13

VS

VH

a, b

a, b

a, b

Scheme 3. General synthesis of 1-azatrienes and aldehyde precursors. Conditions: a) Pd(PPh3)4, LiCl, DMF, 75–120 °C, 71–91%; b) MnO2, CH2Cl2, 59–80%.




(24–26) were also examined (Fig. 5b-d). These derivatives could readily be obtained through either hydrometalation or the cross-coupling protocols established above.16 Azatrienes 19N-21N rapidly produced the dihydropyridines within 10 min. The C6-phenyl substituents similarly accelerated azaelectrocyclization, and clear effects owing to the para-substituents could not be observed under the conditions in Fig. 5b. Moreover, 22N and 23N provided dihydropyridines within 1 and 3 h, respectively (Fig. 5c). Similarly, both the carboxylic acid and amide substituents contributed to the acceleration, but the effects were smaller than the ester group with stronger electron-withdrawing properties. Although cyclization was not observed in 24 and 25 within 60 min

at 23 °C (Fig. 5d), the corresponding O-acetyl congener 26 gave pyridine 27 within 40 min through azaelectrocyclization and subsequent elimination of acetic acid from the intermedi-ary dihydropyridine. The low reactivities of 24 and 25 were presumably as a result of the unfavorable electrostatic repulsion between the oxygen atom at N1 and C5-C6 π-electrons devel-oping at the transition state or the thermodynamic stability of the oximes relative to the cyclized products. Schiess and co-workers have reported that cis-dienone oximes and their benzoate derivatives gave the corresponding pyridines upon heating at 50–160 °C.13 Our unexpected pyridine synthesis can readily be performed in one-pot by treatment of (E)-3-carbonyltrienal with hydroxylamine and acetyl chloride at

CO2Et

NRO

CO2Et

NRO

NRO

NRO

n-Pr

n-Pr

n-Pr

n-Pr

140

20

TIME (min)

60

280 420

100

0

Pro

du

ce

d D

HP

(%

)P

rod

uce

d D

HP

(%

)

Pro

du

ce

d D

HP

(%

)

80

40

12N

14N

(R = TBDMS)

11N

13N

N

R

N

CO2Et

n-Pr n-Pr

R

22N: R = -CO2H20N: R = -OMe 23N: R = -CONHPr

19N: R = H

21N: R = -CO2Me

1N: R = -CO2Me

N

CO2Me

R

CO2Me

NAcO

CO2Me

N

cyclization

27

- AcOH

20

20

40

80

30

20

40

20 30

TIME (min)10

100

60

0

80

60

0TIME (min)

100

10

24: R = OH

(a)

(b) (c)

(d)

25: R = OMe26: R = OAc

R = OAc (26)

Fig. 5. Electrocyclization studies. Reactions were monitored by 1H NMR at 23 °C in [D6]benzene. y-Axis shows the percentage of the produced dihydropyridine while x-axis shows time.



room temperature, and therefore, has potential for natural products synthesis (vide infra).

Rationale for Acceleration of 6π-Azaelectrocyclization by Computational Analysis

The SPARTAN software package was used to rationalize the remarkable substituent effects based on molecular orbital cal-culations of 1N, 6N, and 9N (Figure 6, also see Scheme 1).16 The energies of the azatriene conformations, 2,3-s-cis, 4,5-s-cis, necessary for electrocyclization, were minimized by the semiempirical method (PM3) using SPARTAN version 5.0 (Wavefunction Inc., Irvine CA), and their frontier orbitals and HOMO-LUMO energy gaps were calculated at the HF/6–31G* level. Among the three azatrienes 1N, 6N, and 9N, the energy gap of 1N was the smallest, and the HOMO and LUMO electron densities were suitably arranged around C5-C6 and N1-C4, respectively. Both the C4-carbonyl and the C6-alkenyl groups contributed to the strong interaction between the HOMO of the C5-C6 olefi n and the LUMO of the N1-C4 azadiene in the 1-azatriene system, and thus effec-tively promoted azaelectrocyclization. Such substituent effects are similar to those of the reverse electron demand aza-Diels-Alder reactions where electron defi cient azadienes smoothly react with electron rich dienophiles. Additionally, the analysis is consistent with the experimental results where the aromatic groups at C6 and some other electron-withdrawing groups at C4 accelerated the reaction. Hence, the faster reaction rate of 1N with a stronger electron-withdrawing ester than 22N and 23N (Fig. 5c) is reasonable. Thus, we realized that the substitu-ent effect signfi cantly accelerated 6π-azaelectrocyclization owing to the strong HOMO-LUMO interaction. Using this knowledge, we applied the reaction to natural products synthesis.

6π-Azaelectrocyclization as a New Strategy for Natural Products Synthesis

One-pot Pyridine Synthesis: Formal Synthesis of the Ocular Age Pigment A2-E

Pyridinium bisretinoid “A2-E”, which may be closely related to age-related macular degeneration (AMD),17 was isolated from over 40 aged human eyes as the major orange fl uorophore of ocular age pigments, called lipofuscin. Nakanishi and co-workers have elucidated and realized the biosynthetic pathway of A2-E both in vitro and in vivo as shown in Scheme 4.18 Two equivalents of all-trans-retinal and ethanolamine gave azatriene 28, which participated in 6π-electrocyclization and autoxidation. They also achieved the “complete” chemical syn-thesis of A2-E by double Wittig olefi nation with bis-aldehyde 29.18 As hinted by Nakanishi’s route (Scheme 4), we investi-gated the biomimetic synthesis of A2-E using our smooth azaelectrocyclization.16,19 Our synthesis, shown in Scheme 5, featured two types of “one-pot” procedures for substituted pyridines.

In the fi rst trial for a “one-pot” pyridine synthesis, which was based on previous fi ndings shown in Fig. 5d, aldehyde 30, which was prepared according to the method in Scheme 3, was treated with hydroxylamine in pyridine for 15 min, and then with acetyl chloride for 10 min at room tempera-ture to give desired pyridine 31 in 53% yield.16 As an alterna-tive method for the biomimetic synthesis of A2-E, 30 was reacted with excess lithium bis(trimethylsilyl)amide in THF at room temperature, which cleanly produced the corresponding N-trimethylsilyl-1,2-dihydropyridine within a minute through the Peterson reaction and subsequent smooth azaelectrocyclization.19 Continuous treatment of intermedi-ary 1,2-dihydropyridine with 2,3-dichloro-5,6-dicyano-1,4-benzoquinone (DDQ) successfully constructed the pyridine

CO2Me

Nn-Pr

Nn-Pr

CO2Me

Nn-Pr

Nn-Pr

Nn-Pr

O

OMe

Nn-Pr

O

OMe

5

0.431154(hartrees)

0.4549130.404912

HOMO

2

4

8

1

7

LUMO

6

HOMO-LUMO energy gap

1N

3

9N6N

Fig. 6. Pictorial presentation of the MO calculation results of 1-azatrienes 1N, 6N, and 9N.




nucleus, and the reduction of the crude products by lithium aluminum hydride gave 32 in 77% total yield from 30. Finally, silyl-deprotection and oxidation provided the bis-aldehyde 29. Thus, the formal synthesis of A2-E through the proposed biosynthetic pathway of this natural product was realized.

Library-Directed “One-pot” Solution- and Solid-phase Synthesis of 2,4-Disubstituted Pyridines

Recently, electrocyclization-directed pyridine synthesis was further extended to realize multiple sequences of “one-pot” including even the preparation of the conjugated aldehyde

(Table 1).20 Various 2,4-substituted pyridines could easily be obtained by mixing vinyl stannanes, vinyl iodide, sulfonamide, and a palladium catalyst at 80 °C. This reaction might proceed through (i) Stille coupling, (ii) azatriene formation, (iii) aza-electrocyclization, and (iv) eliminative aromatization path-ways. The protocol was also applicable to the solid-phase synthesis (Table 1). Employing a “traceless” sulfonamide linker enabled a small, highly pure pyridine library to be rapidly prepared without chromatographic separation.20b Thus, a DMF solution of styryl stannanes 33–42, vinyl iodide 43, Pd(PhCN)2Cl2, and LiCl was shaken with the sulfamylbutyryl AM resin (Novabiochem) at 80 °C for 2 h. After the excess

NOH

OHC

N

CHO

N

R

R

OH

N

R

R

OH

CHO

H2NOH

+

28;

+

+

+

R =

A2-E

azaelectrocyclization

[O]

Synthetic intermediate 29

Scheme 4. Biosynthesis of A2-E.

RO

CO2Et

CHO

ROCO2Et

NAcO

ROCO2Et

NAcO

TBDMSOCO2Et

N

ROCO2Et

NTMS

(R = TBDMS)

1) NH2OH•HCl

ROCO2Et

NTMS

TBDMSO

N

2) AcCl / pyr

Deacetylation

53 % from 30

OH

1) LiN(TMS)2

2) DDQ

3) LiAlH4

1) TBAF

2) MnO2

30

31

30

32 77 % from 30

29

(a)

(b)

Scheme 5. One-pot pyridine synthesis and biomimetic synthesis of A2-E.



reagents were washed off, the resulting dihydropyridine-bound resin was treated with DBU to effectively eliminate sulfonic acid, which in turn cleaved pyridines from the resin. After a simple extraction procedure, various 2,4-disubstituted pyri-dines 33p-42p could be effi ciently and rapidly accessed in high purity (Table 1). Except for a few cases, such as entries 7 and 10, the solid-supported protocol provided the corre-sponding pyridines in comparable yields to those obtained by the solution-phase procedure. It is noteworthy that these substituted pyridines were obtained in pure forms without chromatographic separation. Thus, this method could be expanded to library-directed combinatorial synthesis of substi-tuted pyridines.

Highly Stereoselective Asymmetric 6π-Azaelectrocyclization

We also investigated stereoselective asymmetric azaelectrocy-clization as another unique synthetic application.21 Our

strategy was based on controlling the torque selectivity of the disrotatory cyclization, which would be affected by a chiral auxiliary on the nitrogen under mild conditions (Scheme 6). A simple operation, treatment of (E)-3-carbonyl-trienals with chiral amines, provided a new entry for the syn-thesis of chiral dihydropyridine, a useful synthetic unit for natural alkaloids.

We initially selected trienal 1, which contains a bulky 2,6,6-trimethylcyclohexene moiety as an aldehyde, and explored an effective amine that acts both as a “nitrogen source” and as a “chiral auxiliary” to realize high torque selec-tivity of azaelectrocyclization (Scheme 6). Our initial attempt utilizing more than twenty of the readily available amines, which have been recognized as effective chiral auxiliaries in the imine-based asymmetric reaction, was not promising for the present objective. Fortunately, we found that (1S,2R)-(-)-cis-1-amino-2-indanol a quantitatively produced pentacyclic piperidine (-)-1a, as a single stereoisomer (Scheme 6).21 The reaction proceeded via azaelectrocyclization in a completely

Table 1. Library-directed solution and solid-supported one-pot pyridine synthesis.

RSnBu3

OHC

CO2EtI N

R CO2Et

33p-42p

Pd cat43

33-42

MeSO2NH2

or

NH

O

SNH23

N

R CO2Et

R'SO2

O O DBU

entry R productSolution-Phase

Yield (%)aSolid-Phase Yield (%)c entry R product

Solution-Phase Yield (%)a

Solid-Phase Yield (%)c

1

33

33p 90 76 6

NTs

38

338p 77 69

2N

34

2

34p 58b 40d 7NSO2Ph

2

39

39p 49 31

3

N

35

335p 56 76 8

40

40p 54 75

4N

36

36p 67 52d 9O TBS

41

41p 13 22d

5

S

37

37p 76 78 10

42

TBSO 42p 60b 31

aPd(PhCN)2Cl2 (10 mol%), LiCl, DMF, 50–70 °C, 1–4 h then DBU (1.0 equiv), RT, 1 h. bPd2(dba)3 (20 mol%), P(2-furyl)3, LiCl, DMF, 80 °C, 1–4 h then DBU (1.0 equiv), RT, 1 h. cPd(PhCN)2Cl2 (20 mol%), LiCl, DMF, 80 °C 2 h, then DBU (1.0 equiv), THF, RT, 1 h.dPd(PhCN)2Cl2 (20 mol%), LiCl, DMF, 60 °C 5 h, then DBU (1.0 equiv), THF, RT, 1 h.




stereoselective manner, and then isomerization of the dihydro-pyridine yielded the thermodynamically more stable aminoac-etal (-)-1a. To achieve an excellent torque selectivity, the vicinal cis-hydroxy group with respect to the amino group of (-)-a was essential because the reaction with either 1-aminoindane, cis-1-amino-2-methylindane, or cis-1-amino-2-methoxyindane gave a 3 : 1 mixture of diastereomeric dihydropyridines. The hydroxyl also protected dihydropyri-dine as its aminoacetal and contributed to the stabilization; therefore, dihydropyridine equivalent 1a could be easily handled for further functional transformations as shown in the following synthetic applications (vide infra).

Although cis-aminoindanol was an encouraging amine candidate, reactions with the more general aldehydes 11 and 19, which contained linear alkenyl or phenyl substituents, only gave a 3 : 1 diastereomixture of the corresponding piperidines (Table 2, entries 1 and 2). Nevertheless, introducing an alkyl substituent at the 7-position of (-)-a signifi cantly improved the diastereoselectivity. Reactions with the methyl-substituted (-)-b provided piperidines (-)-11b and (-)-19b in diastereo-meric ratios of 5 : 1 and 12 : 1, respectively (entries 3 and 4). Furthermore, applying ethyl-substituted c improved the selec-tivities to 7 : 1 and 20 : 1, respectively (entries 5 and 6). The highest selectivity was obtained with isopropyl congener (-)-d as piperidine (-)-11d and (-)-19d were obtained in ratios of 10 : 1 and 24 : 1 (entries 7 and 8), respectively. It is noteworthy that at a lower temperature of 13 °C, (-)-19d was obtained as almost a single isomer (entry 9). Unexpectedly, reactions with

bulkier tert-butyl-substituted congener e slightly decreased the selectivity (5 : 1 and 17 : 1 for 11e and 19e, respectively, entries 10 and 11).

At the same time as our fi rst report on the asymmetric reaction, Hsung and co-workers successfully employed a stereoselective azaelectrocyclization of their conformationally restricted 1-azatrienes under thermodynamically equilibrated conditions.22 In our case, the observed diastereoselectivity was not a result of thermodynamic equilibration because the equilibrium between the separated diastereomers could not be observed under identical reaction conditions (Table 2). To the best of our knowledge, our report was the fi rst to achieve highly stereoselective asymmetric 6π-azaelectrocyclization of linear 1-azatrienes in a kinetically controlled manner.

Table 3 shows the oxidative cleavage to remove the chiral auxiliary in 1a-19d. Diols derived from 1a-19d by LiAlH4 reduction were treated with manganese dioxide in ether at room temperature followed by silica gel chromatography to provide aminoalcohols (-)-43 and (-)-44 in 55–69% yields.21 A detailed mechanistic investigation21b of this novel phenom-enon revealed that manganese dioxide initially oxidized the nitrogen atom of the diol to produce the intermediary N-oxide. After a Polonovski-type reaction, an acid-mediated dehydration owing to the removal of the reactive benzylic hydrogen located in the anti-parallel direction toward the N-oxygen bond, the hydrolysis of the resulting imminium ion gave the products. The vicinal cis-hydroxy group of the

CO2Me

N

OH

NH2

OHCO2Me

CHO

CO2Me

N

OH N

CO2Me

O

CO2Me

N

OH+

Single isomer

CDCl3, 24 oC

2h, quant.1

[ ]24D -45.4o (c1.1, CHClα 3)

CHO

CO2Et

RNH2

NR

CO2Et

NR

CO2Et

*

Chiral dihydropyridine

2

* *

*

(-)-1a

**

**

Torque selectivity?

(-)-a

(i) For selectivity

(ii) For protection

Scheme 6. Asymmetric azaelectrocyclization.



Table 2. Asymmetric azaelectrocyclization with substituted cis-1-amino-2-indanols.a

R

CO2Et

CHO

X NH2

OH

TBSO

O

R CO2Et

NX

O

R CO2Et

NX

conditionsb

6

19 : R =

7

+

2 2

***

6

11a-19e

11 : R = a: X = H b: X = Me c: X = Et d: X = iPr e: X = tBu

4

entry aldehyde amine product (major isomer) temp (°C) dr (at the 2-position)c

1 11 (-)-a 11a 24 3 : 1d

2 19 (-)-a (-)-19a 24 3 : 1 3 11 (-)-b (-)-11b 24 5 : 1 4 19 (-)-b (-)-19b 24 12 : 1 5e 11 c 11c 24 7 : 1 6f 19 c 19c 24 20 : 1 7 11 (-)-d (-)-11d 24 10 : 1 8 19 (-)-d (-)-19d 24 24 : 1 9 19 (-)-d (-)-19d 13 >40 : 110g 11 e 11e 24 5 : 111h 19 e 19e 24 17 : 1a Unless noted, all reactions quantitatively provided a diastereomeric mixture of two piperidine derivatives with the shown stereochemistry. Relative stereochemistry of the products was determined based on their 1H NMR and NOE experiments by comparing to that of 1a. Stereochemistry of the 6-position of the minor isomers was not determined.b CHCl3, 3h. cDetermined by 1H NMR (400 MHz). d Inseparable mixture of four piperidine derivatives (15 : 5 : 3 : 1) was obtained. Diastereomeric ratio at the 2-position was determined after LiAlH4 reduction of the crude products. e–h Racemic c and e were employed to examine the diastereoselectivity of azaelectrocyclization.

Table 3. Oxidative cleavage of indanol auxiliary.

CO2Et

N

O

XR

N

OH

XR

OH

HN

ROH

(-)-1a-19d*

*

*

then, SiO2

MnO2a

(-)-19: R =

(-)-1: R =

LiAlH4

N

HO

OHR

NO

R

OH

O-

(iii) Facilitating oxidation

H+

HMn(IV)

*

entry substrate product yield of diol (%) yield of aminoalcohol (%)

1 (-)-1a (-)-43 100 692 (-)-19a (-)-44 76b 553 (-)-19b (-)-44 91 554 (-)-19d (-)-44 91c 58a Reaction was performed by treating diols with manganese dioxide (10–20 w/w, chemicals treated, Wako) in ether for a few minutes at room temperature. b,c Total yield from aldehyde 19.




cis-1-amino-2-indanol was essential for a smooth oxidation of the nitrogen under such mild conditons.

Overall, we opened a new route for chiral piperidine synthesis through asymmetric azaelectrocyclization (Scheme 7). cis-Aminoindanols23 acted not only as a “chiral auxiliary” but also as a “nitrogen source” for asymmetric azaelectrocycli-zation. In addition, the cis-hydroxy group of the aminoindanol played the following important roles: (i) helped improve

torque selectivity, (ii) stabilized unstable N-alkyl 1,2-dihydropyridines, and (iii) facilitated the removal of the chiral auxiliary under mild conditions. These advantageous features make this chiral amine a “multi-functional” system for asymmetric electrocyclization. As demonstrated by applying our asymmetric azaelectrocyclization to the natural piperidine synthesis of 20-epiuleine (Scheme 8), this method is simple: mix aldehydes with an amine in chloroform at room tempera-ture, allow the reaction to reach completion, and evaporate the solvent.

Synthesis of 20-Epiuleine

Uleine, 20-epiuleine, and dasycarpidone, which have been principally isolated from Aspidosperma sp., constitute a small family of Strychnos type indole alkaloids (Scheme 8).24 Although many racemic uleine synthesis have been reported, few asym-metric versions have appeared in the literature. One common approach to the racemic 2-azabicyclo[3.3.1]nonane structure, which is found in Strychnos indole alkaloids, includes C2-C16 bond formation of 3-(2-piperidyl)indoles (Scheme 8).24 Hence, we envisioned a stereoselective synthesis for optically active

R

CO2Et

CHO

CO2Et

N

O

NH2

OH

Alkyl

R3

HN

R4

R1

RR

Nitrogen provider

R = Alkenyl, Phenyl

+

Chiral auxiliary

(i) Improved selectivity(ii) Protection of dihydropyridine(iii) Oxidative removal of auxiliary

*

Substituted Piperidine

R2

Alkyl*

*

Alkylation

Alkylation

* * ***Asymmetric

azaelectrocyclization

Scheme 7. Asymmetric azaelectrocyclization and substituted chiral piperidine synthesis.

NTs

SnBu3

X

CO2Et

OH

R NH2

OH

NTs

RN

CO2Et

O

OH

HN

NTs

MeN

NH

O

LiAlH4

MnO2

NTs

CO2Et

CHO

OH

MeN

NTs

OH

OH

NR

NTs

then, SiO2

+

(-)-50b; R = Me, 96%

CHCl3, 24 oC, 3h

2) MnO2, 87 %

(b; R = Me)10 : 1

1) Pd(PPh3)4, LiCl

DMF, 115 oC

*

(-)-50d; R = iso-Pr, 99%

R = Me, 95%(d; R = iso-Pr)

R = iso-Pr, 83%

(47; X = Br)(48; X = I)

1) MnO2

HCHO, NaBH3CN

R = iso-Pr, 74%

ether, rt, 3 min.

2) MeLi(55 % for 2 steps)

Key intermediate (-)-45

quant

4) MnO2

3) NaOH

(40 % for 2 steps)

R = Me, 73%

[ ]24D -3.8o (c 0.3, CHClα 3)

4649

(-)-51

(X = I, 72 %)

Husson et al.

H

HNH

MeN

20

20-epiuleine

2

16

*

*

Scheme 8. Formal synthesis of epiuleine through asymmetric azaelectrocyclization.



2- and 4-substituted piperidine 45 (Scheme 8)25 based on our asymmetric azaelectrocyclization. Thus, aldehyde 49, which was prepared from vinylstannane 46 and vinyl halide 47 or 48 according to the established method in Scheme 3, was reacted with (-)-b or (-)-d to provide corresponding piperidines (-)-50b or (-)-50d in almost quantitative yield and a 10 : 1 diastereose-lectivity. After the minor isomers were separated, (-)-50b or (-)-50d was reduced with lithium aluminum hydride and subseqeunt treatment of the diol with manganese dioxide gave the desired aminoalcohol (-)-51 in 73% and 74% yields, respectively. Finally, (-)-51 was converted into (-)-45 by simple functional group manipulations as shown in Scheme 8. Because racemic 45 has already been converted into natural 20-epiuleine by Husson and co-workers,25 we achieved the formal synthesis of this natural alkaloid.21

In addition to the asymmetric synthesis described above, we have also developed a more sophisticated approach to sub-stituted chiral piperidines using a similar protocol as that for pyridine synthesis described above using a “one-pot” process directly from the vinyl stannanes, halides, chiral amino indanol, and Pd(0) catalysts.26 This method was also applied to the synthesis of the dendroprimine, an indolizidine natural alka-loid containing the 2,4,6-trisubstituted piperidine motif as well as other substituted chiral piperidines, which are described in detail in our original publications.26 Thus, our smooth azaelectrocyclization can be regarded as a new strategy for natural product syntheses.

In summary, we discovered remarkable substituent effects that accerelate disrotarory 6π-azaelectrocyclization through an inhibitory mechanism of the aldehyde terpenoids, natural products derivatives. We thoroughly investigated these effects based on molecular orbital interactions, and used this knowl-edge to develop a new synthetic strategy for natural and phar-macologically important compounds. So to circulate the triangle depicted in Fig. 1, the next question became how can we use this information from our synthetic applications on electrocyclization reactivity for bio-directed research? We eventually directed our attention to the research fi eld of molec-ular imaging of biomolecules by effectively applying our smooth azaelectrocyclization to label lysines.

6π-Azaelectrocyclization-based Microgram-scale Labeling: Biomolecule-Based In Vivo Imaging

Molecular imaging is an important topic that has garnered signifi cant attention in the fi elds of chemical biology, drug discovery, and diagnosis. Positron emission tomography (PET) is a common non-invasive method, which quantitatively visu-alizes the locations and levels of radiotracer accumulation with high imaging contrast.27 2-Deoxy-2-[18F]fl uoro-D-glucose (18F-FDG) has become an important small molecule-based

PET tracer, which is often used to detect primary and meta-static cancers.28 However, 18F-FDG PET, which only measures the glycolysis inside cells, can often be confounded by benign diseases as well as by active infl ammations, which also exhibit an elevated uptake of this tracer, leading to ambiguous conclu-sions. To circumvent these problems associated with 18F-FDG, a variety of new small-molecule PET probes have been devel-oped. An interesting possibility for cancer diagnosis is to use biomolecules as molecular imaging probes27,28 because the desired peptides, proteins, monoclonal antibodies, and oligo-nucleotides, exhibiting a high binding affi nity to the target receptors, antigens, and nucleic acids being specifi cally overex-pressed in or on tumor cells, can be easily chosen or designed. Alternatively, PET imaging of biomolecules can visualize unknown in vivo kinetics where an important biological pathway is involved, which may lead to the discovery of prom-ising biomolecule-based drug candidates.

Generally, PET imaging of these biomolecules is achieved by conjugation to metal chelating agents such as DOTA (1,4,7,10-tetraazacyclodecane-1,4,7,10-tetraacetic acid) or DTPA (diethylenetriaminepentaacetic acid) prior to radiola-beling with a suitable β+ emitting metal.28 Chelating agents can be introduced either during the solid-phase synthesis of pep-tides or more directly through reaction of lysine residues or N-terminal amino groups with DOTA-O-sulfosuccinimidyl ester.28,29 However, the latter method, which uses activated DOTA ester, usually proceeds slowly and requires as much as a few milligrams of sample to maintain high reaction concen-trations. Additionally, the conjugate effi ciency is not high, and more importantly, labeling under such high reagent concentra-tions and long reaction times indescriminately modifi es key lysines, and eventually kills the function of the biomolecules. Because PET experiments require only small amounts of tracers, i.e., p-μ g, and important biomolecules are occas-sionally obtained only in small amounts, a subpicogram-scale and non-destructive conjugation methodology would greatly expand the applicability of PET imaging. Considering such a situation in molecular imaging fi elds, we devised the idea of applying our rapid 6π-azaelectrocyclization to lysine labeling,30 which in turn led to effi cient PET- and/or fl uorescence-base in vivo imaging protocols.

Although various biomolecule labeling methods employ-ing advanced organic reactions, including Sharpless/Meldal Click reaction and Bertozzi’s click and/or Staudinger ligation, have been reported,31 an effi cient, rapid, and mild protocol without using toxic metals that is applicable in various buffer solutions had yet to be discovered. As mentioned previously, we have found that unsaturated (E)-ester aldehydes quantita-tively reacted with primary amines, including lysine, within a very short time in various buffer solutions to yield 1,2-dihydropyridines as irreversible products. To the best of our knowledge, our reaction, where unsaturated aldehyde 1 reacts




with enzyme to modify several Lys residues within 15 min at room temperature (see the MALDI-TOF-MS in Fig. 3), is the fastest conjugation reaction with lysines in water.6 Thus, we wondered if this reaction would work as a new protocol for Lys labeling.

Development of Non-destructive Lys-labeling Kit “Stella+” by 6π-Azaelectrocyclization

Inspired by the notable electrocyclization reactivity during our enzyme inhibition study and in the synthetic applications described above, a new DOTA-labeling probe, DOTA-(E)-ester aldehyde 55a, was designed and synthesized (Scheme 9).30 Thus, THP-protected (Z)-vinyl bromide 53 and glycine linked (E)-stannane 52 were heated to 110 °C in the presence of Pd2(dba)3, P(2-furyl)3, and LiCl in DMF to provide the Stille coupling product, which was subsequently treated with 3 m hydrochloric acid to give aminoalcohol 54 in 73% yield in two steps. Currently, compound 54 can be prepared on a 50 g scale. This scaled up reaction allowed us to develop a complete electrocyclization-labeling process as a labeling kit “Stella+”, which can be utilized for bio-directed research with broader applications (vide infra).32 This is a good example to demonstrate how important it is to establish “general prepara-tion chemistry” prior to developing synthetic and/or biological applications. All the work described in this review would not be possible without a thorough establishing of the Stille cou-pling method for unsaturated aldehydes.16 Selective acylation

of the amino group in 54 was achieved in 58% yield by the reaction with DOTA-O-succinimidyl ester in the presence of Et3N in DMF. Finally, the allylic alcohol was oxidized by Dess-Martin periodinane in DMF/CH2Cl2 (3 : 10) to afford desired aldehyde 55a, which was immediately used for labeling in a DMF solution after size-partitioning fi ltration and removal of CH2Cl2. Moreover, the established route in Scheme 9 can also introduce various functionalities on the amino group in 54 to easily provide general probes. For example, coumarin and TAMRA fl uorescence probes 55b and 55c as well as other imaging reporters or biologically useful tags have been prepared.30,32

With probes 55a–c in hand, the reactivity toward lysine residues of various biomolecules, namely, somatostatin as a peptide, albumin, orosomucoid, and anti-GFP antibody as proteins were tested (Table 4).30 To our satisfaction, HPLC analysis indicated that the reaction of somatostatin (170 μg) with 100 equivalents of DOTA probe 55a at room tempera-ture for 30 min provided mono-DOTA labeled somatostatin in 96% yield (entry 1). Interestingly, of the two lysines in somatostatin, only the lysine not involved in receptor binding was modifi ed under these labeling conditions. In con-trast, conventional labeling conditions using the DOTA acti-vated ester required a much longer reaction time (over 24 h) compared to the conditions in entry 1, and employing the DOTA activated ester indiscriminantly modifi ed both lysine residues and the N-terminus, and the intact peptide was also recovered. Thus, the high reactivity of 55a enabled lysine-

+

1) Pd2(dba)3, P(2-furyl)3, LiCl

52

H2NNH

O

CO2Et

OH

BocHNNH

OSnBu3

Br

CO2Et

OTHP53

54

a), b) HN

NH

O

CO2Et

OH

O

R

OEt2N O

55b

55a-c

DMF, 110 oC, 73 %

2) 3M HCl, MeOH/H2O (1 : 1)rt, quant

O

Me2N

Me2N

CO2

NH

O

55c

R =

N

N

N

N

CO2HHO2C

HO2C

55a

Scheme 9. Preparation of lysine-labeling probes. a) DOTA-OSu, Et3N, DMF, RT, 58%; coumarin-OSu, CH2Cl2, RT, 68%; TAMRA-OSu, DMF/CH2Cl2 (1 : 1), RT, 74%; b) Dess-Martin periodinane, DMF/CH2Cl2 (3 : 10), RT or polystyrene-supported IBX, DMF/CH2Cl2 (1 : 1), RT.



azaelectrocyclization modifi cation to be completed in a very short time, which resulted in selective labeling of the more accessible lysine residue. Similarly, incubation of human serum albumin (100 μg) with 25 equivalents of probe 55a for just 10 min followed by quick size-partitioning gel-fi ltration suc-cessfully provided the modifi ed protein that incorporated fi ve DOTA units (entry 2). The present labeling by azaelectrocy-clization proceeded in nearly quantitative yield based on the lysine residues at both high (>10−3 M) and low (10−5 M − 10−6 M) concentrations of probe 55a. Consequently, the number of DOTA units introduced into a protein can be precisely controlled by adjusting the number of equivalents of 55a. For example, under 10 min reaction conditions, the use of 10 equivalents of 55a provided the labeled protein with two molecules of DOTA, whereas employing 500 equivalents introduced 20 DOTA molecules (entries 3 and 4).

Furthermore, the different effi ciencies between our elec-trocyclization and the conventional succinimidyl ester method could be evaluated by labeling various concentrations of HSA (10 min at RT and 10 equivalents of the reagents with respect to the protein). Although one and two molecules of DOTA were introduced into the HSA protein at concentrations of 10−6 m and 10−7 m, respectively, the conventional method

failed to label HSA even at a concentration of 10−5 m under the identical incubation conditions. Hence, our electrocycliza-tion protocol is 100–1000 times more effi cient.

To further display the performance of this labeling method, proteins, which are available in only small amounts (62 μg of glycoproteins: orosomucoid and asialoorosomu-coid), were labeled successfully and ∼2–3 units of DOTA were incorporated by incubating the respective protein with 10 equivalents of 55a for 30 min (entries 5 and 6). Additionally, paramagnetic metal Gd3+ could be chelated in the DOTA unit of 55a prior to peptide labeling (entry 7). This new process should allow valuable and/or unstable materials, which might decompose during metal-incorporation to DOTA, to be labeled.

The present electrocyclization protocol is also applicable to rapid fl uorescent labeling, 10–100 μg of albumin and anti-GFP mAb were labeled with coumarin 55b and TAMRA probes 55c, respectively (entries 8 and 9). As little as 2 μg of anti-GFP antibody (∼10 pmol) was successfully labeled by 55c in 30 min (entry 10); 20 equivalents of 55c preferentially labeled the more accessible Fc fragment with two fl uorophore molecules, while retaining its GFP recognition activity with 90% of the intact mAb.

Table 4. Labeling effi ciency by azaelectrocyclization.

55a : R = DOTA55b : R = coumarin

conditionsa

HN

NH

O

CO2Et

OH

O

R

55c : R = TAMRA-X

NH2

N

CO2EtHN

NH

O

O

R

N

CO2Et

entry probe biomolecule [μg] conc. of biomolecule [M] equiv. of probe t [min] number of labeled lysc

1b 55a somatostatin (170) 5.5 × 10−4 100 30 1 2 55a albumin (100) 5.0 × 10−5 25 10 5 3 55a albumin (100) 5.0 × 10−5 10 10 2 4 55a albumin (100) 5.0 × 10−5 500 10 20 5 55a orosomucoid (62) 4.5 × 10−6 10 30 2 6 55a asialo-orosomucoid (62) 4.5 × 10−6 10 30 3 7b 55a(Gd)d somatostatin (32) 3.6 × 10−4 31 30 1 8 55b albumin (120) 1.7 × 10−5 7 30 1e

9 55c albumin (12) 2.1 × 10−5 7 30 1f

10 55c anti-GFP mAb (2.0) 1.1 × 10−6 20 30 2f

a Unless otherwise noted, reactions were performed in 0.1 m phosphate buffer (pH = 7.4) at 24 °C. b in H2O. c Calculated by γ-counter of 57Co introduced to DOTA according to the method of Meares. d Gd was introduced by the reaction of with 0.1 M GdCl3 in H2O. e,f Estimated by emission spectra at 470 nm (coumarin) and 555 nm (TAMRA).




The entire process of the labeling procedure has been developed into a “labeling-kit”. Although we have prepared unsaturated aldehyde probes such as 55a and 55b through the oxidation of stable and therefore, “storable” alcohol precursors in a reaction fl ask (Scheme 9), such a specialized procedure can only be performed in synthetic organic laboratories. Hence, for a broader and more general application of labeling reagents to biology-directed research, an Eppendorf-tube pro-tocol was realized by oxidizing the alcohols with the polysty-rene-loaded IBX (iodobenzoic acid, Scheme 9), fi ltration, and subseqeunt treatment with biomolecule samples. Using a sim-plifi ed “kit” procedure, which is now available as “Stella+”,32 probe 55c as well as other unsaturated aldehyde probes can be conveniently and reproducibly prepared for labeling use in both organic chemistry and biology laboratories.

Thus, we successfully developed a new lysine-based label-ing of biomolecules on the basis of rapid 6π-azaelectrocycliza-tion. Without inactivating the biomolecules, both DOTA as a metal chelating agent (either for MRI, PET, or other radio-pharmaceutical purposes, such as SPECT with γ-emitters) as well as fl uorescent groups can be effi ciently and selectively introduced to lysine residues of samples at the few microgram level in either phosphate or Tris-HCl buffer solutions within 30 min at room temperature. The effi ciency of our rapid aza-electrocyclization protocol depends on the steric accessibility of the primary amino groups. Although reactions with internal lysines in the protein tertiary structures as well as the N-termi-nal amine (secondary amine) are very slow (>5 h at 24 oC),21a lysines at the protein’s surface react rapidly (10–30 min at 24 oC);6 therefore, labeling occurs preferentially at surface positions. Moreover, dihydropyridines as the electrocycliza-tion products, which retain cationic charges as those of the inherent lysines, should contribute to the retention of the protein activity.

Positron Emission Tomography (PET) of Biomolecules: First Visualization of Somatostatin Accumulation to Pancreas and Sialic Acid-Dependent Circulatory Residence of Glycoproteins

To verify the applicability of our new labeling method to in vivo imaging, the DOTA-somatostatin adduct obtained in the above labeling study was treated with 68Ga3+ as a β+ emitter for PET.28,30 Somatostatin analogs-based PET imaging and radio-therapy of tumors are of great interest because somatostatin receptors (SSTRs) are overexpressed in neuroendocrine tumors, i.e., gastroenteropancreatic, small cell lung, breast, and some-times tumors in the nervous systems.28,33 Nevertheless, soma-tostatin itself was believed to be extremely unstable in vivo (stability in rat serum is just 5 min), and previous imaging studies of somatostatin receptors were all for cases using metabolically stable analogs, such as [DOTA-DPhe1,Tyr3]-

octreotide.28 Thus, this probe was used for imaging of tumor-bearing mice, non-human primates, and human patients. However, a current study on microPET imaging of [68Ga]DOTA-somatostatin in rabbit successfully visualized the tracer accumulated in rabbit pancreas 2–4 h after injection (Fig. 7a). Given this new evidence of somatostatin stability in rabbit as well as its clear accumulation in the pancreas (somatostatin receptors are expressed in the pancreas, kidney, and gastroin-testinal tract), this study might provide an intriguing oportu-nity to use somatostatin itself as a diagnostic tracer of endocrine tumors.28,30

Furthermore, effi ciently labeled microPET images of [68Ga]DOTA-orosomucoid and asialoorosomucoid adducts (Table 4) were investigated in rabbits (Fig. 7b). Although FDG is a common monosaccharide PET tracer for cancer diagnosis, PET of oligosaccharides and glycoproteins is a totally unex-plored fi eld.28 The PET of these two proteins successfully visualized asialo-glycoprotein being cleared through the kidney and via the liver to the gallbladder faster than orosomucoid (Fig. 7b); thus, achieving the fi rst visualization of sialic acid-dependent circulatory residence of glycoproteins. The results agree well with the well-known hypothesis about the clearance of the proteins through the asialoglycoprotein receptor in the liver.34 That is, the sialic acid residue on the non-reducing end of the oligosaccharides contributes to the stability of glycopro-teins in the blood, whereas the sialidase-mediated production of asialoglycoproteins bearing a galactose residue at the non-reducing end of the oligosaccharides is responsible for the metabolic pathway. Thanks to the development of rapid 6π-azaelectrocyclization, these promising PET images suggest future uses for these glycoproteins in pharmacological and/or clinical applications.28,30

Site-Selective and Non-Destructive Protein Labeling via Azaelectrocyclization-Induced Cascade Reactions

Our new electrocyclization method precisely controls the introduction of DOTA or fl uorescence labels onto lysines in target proteins. By adjusting the reaction concentration, labels can be introduced at a level of one or two molecules per protein while retaining the activity of the biomolecules. However, when lysines critical for receptor binding are situated at the target protein’s most accessible site, a signifi cant decrease in activity may occur. For such cases, a site-selective and non-destructive method for protein labeling has been realized by directing the reactive groups (unsaturated aldehydes) to a specifi c site using a small-molecule ligand of the protein (Scheme 10).35 To demonstrate this, we selected Human Serum Albumin (HSA, MW = 66,000) as a target protein because HSA is readily available and it contains as many as 59 lysines, which makes it a very challenging target to test selective labeling using the azaelectrocyclization protocol. As outlined



in Scheme 10, fl uorescence probe 56 was applied; the fl uorescent reporter group, NBD (7-nitrobenz-2-oxa-1,3-diazole), was introduced at the C9 position of the probe, while the ligand of HSA, 7-diethylaminocoumarin, was attached at the C3 position by an ester linkage because 7-diethylaminocoumarin strongly binds to HSA through hydrophobic binding pockets.

Incubation of HSA with probe 56 selectively labeled Lys137 in subdomain IB situated close to the hydrophobic ligand-binding site via Schiff base formation (stage A) and subsequent rapid azaelectrocyclization (stage B) where the position of Lys was unambiguously determined by MALDI-TOF-MS peptide mapping analysis after the lysyl endopepti-dase treatments.35 Subsequently, the electrocyclization product, the 1,2-dihydropyridine derivative, could be readily aroma-tized to the pyridinium ion by autooxidation (stage C), which in turn, accelerated the hydrolysis of the ester linkage at the C3 position (stage D) to produce a zwitterion. This cascade process of oxidation-hydrolysis could successfully recover the ligand-binding site in the NBD-labeled protein. Therefore, owing to the susceptibility of the electrocyclization products

toward autooxidation, the ligand on the labeled protein could automatically be released under both neutral and physiological conditions. Hence, the results described herein have potential to selectively label target proteins in cell lysates as well as to label specifi c positions on cell surfaces.

Conclusion

As mentioned in this review, we have found the sub-stituents on 1-azatrienes affect the acceleration of 6π-azaelectrocyclization in the inactivation mechanism of the enzyme by the natural product derivatives. The new reac-tivity profi le of azaelectrocyclization has been thoroughly investigated by analog synthesis and reactivity studies, and rationalized by Frontier Orbital Theory. Quantitative azaelec-trocyclization can be realized within 10 min at room tempera-ture. One-pot pyridine syntheses, including the library-directed protocol on the solid supports and asymmetric “torque-selective” cyclization, have been developed and success-fully applied to natural products syntheses, i.e., pyridinium

Fig. 7. a) PET imaging of 68Ga-DOTA-somatostatin in rabbit (axial view, overlapped with CT). Images represent time-dependent changes of accumulation in the liver (large arrow), kidney (small arrow), and pancreas (arrowhead) after injection. b) Dynamic microPET images of 68Ga-DOTA-glycoproteins in rabbits (axial view, overlapped with CT). Time course of accumulation of 68Ga-DOTA-orosomucoid (upper) and 68Ga-DOTA-asialoorosomucoid (lower) in some peripheral organs (axial views).




containing retinoid metabolites or substituted-piperidine alkaloids. The knowledge gained during our synthetic investi-gations on azaelectrocyclization reactivity was used to under-stand biofunctions in vivo such as those of peptides and glycoproteins using PET imaging. The advantage of rapid azaelectrocyclization-based labeling under the mild conditions is that the native functions are not affected. Hence, the authors believe that the “triangle” shown in Fig. 1, a novel category in chemical biology, and more specifi cally, in synthetic biology, constitutes unique and valuable research at the chemistry/biology interface. Thus, the “triangle” may lead to the under-standing of new reactions and reactivities not only as crucial functions in a important biological mechanism, but also as an approach to synthetic and biology-directed research.

The authors are grateful to all the co-authors of the references cited in this review. We especially thank Professor Kiyoshi Ikeda and Dr. Shinobu Fujii at Osaka University of Pharmaceutical Sciences, Japan, for the biochemical experiments. We also acknowledge Professor Yasuyoshi Watanabe at RIKEN, Japan, for directing the molecular imaging research. The work cited in this review were in part fi nancially supported by Grants-in-Aid for Scientifi c Research No. 19681024 and 19651095 from the Japan Society for

the Promotion of Science, Grant-in-Aid for Science Research on Priority Areas 16073222 from the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Molecular Imaging Research Program from MEXT, Matching Fund Subsidy for a Private University, Collaborative Development of Innovative Seeds from Japan Science and Technology Agency (JST), New Enegry and Industrial Technology Development Organization (NEDO, project ID: 07A01014a), and a Research Grant from Yamada Science Foundation.

REFERENCES

[1] Dennis, E. A. The Enzymes, Vol. 16 (Ed.: Boyer, P. D.) Academic Press, NewYork, 1983, p. 307.

[2] Pace-Asciak, C. R.; Smith, W. L. The Enzymes, Vol. 16 (Ed.: Boyer, P. D.) Academic Press, NewYork, 1983, p. 543.

[3] Noel, J. P.; Bingman, C. A.; Deng, T.; Dupureur, C. M.; Hamilton, K. J.; Jiang, R.-T.; Kwak, J.-G.; Sekharudu, C.; Sundaralingam, M.; Tsai, M.-D. Biochemistry 1991, 30, 11801.

[4] (a) Katsumura, S.; Han, Q.; Kadono, H.; Fujiwara, S.; Isoe, S.; Fujii, S.; Nishimura, H.; Ikeda, K. Bioorg. Med. Chem. Lett. 1992, 2, 1263; (b) Katsumura, S.; Hart, Q.; Fujiwara, S.; Isoe,

rapid 6π-azaelectrocyclization

O

N

HSA

NH2

coumarinbinding site

NH2

rt, < 30 min

Schiff base formation

autooxidation

hydrolysis

Lys137

O

OOO

NH

O

OO

Et2N

HN

O

NH N O

N

NO2

HO

O OO

HN

O

OO NEt2

NH

OHN

NO

N

O2NH

HNO

NH N O

N

NO2

OOHN

O

OO

Et2N

O

N

HNO

NH N ON

NO2

O

recovering binding site

56

O

N

HN O

NH N ON

NO2

O

ONH

O

O OEt2N

(stage A)

9

3

(stage B)

(stage C)

(stage D)

Scheme 10. Site-selective and non-destructive labeling by azaelectrocyclization.



S.; Nishimura, H.; Inoue, S.; Ikeda, K. Bioorg. Med. Chem. Lett. 1992, 2, 1267; (c) Fujii, S.; Tahara, Y.; Toyomoto, M.; Hada, S.; Nishimura, H.; Inoue, S.; Ikeda, K.; Inagaki, Y.; Katsumura, S.; Samejima, Y.; Omoil-Satoh, T.; Takasaki, C.; Hayashi, K. Biochem. J. 1995, 308, 297.

[5] Potts, B. C. M.; Faulkner, D. J.; Jacobs, R. S. J. Nat. Prods. 1992, 55, 1701.

[6] (a) Tanaka, K.; Mori, H.; Fujii, S.; Itagaki, Y.; Katsumura, S. Tetrahedron Lett. 1998, 39, 1185; (b) Tanaka, K.; Kamatani, M.; Mori, H.; Fujii, S.; Ikeda, K.; Hisada, M.; Itagaki, Y.; Katsumura, S. Tetrahedron 1999, 55, 1657; (c) Tanaka, K.; Katsumura, S. J. Synth. Org. Chem. Jpn. 1999, 57, 876.

[7] Katsumura, S.; Hori, K.; Fujiwara, S.; Isoe, S. Tetrahedron Lett. 1985, 26, 4625.

[8] de Lera, A. R.; Reischl, W.; Okamura, W. H. J. Am. Chem. Soc. 1989, 111, 4051.

[9] Marvell, E. N. Thermal Electrocyclic Reactions; Academic Press: New York, 1980.

[10] (a) Marvell, E. N.; Caple, G.; Schatz, B.; Pippin, W. Tetrahe-dron 1973, 29, 3781; (b) Marvell, E. N.; Caple, G.; Delphey, C.; Platt, J.; Polston, N.; Tashiro, J. Tetrahedron 1973, 29, 3797.

[11] Spangler, C. W.; Jondahl, T. P.; Spangler, B. J. Org. Chem. 1973, 38, 2478.

[12] (a) Palenzuela, J. A.; Elnagar, H. Y.; Okamura, W. H. J. Am. Chem. Soc. 1989, 111, 1770; (b) Zhu, G.-D.; Okamura, W. H. Chem. Rev. 1995, 95, 1877; (c) Okamura, W. H.; de Lera, A. R. Comprehensive Organic Synthesis Vol. 5,(Eds. Trost, B. M., Fleming, I.), Paquette, L. A., Volume Ed.; Pergamon Press, London, 1991, Chapter 6.2, pp 699–750.

[13] Schiess, P.; Chia, H. L.; Ringele, P. Tetrahedron Lett. 1972, 313.

[14] (a) Cheng, Y.-S.; Lupo, A. T., Jr.; Fowler, F. W. J. Am. Chem. Soc. 1983, 105, 7696; (b) Wyle, M. J.; Fowler, F. W. J. Org. Chem. 1984, 49, 4025.

[15] Maynard, D. F.; Okamura, W. H. J. Org. Chem. 1995, 60, 1763.

[16] Tanaka, K.; Mori, H.; Yamamoto, M.; Katsumura, S. J. Org. Chem. 2001, 66, 3099.

[17] Tso, M. O. M. Invest. Ophthalmol. Visual Sci. 1989, 30, 2430.[18] Sakai, N.; Decatur, J.; Nakanishi, K. J. Am. Chem. Soc. 1996,

118, 1559.[19] Tanaka, K.; Katsumura, S. Org. Lett. 2000, 2, 373.[20] (a) Kobayashi, T.; Hatano, S; Tsuchikawa, H.; Katsumura, S.

Tetrahedron Lett. 2008, 49, 4349; (b) Sakaguchi, T.; Kobayashi, T.; Hatano, S.; Tsuchikawa, H.; Fukase, K.; Tanaka, K.; Katsumura, S. Chem. Asian J. 2009, 4, 1573.

[21] (a) Tanaka, K.; Katsumura, S. J. Am. Chem. Soc. 2002, 124,

9660; (b) Tanaka, K.; Kobayashi, T.; Mori, H.; Katsumura, S. J. Org. Chem. 2004, 69, 5906; (c) Tanaka, K.; Katsumura, S. J. Synth. Org. Chem. Jpn. 2005, 63, 697.

[22] (a) Hsung, R. P.; Wei, L.-L.; Sklenicka, H. M.; Douglas, C. J.; McLaughlin, M. J.; Mulder, J. A.; Yao, L. J. Org. Lett. 1999, 1, 509; (b) Sklenicka, H. M.; Hsung, R. P.; Wei, L.-L.; McLaughlin, M. J.; Gerasyuto, A. I.; Degen, S. J. Org. Lett. 2000, 2, 1161; (c) Sklenicka, H. M.; Hsung, R. P.; McLaugh-lin, M. J.; Wei, L.-l.; Gerasyuto, A. I.; Brennessel, W. B. J. Am. Chem. Soc. 2002, 124, 10435; (d) McLaughlin, M. J.; Hsung, R. P.; Cole, K. P.; Hahn, J. M.; Wang, J. Org. Lett. 2002, 4, 2017; (e) Luo, S.; Zifi csak, C. A.; Hsung, R. P. Org. Lett. 2003, 5, 4709.

[23] Kobayashi, T.; Tanaka, K.; Miwa, J.; Katsumura, S. Tetrahe-dron Asymmetry 2004, 15, 185.

[24] Bosch, J.; Bonjoch, J. Pentacyclic Strychnos Indole Alkaloids. In Studies in Natural Products Chemistry Vol. 1 (Ed. Atta-ur-Rahman) Elsevier, Amsterdam, 1988, pp 31–88.

[25] Grierson, D. S.; Harris, M.; Husson, H.-P. Tetrahedron 1983, 39, 3683.

[26] (a) Kobayashi, T.; Nakashima, M.; Hakogi, T.; Tanaka, K.; Katsumura, S. Org. Lett. 2006, 8, 3809; (b) Kobayashi, T.; Hasegawa, F.; Tanaka, K.; Katsumura, S. Org. Lett. 2006, 8, 3813.

[27] Gambhir, S. S. Nat. Rev. Cancer 2002, 2, 683.[28] (a) Tanaka, K.; Fukase, K. Org. Biomol. Chem. 2008, 6, 815;

(b) Tanaka, K.; Fukase, K. Mini-Rev. Org. Chem. 2008, 5, 153–162.

[29] Lewis, M. R.; Kao, J. Y.; Anderson, A.-L. J.; Shively, J. E.; Raubitschek, A. Bioconjugate Chem. 2001, 12, 320.

[30] (a) Tanaka, K.; Masuyama, T.; Hasegawa, K.; Tahara, T.; Mizuma, H.; Wada, Y.; Watanabe, Y.; Fukase, K. Angew. Chem. Int. Ed. 2008, 47, 102; (b) Tanaka, K.; Masuyama, T.; Minami, K.; Fujii, Y.; Hasegawa, K.; Tahara, T.; Mizuma, H.; Wada, Y.; Watanabe, Y.; Fukase, K. Peptide Science 2007, 91.

[31] (a) Chen, I.; Ting, A. Y. Curr. Opin. Biotech. 2005, 16, 35; (b) Miller L. W.; Cornish, V. W. Curr. Opin. Chem. Biol. 2005, 9, 56; (c) Takaoka, Y.; Tsutsumi, H.; Kasagi, N.; Nakata, E.; Hamachi, I. J. Am. Chem. Soc. 2006, 128, 3273.

[32] Labeling kit “STELLA+” is available from Kishida Chemical Co., Ltd., http://www.kishida.co.jp/.

[33] Maecke, H. R.; Hofmann, M.; Haberkorn, U. J. Nucl. Med. 2005, 46, 172S.

[34] Morell, A. G.; Irvine, R. A.; Sternlieb, I.; Scheinberg, I. H.; Ashwell, G. J. Biol. Chem. 1968, 243, 155.

[35] Tanaka, K.; Fujii, Y.; Fukase, K. ChemBioChem 2008, 9, 2392.

New strategy in synthetic biology: from enzyme inhibition and natural products synthesis to PET...

Documents

Transcript of New strategy in synthetic biology: from enzyme inhibition and natural products synthesis to PET...