Facile and rapid access to linear and truncated microcystin analogues for the implementation of immunoassays

View Article Online / Journal Homepage / Table of Contents for this issue

PAPER

www.rsc.org/obc | Organic & Biomolecular Chemistry

Facile and rapid access to linear and truncated microcystin analogues for the

implementation of immunoassays†

G. Clave´,^a,bC. Ronco,^a,bH. Boutal,^cN. Kreich,^cH. Volland,^cX. Franck,^aA. Romieu*^a,band P.-Y. Renard*^a,b,d

Received 28th September 2009, Accepted 31st October 2009

First published as an Advance Article on the web 8th December 2009

DOI: 10.1039/b920193a

A series of simpliﬁed microcystin-LR analogues based on Adda [(2S,3S,8S,9S,4E,6E)-3-amino-9-

methoxy-2,6,8-trimethyl-10-phenyldecadienoic acid] or its corresponding aldol precursor linked to a

polypeptide moiety have been synthesised and assessed for their binding afﬁnity by the monoclonal

antibody mAb MC159, an anti-microcystin-LR mAb recently selected by us for the detection of

microcystins through various immunoassay formats. Some modiﬁcations have been brought to the

enantiospeciﬁc synthesis of N-Boc-Adda developed by Pearson et al. (Org. Lett., 2000, 2, 2901) which

enabled us to access in an economical and time-saving manner a small library of MC-LR linear

analogues. Among which Adda was chosen to synthesise, as an illustrative example, a ﬂuorescent probe

derived from this b-amino acid. This probe was subsequently solid-phase immobilised by means of

oxime ligation in order to lead to biochips suitable for microcystin detection through the SPIT-FRI

method.

Introduction

Microcystins (MCs) are well-known toxic cyclic heptapeptides

produced by the cyanobacterial genera Microcystis, Oscillatoria,

Anabaena and Nostoc found in fresh and brackish waters.¹These

cyanotoxins show potent hepatotoxicity and tumor-promoting

activity through inhibition of protein phosphatases PP-1 and

PP-2A. Thus, in countries where freshwater (i.e., drinking and

surface water) is prone to contamination by cyanobacterial

blooms, MCs have been responsible for a variety of adverse

effects on animal and human health.²The chemical structure

of MCs comprises two variable L-amino acids X and Y, three

D-amino acids (alanine, methylaspartic acid and glutamic acid)

and two unusual amino acids, N-methyldehydroalanine (Mdha)

and (2S,3S,8S,9S,4E,6E)-3-amino-9-methoxy-2,6,8-trimethyl-

10-phenyldecadienoic acid (Adda) which play an important role

in their biological activity.

MCs differ primarily in the two L-amino acids X and Y (Fig. 1),

and over 80 variants have been reported to date. Among the vari-

ants fully identiﬁed, microcystin-LR (MC-LR, X = Leu, Y = Arg),

also known as the “fast death factor”, is among the most frequent

and toxic microcystin congeners.³Thus, fast quantitative detection

of this toxin is essential to ensure water quality, and therefore

Fig. 1 General structure of microcystins (X and Y: variable L-amino

acids).

human health. Currently, several detection methods based on

analytical separation methods (i.e., reversed-phase HPLC (RP-

HPLC) coupled with mass spectrometry and/or UV detection),⁴

bio-assays,⁵biochemical assays⁶and immunoassays (i.e., direct

competitive ELISA using monoclonal or polyclonal antibodies

(mAbs or pAbs))⁷are in use. Recently, new technologies employing

recombinant antibodies⁸and molecularly imprinted polymers⁹

have been exploited to develop bioassays and biosensors for MCs.

All these methods have their own advantages and drawbacks de-

pending on their sensitivity, speciﬁcity, cross-reactivity (for other

MCs in the context of immunoassays), cost and implementation

procedure but most of them are slow, technically demanding and

often require extensive sample processing prior to analysis.¹⁰To

circumvent this latter deﬁciency, Kim et al. have developed an

ultra-rapid one-step ﬂuorescence immunochromatic assay which

was used in portable biosensor systems suitable for checking

the presence of MCs in drinking or surface water.^11aWe further

developed such an immunochromatic assay yet further improved

by the use of immunoliposomes. Interestingly, samples can be

analysed for MCs at the test site in real-time within 15 min

and in a sensitive and reproducible manner. The weakness of

this method is that the antibody-immobilised and microcystin-

immobilised systems exhibit poor stability over time due to

the non-covalent immobilisation of these peptides and proteins

^aEquipe de Chimie Bio-Organique, COBRA-CNRS UMR 6014 & FR 3038,

rue Lucien Tesnie`re, 76131 Mont-Saint-Aignan, France. E-mail: pierre-

yves.renard@univ-rouen.fr, anthony.romieu@univ-rouen.fr; Fax: +33 2-35-

52-29-71; Tel: +33 2-35-52-24-14 (or 24-15)

^bUniversite´ de Rouen, Place Emile Blondel, 76821 Mont-Saint-Aignan,

France

^ciBiTecS, Laboratoire d’Etudes et de Recherches en Immuno-analyse,

Commissariat a` l’Energie Atomique, Gif-sur-Yvette, F-91191, France

^dInstitut Universitaire de France, 103 Boulevard Saint-Michel, 75005, Paris,

France

† Electronic supplementary information (ESI) available: ESI mass spectra

of compounds 19, 20 and related intermediate conjugates. See DOI:

10.1039/b920193a

676 | Org. Biomol. Chem., 2010, 8, 676–690

This journal is

The Royal Society of Chemistry 2010

©

View Article Online

onto the nitrocellulose membrane used as the bioassay support.

Recently, we have reported the Solid-Phase Immobilised Tripod for

Fluorescent Renewable Immunoassay (SPIT-FRI), a new promis-

ing method for sensitive detection and quantiﬁcation of various

analytes that possesses the same advantages as described above

for the immunochromatic assay.¹²One of the main advantages

is that the SPIT-FRI assay has been conceived in order to be

used in continuous ﬂow, with a localised analysis point, and thus

could easily be multiplexed. With that goal in mind to improve

the stability of the immunosensor, we have developed a novel

heterotrifunctional peptide-based cross-linking reagent aimed at

the covalent immobilisation of the ﬂuorescent “tripod” which

supports an epitope similar or close to the target analyte and

recognised by a speciﬁc antibody, on functionalised aldehyde

surfaces via a highly chemoselective oxime ligation.¹³This method

was validated using a clinically relevant neuropeptide, substance

P, as a model of target analyte to be measured. At present, we want

to apply SPIT-FRI to the most challenging and important targets

such as MC-LR. Since implementation of such an immunoassay

requires the production and selection of mAbs directed against this

microcystin, we have recently started a research program devoted

to the synthesis of linear and truncated MC-LR analogues that

possess the unusual side-chain of Adda. Indeed, in the course

of the evaluation of the SPIT-FRI immunoassay FRET-based

system (Fig. 2), it is essential to have access to a small library of

MC-LR structural analogues which are recognised by mAbs with a

representative range of cross-reactivites (10–80%). Since the SPIT-

FRI assay is based on the efﬁcient immobilisation of the mAb on

the solid surface through strong binding onto the toxin analogue

covalently grafted on the solid-phase, this analogue should display

a high afﬁnity towards mAb. Yet, if the toxin is present in the

medium, the ﬂuorescent signal in uncovered by the competitive

displacement of the mAb from the solid surface through stronger

binding of the mAb onto the toxin itself, the differential binding

of the mAb (cross-reactivity) between the toxin and its analogue

is thus one of the key features of the SPIT-FRI system. In order to

be able to evaluate the SPIT-FRI with a wide range of analogues,

ideally, each compound of this small library must be synthesised in

reasonable amounts (~10 mg), rapidly and inexpensively. However,

the structural complexity of the Adda scaffold makes this objective

not trivial despite the number of syntheses of this unusual

b-amino acid already reported in the literature over the past

decade.¹⁴

Fig. 2 Different steps of the SPIT-FRI procedure.

Results and discussion

In this paper, we describe an efﬁcient asymmetric synthesis

of Adda by a stepwise linear approach allowing a rapid and

straightforward addition of polypeptides on the Adda scaffold;

some practical improvements have been brought to the 13-step

synthesis reported by Pearson et al.¹⁵aimed at signiﬁcantly

reducing reagent costs and synthesis time. The corresponding

N-Boc-protected derivative N-Boc-Adda and aldol precursor 10

were synthesised and used to rapidly prepare four additional

linear and truncated MC-LR analogues through standard peptide-

coupling reactions or trans-amidiﬁcation reactions. In addition

to these synthetic aspects, cross-reactivity assays towards mAb

anti-MC MC159 binding afﬁnity have enabled us to select one

of these MC-LR analogues which was ﬂuorescently labelled

and subsequently immobilised on an aldehyde-functionalised

glass surface. Indeed, such peptide microarrays are required for

detecting MC-LR by SPIT-FRI.

Structural features of targeted MC-LR analogues

We recently described a procedure that allowed us to produce, iso-

late and characterise 52 different mAbs anti-MC.¹⁶The best results

were obtained with mAb MC159 that recognises different variants

of MCs indicating that the Adda unit plays an essential role in the

recognition by this latter antibody. We decided to synthesise linear

MC-LR analogues derived from the “north-western” part of this

macrocycle which according to our preliminary screening results

deﬁnes the minimum epitope required for antibody recognition.

Thus, these Adda-containing peptides should be readily accessible

through a convergent synthetic strategy based on solution-phase

peptide couplings or related reactions and involving suitable

protected derivatives of D-glutamic acid, N-methylalanine (i.e.,

the saturated analogue of Mdha found in MCs) and Adda.

This journal is

The Royal Society of Chemistry 2010

Org. Biomol. Chem., 2010, 8, 676–690 | 677

©

View Article Online

Synthesis of N-Boc-Adda

10 by free N-terminal peptides might be considered (vide infra).

Such asymmetric aldol additions were achieved using chlorotita-

nium enolates of N-acyl thiazolidinethiones under the conditions

reported by Crimmins and She (i.e., combination of TiCl₄, DIEA

and NMP),¹⁹and proceeded with high diastereoselectivity for the

“Evans syn” product. Furthermore, titanium chloride is 60 times

less expensive than dibutylboron triﬂate (Bu₂BOTf), the most

popular reagent used to generate the boron enolates involved in

stereoselective aldol condensations. Thus, such reactions could be

carried out on a large scale (100 mmol) without causing prohibitive

costs. N-Propionyl-thiazolidin-2-thiones 1 and 9 were prepared by

cyclisation of the corresponding chiral b-aminoalcohols with car-

bon disulﬁde, following a literature procedure,²⁰and subsequent

N-acylation with propionyl chloride. N-Boc-Adda was synthesised

through the 13-step synthetic procedure depicted in Scheme 1.

(R)-N-Propionyl-4-phenylthiazolidine-2-thione 1 (prepared from

D-phenylglycine) was allowed to react with phenylacetaldehyde in

the presence of TiCl₄, DIEA and NMP, producing syn aldol 2 in

quantitative yield with complete control of diastereoselectivity.

Puriﬁcation by simple liquid–liquid extractions provided this

compound with satisfactory purity. Thereafter, thanks to the

good leaving group ability of thiazolidinethione moiety, 2 was

readily converted into the corresponding Weinreb amide under

During the past 20 years, signiﬁcant efforts have been devoted to

the asymmetric synthesis of both Adda (as a single stereoisomer)

and Adda conjugates to prepare MCs, MC analogues and Adda

related speciﬁc protein phosphatase inhibitors. Since one of

the current research interests of our group is the development

of new synthetic methodologies involving stereoselective aldol

reactions,¹⁷we decided to prepare N-Boc-Adda using the high-

yield and convenient synthetic route developed by Pearson et al.

The published protocols are efﬁcient, yet, taking into account

the experienced fragility of Adda derivatives, with both goals in

mind to have milder derivatisation conditions, and cheaper access

to signiﬁcant quantities of N-Boc-Adda we proposed practical

modiﬁcations. Those modiﬁcations are especially focused on the

two Evans aldol reactions chosen to elaborate the function-

alised chiral side-chain and to set the four stereocenters of this

b-amino acid. Instead of using a chiral auxiliary belonging to the

family of oxazolidinones, we preferred thiazolidinethione deriva-

tives, which display a better leaving group ability which greatly

facilitates their removal and/or nucleophilic displacement under

mild conditions.^17,18Consequently, direct access to original MC

analogues by trans-amidiﬁcation of fragile Adda aldol precursor

Scheme

1

Synthetic route to N-Boc-Adda. Reagents and conditions: i) TiCl₄, DIEA, NMP, phenylacetaldehyde, CH₂C₂, -78 ^◦C;

ii) N,O-dimethylhydroxylamine hydrochloride, imidazole, CH₂Cl₂; iii) NaH, MeI, THF, -78 ^◦C; iv) DIBAL-H, THF, -78 ^◦C; v) (carbethoxy-

ethylidene)triphenylphosphorane, toluene, reﬂux; vi) DIBAL-H, THF, -78 ^◦C; vii) MnO₂, toluene, 50–55 ^◦C; viii) (carbethoxymethyli-

dene)triphenylphosphorane, toluene, reﬂux; ix) DIBAL-H, THF, -78 ^◦C; x) MnO₂, toluene, 50–55 ^◦C; xi) TiCl₄, DIEA, NMP, CH₂Cl₂, -78 ^◦C;

xii) N-Boc hydroxylamine, imidazole, CH₂Cl₂, rt. xiii) Ph₃P, DIAD, THF, -78 ^◦C then sodium naphthalide, THF, -78 ^◦C.

678 | Org. Biomol. Chem., 2010, 8, 676–690

This journal is

The Royal Society of Chemistry 2010

©

View Article Online

mild conditions by treatment with N,O-dimethylhydroxylamine

(HCl salt) in the presence of imidazole and DMAP. Therefore,

the enantiopure thiazolidine-2-thione moiety, contrary to the

corresponding oxazolidinone derivative used by Pearson et al.,

can be considered both as a chiral auxiliary and as an activated

amide (also named “activated ester” in the terminology of peptide

synthesis), preventing both the use of drastic conditions for its

removal and the subsequent activation of the resulting carboxylic

acid. Subsequent methylation of the free secondary alcohol

performed with methyl iodide and NaH, and DIBAL-H reduction

of Weinreb amide moiety led to the corresponding aldehyde which

was engaged in a ﬁrst Wittig reaction with (carbethoxyethyli-

dene)triphenylphosphorane to provide oleﬁn 4. This latter a,b-

unsaturated ester was then subjected to a two-step reduction–

oxidation sequence to give the aldehyde 6. The reduction step with

DIBAL-H was achieved under the same conditions as used for the

Weinreb amide. MnO₂-mediated oxidation was optimised because

in our hands, the reaction conditions reported by Pearson et al.

(i.e., 0 ^◦C in CH₂Cl₂) failed to yield 6. Due to the lack of informa-

tion about the source of MnO₂, we screened several batches from

different providers toﬁndthe mostactive reagent. In our hands, the

best results were obtained using MnO₂from Alfa Aesar (activated,

tech., Mn 58% min) in suspension in dry toluene at 50–55 ^◦C for

2 h, to obtain the targeted aldehyde 6 in a quantitative yield.

These conditions prevented its subsequent partial polymerisation

observed with other sources of activated MnO₂, or at higher

temperatures or with longer reaction times when less activated

MnO₂sources were used. A second Wittig reaction between 6 and

(carbethoxymethylidene)triphenylphosphorane provided ester 7,

completing the diene system of Adda. A second reduction–

oxidation sequence performed under the same conditions as

stated above, gave aldehyde 8. Aldol reaction between 8 and (S)-

N-propionyl-4-benzyl-thiazolidine-2-thione 9 was again achieved

according to the Crimmins protocol to provide syn aldol 10.

For this second asymmetric aldol reaction, the chiral auxiliary

derived from L-phenylalanine was preferred because this latter

amino acid was two times less expensive than L-phenylglycine.

Thereafter, the thiazolidinethione moiety of 10 was displaced by

N-Boc hydroxylamine to give N-protected hydroxamic acid 11 in

70% overall yield (for the last four steps). This latter compound

underwent an intramolecular Mitsunobu reaction which resulted

in the formation of an isoxazolidin-5-one intermediate whose

N–O bond was reductively cleaved with sodium naphthalide to

give N-Boc-Adda. Puriﬁcation of this N-protected amino acid was

achieved by silica gel column chromatography. All spectroscopic

data, especially NMR and mass spectrometry were in agreement

with the structure assigned. It is important to note that the 5-step

synthetic sequence used for the conversion of ester 7 to N-Boc-

Adda must be performed within a short period of time because we

observed signiﬁcant degradation of the intermediate compounds

8, 10 and 11, even after few months of storage at -20 ^◦C

when kept under an argon atmosphere. Surprisingly, the observed

poor stability of these latter intermediates was not reported by

Pearson et al. To sum up, the present synthetic approach has

enabled us to prepare N-Boc-Adda on a 100 mg scale with a

reasonable 5% overall yield; practical modiﬁcations brought to

the Pearson’s synthesis allowed this total synthesis to be achieved

within two weeks (if chiral auxiliaries are already available), in a

perfectly reproducible manner whatever the experimental skill of

the chemist (from undergraduate student to post-doctoral fellow).

Yields given were obtained on a signiﬁcant larger scale than the

reactions conditions described by Pearson et al., and have been

counterchecked by two independent chemists (i.e., GC and CR).

Moreover, we faced unreported stability/puriﬁcation issues, which

can also explain the signiﬁcantly lower global yield.

Synthesis and antibody cross-reactivity of MC-LR analogues

As previously mentioned, we have recently produced and charac-

terised monoclonal antibodies directed against MC-LR. Among

the 52 mAb obtained, mAb MC159 was selected due to its

broad speciﬁcity to develop a sensitive competitive enzyme

immunoassay (EIA) aimed at quantifying MCs and nodularins

in water samples and an innovative immunoliposome-based im-

munochromatographic assay.^11b,16To simplify the implementation

of SPIT-FRI, which requires the production step of “tripod” in

substantial amount, structurally simpler analogues of the whole

targeted toxins are desirable. Moreover, since one of the questions

remaining to be addressed in the SPIT-FRI assay is the required

cross-reactivity for binding between the analogue grafted on

the tripod and the toxin, in order to have an efﬁcient speciﬁc

displacement of the mAbs by the toxin itself, a small library

of easily accessible analogues of the toxins was required. Thus,

we ﬁrst planned the synthesis of a tripeptide derived from Adda,

D-glutamic acid and N-methylalanine. In order to widen the size

of the library, without introducing too many structural differences

in the global structures, before starting the total synthesis of this

Adda conjugate according to standard peptide synthesis protocols,

we explored an original coupling strategy based on the aminolysis

of aldol 10 (Scheme 2). Indeed, this intermediate in the synthesis of

Adda possesses almost all structural attributes of the unusual side-

chain of this b-amino acid (except for the stereochemistry of C-3

which is inverted and for the nature of substituent at C-3: the amino

group is replaced by a hydroxyl one). This intermediate can also be

used as an activated ester as we previously described, thanks to the

good leaving group ability of the thiazolidine-2-thione moiety. This

heterocyclic system has been successfully used in solution-phase

peptide synthesis in the early 80’s,²¹but since, few applications of

this peptide coupling additive have been reported.²²To demon-

strate its usefulness and to obtain rapidly two simpliﬁed MC-LR

analogues 14 and 15, we have investigated the trans-amidiﬁcation

of aldol 10 by dipeptides H-D-Glu(MeAla-OMe)-OMe 12 and H-

D-Glu(Ala-OMe)-OMe 13. These latter dipeptide methyl diesters

were readily synthesised by coupling between the commercial

amino acids Boc-D-Glu-OMe and H-MeAla-OMe (or H-Ala-

OMe) and subsequent Boc removal by acidolysis with TFA. To

overcome the coupling difﬁculties generally encountered with

N-methylated amino acids, PyBrOP/DIEA activation was used

for the preparation of 14.²³The more conventional peptide

coupling reagent BOP was effective for synthesising 15.²⁴By

analogy with the synthesis of intermediates 3 and 11 involved in the

preparation of N-Boc-Adda, trans-amidiﬁcation of aldol 10 with

the TFA salt of primary amine 14 (or 15) was carried out in THF at

room temperature, in the presence of imidazole (2 equiv.) and a cat-

alytic amount of DMAP. Contrary to oxazolidinone derivatives,

the reaction does not require the use of a Lewis acid such as AlMe₃

which may have deleterious effects on the structure of peptide

partner used. Puriﬁcation by silica gel column chromatography

This journal is

The Royal Society of Chemistry 2010

Org. Biomol. Chem., 2010, 8, 676–690 | 679

©

View Article Online

Scheme 2 Synthetic reactions used for the preparation of acyclic microcystin analogues 14–16 from aldol 10. Reagents and conditions:

i) H-D-Glu(MeAla-OMe)-OMe 12, imidazole, DMAP, THF, rt; ii) aq. LiOH, CH₃OH, rt; iii) H-D-Glu(Ala-OMe)-OMe 13, imidazole, DMAP, THF, rt;

iv) aq. LiOH, CH₃OH, rt.

has enabled us to isolate the resulting fully-protected tripeptides

in moderate yields and proved smooth enough so that we were

also able to recover the released chiral auxiliary. As the ﬁnal step,

the methyl esters were removed by short treatment with LiOH

in H₂O–CH₃OH to give the target microcystin analogues 14 and

15 which were isolated in their pure form by semi-preparative RP-

HPLC. Their structures were conﬁrmed by detailed measurements,

including ESI mass spectrometry and NMR analyses. However,

if 15 proved to be present as a single diastereomer, this was

not the case with its N-methylated analogue 14. NMR analyses

revealed that complete epimerisation of N-methylalanine residue

had taken place giving rise to a 1 : 1 mixture of diastereomers for

analogue 14. In our case, the signal splitting observed in proton

NMR spectrum of 14 cannot be assigned to a mixture of two

amide rotamers because such spectral behaviour did not disappear

on warming and was not observed with the parent methyl

diester derived from the same peptide scaffold. Furthermore, the

racemisation of N-substituted-N-methylamino acid methyl esters

during saponiﬁcation with LiOH has been already reported in

the literature.²⁵All our attempts to separate each one of the

diastereomers by semi-preparative RP-HPLC failed, and using

the mixture in a competitive EIA experiment, a signiﬁcant cross-

reactivity of around 25% (IC₅₀219 pM) for binding to mAb

MC159 as compared with MC-LR was obtained. This result

clearly shows that the Adda side-chain plays a pivotal role in

the recognition of MCs by mAb MC159. This structure/cross-

reactivity relationship study was extended to microcystin analogue

15 and b-hydroxy acid 16, this latter compound being obtained by

direct saponiﬁcation of aldol 10. Monoclonal antibody MC159

recognises the greatly simpliﬁed MC analogues 15 and 16 with

~10% of cross-reactivity for binding afﬁnity as compared to

MC-LR (12% for 15 and 13% for 16, IC₅₀614 pM and 409 pM

respectively). These results demonstrate that the methyl residue

on the amino group of the alanine residue is important, but not

pivotal, for the recognition of our analogues by the selected mAb.

Conversely, the presence of the D-glutamic acid residue in the

structure has no inﬂuence on the binding afﬁnities of the selected

analogues for the mAb MC159. These three acyclic analogues

of MCs, which could be readily obtained from the N-Boc-Adda

synthesis intermediate 10 in the yield range 30–70% (for one or

two steps), offer a ﬁrst small library of MC analogues. They will be

further used to evaluate the SPIT-FRI scope and limitations, and

could offer an economical alternative to Adda-based analogues of

MCs. Next, we turned our attention to the synthesis and cross-

reactivities studies of Adda peptidyl conjugates (Scheme 3) in

order to widen the scope of cross-reactivities, but also since 14 and

15 do not perfectly fulﬁl all requirements of an useful epitope for

the SPIT-FRI technology. Indeed, the presence of two carboxylic

acid functions within their structure and availability of 14 only as

a mixture of diastereomers makes more tricky the chemoselective

conjugations to a ﬂuorescent label and functionalised solid-

phases in an efﬁcient manner (i.e., good yield and high degree

of purity for the resulting tri-component conjugate). To avoid

epimerisation of N-methylalanine residue occurring during the

ﬁnal ester deprotection step, tripeptide 17 derived from Adda,

D-glutamic acid and alanine was preferred to Adda-containing

tripeptide H-Adda-iso-D-Glu(MeAla)-OH initially considered.

Furthermore, we thought that the negative effect of the lack

of this methyl group on the recognition by mAb MC159 could

be counterbalanced by the positive effect expected with the

presence of Adda instead of its aldol precursor. As previously

described, the conjugation of N-Boc-Adda to amino acids or

680 | Org. Biomol. Chem., 2010, 8, 676–690

This journal is

The Royal Society of Chemistry 2010

©

View Article Online

Scheme 3 Synthetic reactions used for the preparation Adda and tripeptide 17 from N-Boc-Adda. Reagents and conditions: i) HATU, 2,4,6-collidine,

H-D-Glu(Ala-OMe)-OMe, DMF, rt; ii) TFA, CH₂Cl₂, reﬂux; iii) aq. LiOH, CH₃OH, rt.

related peptides through amidifcation reaction is not a trivial

synthetic task and often requires non standard peptide coupling

conditions. Representatively, satisfactory results to get the fully-

protected tripeptide were only obtained through the use of the

optimised protocol described by Gulledge et al.:²⁶pre-activation

of carboxylic acid of N-Boc-Adda with uronium salt HATU in the

presence of 2,4,6-collidine followed by aminolysis of the formed

OAt esters by H-D-Glu(Ala-OMe)-OMe 13. Finally, the Boc and

methyl ester groups were sequentially removed by treatment with

TFA and LiOH respectively. Puriﬁcation was achieved by semi-

preparative RP-HPLC to give the novel acyclic MC analogue 17 in

24% overall yield for the three steps. Monoclonal antibodyMC159

binds to this analogue with 45% cross-reactivity as compared to

MC-LR (IC₅₀122 pM). This good result clearly shows that the

stereostructural features of C-2 and C-3 positions of Adda amino

acid are essential for designing simpliﬁed acyclic MC analogues

well-recognised by mAb MC159. In addition to this result, we have

also determined the cross-reactivity of mAb MC159 for binding

MC-LR toward the simplest synthetic analogue of MCs, Adda

itself. TFA-mediated deprotection of N-Boc-Adda followed by

semi-preparative RP-HPLC puriﬁcation provided this b-amino

acid in a pure form. Cross-reactivity for binding mAb MC159 of

this analogue relative to MC-LR is 58% (IC₅₀94 pM). Thus, with

these two additional Adda-based derivatives, we have a complete

panel of MC-LR analogues with cross-reactivity for MC159

binding varying from 12% to 58% required for the implementation

of SPIT-FRI. Since Adda is synthetically accessible more rapidly, it

was preferred for the preparation of a ﬁrst “tripod”, the key reagent

in this original competitive immunoassay, with the ﬁrst goals in

mind: (1) to evaluate the whole set of chemical steps required for

the tripod synthesis using a fragile toxin analogue, (2) to have an

“universal” tripod which could be applied for the detection of any

Adda-based toxin (if the appropriate mAb is produced), and (3)

evaluate the solid-phase attachment efﬁcacy.

solid-phase (Scheme 4). Thus, it is essential to introduce two or-

thogonal reactive groups (for chemoselective ﬂuorescent labelling

and subsequent solid-phase immobilisation) within the Adda

structure without altering its recognition by mAb MC159. To our

knowledge, there are no reports in the literature for ﬂuorescent

labelling of this unusual b-amino acid. Several ﬂuorescent probes

of MC-LR have been prepared but the labelling strategies are

based on the chemoselective derivatisation of the side-chain of

an amino acid other than Adda: (1) conjugation under basic con-

ditions of a tetramethylrhodamine 1,3-diketone derivative to the

arginine side-chain²⁷or (2) a two-step process involving a Michael

type addition of an aminothiol (i.e., cysteamine or cysteine)

to the a,b-unsaturated carbonyl of the Mdha residue followed

by the reaction of its amino group with N-hydroxysuccinimidyl

(NHS) of a terbium cryptate²⁸or isothiocyanate derivative of

ﬂuorescein (FITC).²⁹Consequently, we have explored the synthesis

of an aldehyde-reactive Adda-based ﬂuorescent “tripod” through

the four-step sequential derivatisation of an in-house-developed

heterotrifunctional cross-linking reagent (Scheme 4).¹³Indeed,

this generic bio-labelling reagent contains three orthogonal reac-

tive groups: an amine-reactive N-hydroxysuccinimidyl carbamate

moiety, an aldehyde/ketone-reactive aminooxy group (protected

as phthalimide) and a thiol group (protected as disulﬁde) with

a propensity to form urea, oxime and thioether linkages under

mild conditions respectively. First, Adda readily reacted through

its free amine, with a slight excess of reagent 18 in NMP in

the presence of DIEA to obtain Adda-peptidyl conjugate 19. As

previously observed with other amine-containing compounds, this

acylation reaction was found to be fast and efﬁcient. Puriﬁcation

was achieved by RP-HPLC to give 19 in 48% yield. Such a reaction

could be favorably performed in alkaline aq. buffers (i.e., NaHCO₃

or borate buffer, pH 8–9) but the high degree of hydrophobicity of

the Adda side-chain prevents its solubilisation in such a reaction

medium. Thereafter, the phthaloyl and SEt protecting groups

were cleanly removed by using the two-step sequential synthetic

procedure previously described by us.¹³Finally, the chemoselective

ﬂuorescent labelling of intermediate free-sulfhydryl peptide was

achieved by Michael addition of the newly formed thiol group to

the maleimide moiety of a water-soluble analogue of rhodamine

Preparation of the ﬁrst “microcystin tripod” through the sequential

derivatisation of a heterotrifunctional cross-linking reagent

Once an MC-LR analogue has been selected (i.e., Adda), the next

step for the implementation of the SPIT-FRI detection method

is the design and synthesis of an Adda-derived ﬂuorescent probe

which will be able to react readily with an aldehyde-functionalised

R

6G (FluoProbe^ꢀ532, FP532). Noteworthily, as we previously

described, no reaction of the aminooxy group was observed in

these conditions, even with an excess of maleimide ﬂuorescent

This journal is

The Royal Society of Chemistry 2010

Org. Biomol. Chem., 2010, 8, 676–690 | 681

©

View Article Online

Scheme 4 Synthetic reactions used for the preparation of ﬂuorescent “microcystin-tripod” 20. Reagents and conditions: i) Adda, DIEA, NMP, rt;

ii) hydrazine monohydrate, CH₃OH, rt; iii) DTT, 0.1 M borate buffer (pH 8.2), rt; iv) FluoProbes^ꢀR532A maleimide, 0.1 M NaHCO₃buffer (pH 8.5),

CH₃CN, rt.

thiol reagent. The resulting aminooxy-containing Adda-derived

ﬂuorescent probe 20 was isolated in a pure form by RP-HPLC. The

modest 5% overall yield (for the three steps) is explained both

by the non-quantitative yield of the aminooxy deprotection step

(~50%) and by signiﬁcant losses of intermediate conjugates during

the chromatographic puriﬁcation and lyophilisation steps. Indeed,

this is the main unavoidable drawback of efﬁcient bioconjugation

reactions performed at the micromole scale. The structure of 20

was conﬁrmed by ESI mass spectrometry (see ESI†). Further

evidence in favor of the integrity of “microcystin tripod” 20,

especially its ﬂuorescent label and the aminooxy functionality,

was provided by immobilisation experiments. Thus, this probe

was grafted on an aldehyde-functionalised glass surface using the

same protocol as previously described for “substance P-tripod”.¹³

After washings, an intense ﬂuorescence signal was observed and

proved the efﬁcacy of the oxime ligation to covalently graft Adda

related “tripods” to a solid surface (Fig. 3a). A spotting experiment

with the non-reactive oxime derivative (obtained by mixing 20

with acetone) gave no signiﬁcant signal (Fig. 3b). Furthermore,

the concentration of the aminooxy probe was varied (1–170 mM)

and comparison of the intensities of ﬂuorescence conﬁrmed that

the amount of anchored “tripod” increases with increasing tripod

concentration until saturation of surface reactive sites is reached

Fig.

3 Immobilisation of ﬂuorescent “microcystin-tripod” 20 on

around 80 mM (Fig. 3c). This result proved the efﬁcient and

aldehyde-functionalised silica surface by oxime ligation.

682 | Org. Biomol. Chem., 2010, 8, 676–690

This journal is

The Royal Society of Chemistry 2010

©

View Article Online

concentration-dependant immobilisation process of 20 involving

its aminooxy group. We thus proved the efﬁcacy of the solid-

phase immobilisation of such an Adda-based ﬂuorecent probe.

Since the ﬂuorescent Adda derivative 20 is signiﬁcantly recognised

by mAb MC159 (IC₅₀211 pm, molar cross-reactivity compared

to MC-LR 26%), the present microarrays composed of smaller

Adda conjugates should be suitable to investigate activation and

competition steps of SPIT-FRI, for instance, aimed at detection

and quantifying MC-LR in continuous ﬂow. The corresponding

results will be reported in due course.

ethanol. All solvents were dried following standard procedures

(CH₃CN: distillation over CaH₂, CH₂Cl₂: distillation over P₂O₅,

DMF: distillation over BaO, NMP: distillation over BaO, THF:

distillation over Na⁰/benzophenone, toluene: distillation over

Na⁰). DIEA and TEA were distilled from CaH₂and stored over

BaO. Activated MnO₂(tech., Mn 58% min) and protected amino

acids were obtained from Alfa Aesar and Bachem respectively.

The thiol-reactive water-soluble analogue of rhodamine 6G

R

“FluoProbes^ꢀ532A maleimide” was purchased from Interchim.

Chiral thiazolidinethiones 1 and 9 were synthesised using literature

procedures.^20,32The HPLC-gradient grade acetonitrile (CH₃CN)

and methanol (CH₃OH) were obtained from Acros or Fisher

Scientiﬁc. Buffers and aq. mobile-phases for HPLC were prepared

using water puriﬁed with a Milli-Q system (puriﬁed to 18.2 MX

cm). ¹H and ¹³C NMR spectra were recorded on a Bruker

DPX 300 spectrometer (Bruker, Wissembourg, France). Chemical

shifts are expressed in parts per million (ppm) and relative to

tetramethylsilane from CD₃CN (d_H= 1.96, d_C= 1.79 (CH₃),

Conclusions

In this paper, we have described the synthesis of four novel acyclic

analogues of microcystin-LR bearing Adda amino acid (or its

aldol precursor) as the key component. Some practical changes

have been brought to the linear total synthesis of this unusual

b-amino acid initially developed by Pearson et al. to get rapidly

synthetic intermediates (aldol 10 and N-Boc-Adda) useful for the

preparation of targeted MC analogues through peptide coupling

or trans-amidiﬁcation reactions. Their use in a comprehensive

structure/cross-reactivity relationship study toward mAb anti-

MC MC159 binding ability has enabled us to conclude that the

use of the greatly simpliﬁed macrocyclic heptapeptide of MC-LR

is possible without suppressing a good recognition of the

corresponding analogues by anti-MC mAbs. These results are

valuable especially for selecting suitable analytes required for

the development of immunoassays aimed at the non-selective

(alert) or selective (toxin identiﬁcation) detection of MCs and

other cyanobacterial toxins such as nodularins, which also display

the Adda b-amino acid unit. Interestingly, the two efﬁcient

synthetic strategies leading to the Adda conjugates 14, 15 and

17 are also promising methods that may prove useful in the near

future for the rapid production of a new generation of speciﬁc

phosphatase inhibitors. They constitute good alternatives to the

general synthesis of Adda-containing peptides reported by Cundy

et al. which relies on the ring opening of an Adda-derived b-

lactam by various aminoesters.³⁰Furthermore, the use of an “in

house” heterotrifunctional cross-linking reagent has enabled us to

prepare the ﬁrst ﬂuorescent conjugate of Adda. Interestingly, the

presence of an additional reactive group (i.e., aminooxy group),

within the peptide scaffold of this conjugate allows the grafting

of another reporter group (e.g., a second ﬂuorophore to create a

FRET pair). Thus, this latter synthetic tool could be the starting

point for the design of activity-based probes (ABPs)³¹to study and

dissect serine/threonine phosphatase activity.

118.26 (CN)), CDCl₃(d_H= 7.26, d_C= 77.36) or CD₃OD (d_H

=

3.31, d_C= 49.00).³³J values are in Hz. Infrared (IR) spectra

were recorded as thin-ﬁlm on sodium chloride plates or KBr

pellets using a Perkin Elmer FT-IR Paragon 500 spectrometer

with frequencies given in reciprocal centimetres (cm^-1). Optical

rotations were measured by using a Perkin Elmer 341 polarimeter

(sodium D ray at l = 589 nm). Elemental analyses were carried out

on a Carlo Erba EA 1110 CHNS–O instrument. UV-visible spectra

were obtained on a Varian Cary 50 scan spectrophotometer.

Chromophore-containing compounds were quantiﬁed by UV-

visible spectroscopy at the l_maxusing the corresponding tabulated

molar extinction coefﬁcient. Fluorescence spectroscopic studies

were performed with a Varian Cary Eclipse spectrophotometer.

Analytical HPLC was performed on a Thermo Electron Surveyor

instrument equipped with a PDA detector. Semi-preparative

HPLC was performed on a Finnigan SpectraSYSTEM liquid

chromatography system equipped with UV-visible 2000 detector.

Mass spectra were obtained with a Thermo Finnigan LCQ Ad-

vantage Max (ion-trap) apparatus equipped with an electrospray.

The monoclonal antibodies anti-MC were obtained as previously

described and the corresponding cross-reactivity experiments were

performed by using standard conditions.¹⁶

High-performance liquid chromatography separations

Several chromatographic systems were used for the analytical

experiments and the puriﬁcation steps. System A: RP-HPLC

(Thermo Hypersil GOLD C₁₈column, 5 mm, 4.6 ¥ 150 mm)

with CH₃CN and 0.1% aqueous triﬂuoroacetic acid (aq. TFA,

0.1%, v/v, pH 2.0) as eluents [20% CH₃CN (5 min), followed by

linear gradient from 20 to 90% (35 min) of CH₃CN] at a ﬂow rate

of 1 mL min^-1. UV detection was achieved at 254 nm. System

B: System A with the following gradient [0% CH₃CN (5 min),

followed by linear gradient from 0 to 60% (30 min) of CH₃CN].

Visible detection was achieved at 550 nm. System C: RP-HPLC

(Thermo Hypersil GOLD C₁₈column, 5 mm, 10 ¥ 250 mm) with

CH₃CN and 0.1% aq. TFA as eluents [30% CH₃CN (5 min),

followed by linear gradient from 30 to 60% (40 min) of CH₃CN] at

a ﬂow rate of 4 mL min^-1. UV detection was achieved at 260 nm.

System D: System C with the following gradient [20% CH₃CN

(5 min), followed by linear gradient from 20 to 30% (5 min) and

Experimental section

General

Column chromatography puriﬁcations were performed on silica

gel (40–63 mm) from SdS. TLC were carried out on Merck DC

Kieselgel 60 F-254 aluminium sheets. Compounds were visualised

by one or more of the following methods: (1) illumination with

a short wavelength UV lamp (i.e., l = 254 nm), (2) spray with

a 0.2% (w/v) ninhydrin solution in absolute ethanol, (3) spray

with a 3.5% (w/v) phosphomolybdic acid solution in absolute

This journal is

The Royal Society of Chemistry 2010

Org. Biomol. Chem., 2010, 8, 676–690 | 683

©

View Article Online

30 to 70% (40 min) of CH₃CN]. System E: RP-HPLC (Waters

Xterra MS C₁₈, 5 mm, 7.8 ¥ 100 mm) with CH₃CN and 0.1% aq.

TFA as eluents [20% CH₃CN (5 min), followed by linear gradient

from 20 to 30% (5 min) and 30 to 60% (30 min) of CH₃CN] at a ﬂow

rate of 2.5 mL min^-1. UV detection was achieved at 250 nm. System

F: System E with the following gradient [0% CH₃CN (5 min),

followed by linear gradient from 0 to 40% (5 min) and 40 to 70%

(30 min) of CH₃CN]. System G: System E with the following

gradient [0% CH₃CN (5 min), followed by linear gradient from

0 to 30% (5 min) and 30 to 70% (35 min) of CH₃CN]. System

H: System E with the following gradient [20% CH₃CN (5 min),

followed by linear gradient from 20 to 30% (5 min) and 30 to 60%

(30 min) of CH₃CN]. Visible detection was achieved at 550 nm.

CH₃), 3.45 (s, 3H, CH₃), 3.63 (s, 1H, OH), 4.06–4.13 (m, 1H,

CH), 7.19–7.22 (m, 5H, Ph); ¹³C NMR (75.5 MHz, CDCl₃): d

10.2, 31.8, 37.2, 40.0, 61.2, 72.9, 126.3, 128.4, 129.1, 138.3, 177.9.

(c) Alkylation reaction: Iodomethane (40.1 mL, 636.4 mmol) was

dissolved in dry THF (140 mL) and the resulting solution was

stirred with solid Na₂CO₃(1.38 g, 13 mmol) for 10 min. The

mixture was cooled to 4 ^◦C, ﬁltered through cotton and added

to the Weinreb amide previously obtained (15.1 g, 64 mmol).

Thereafter, 2.49 g (63.6 mmol) of NaH (60% dispersion in mineral

oil) was added at 4 ^◦C. The reaction mixture was stirred at room

temperature for 4 h. The reaction was checked for completion by

TLC (cyclohexane–AcOEt, 6 : 4, v/v). Further amount of NaH

(60% oil, 1.24 g, 31.8 mmol) was added and the reaction mixture

was stirred at room temperature for further 2 h. The reaction was

checked for completion by TLC (cyclohexane–AcOEt, 6 : 4, v/v).

After cooling to 4 ^◦C, the excess of NaH was destroyed by addition

of aq. sat. NH₄Cl (150 mL). The mixture was extracted twice

by AcOEt (3 ¥ 200 mL). The organic layer was washed with

brine (250 mL), dried over Na₂SO₄, ﬁltered and concentrated

under reduced pressure. The resulting residue was puriﬁed by

chromatography on a silica gel column (300 g) with a mixture

of cyclohexane–AcOEt (6 : 4, v/v) as the mobile phase to give 3

as a yellow oil (10.85 g, 43.2 mmol, overall yield for the three

steps 43%). Spectroscopic data are identical to those reported by

Pearson et al.

(2R,3S)-N,3-Dimethoxy-N,2-dimethyl-4-phenylbutanamide (3)

(a) Crimmins aldolisation: Chiral N-propionylthioazolidinethione

1 (26.2 g, 104 mmol) was dissolved in dry CH₂Cl₂(250 mL) and

the resulting solution was cooled to -78 ^◦C with a bath of dry ice

in acetone. TiCl₄(12 mL, 109.4 mmol) was added dropwise and

the resulting mixture allowed to stir for 15 min in order to form

the titanium complex. Then, freshly distilled DIEA (18.9 mL,

114.4 mmol) was added and the mixture was stirred for 40 min

under the same conditions to form the corresponding titanium

enolate. Thereafter, freshly distilled NMP (20.0 mL, 208 mmol)

was added. After 10 min, freshly distilled phenylacetaldehyde

(13.4 mL, 114.4 mmol) was added. After keeping the solution

at -78 ^◦C for 30 min, the reaction mixture was allowed to

warm to 4 ^◦C and stirred at this temperature for a further 1 h.

The reaction was checked for completion by TLC (cyclohexane–

AcOEt, 7 : 3, v/v) and quenched by the addition of aq. 50% NH₄Cl

(300 mL). The organic layer was washed twice by aq. 1.0 N HCl (2 ¥

300 mL), aq. 10% Na₂CO₃(300 mL) and ﬁnally brine (300 mL)

before drying over Na₂SO₄. Aldol 2 was obtained as a yellow

solid and used in the next step without further puriﬁcation. R_f

(cyclohexane–AcOEt, 7 : 3, v/v) 0.61; [a] -164 (c 0.6 CHCl₃); IR

(KBr): n_max698, 772, 1047, 1156, 1221, 1258, 1316, 1453, 1496,

(4S,5S,2E)-2,4-Dimethyl-5-methoxy-6-phenylhex-2-en-1-ol (5)

(a) Reduction: Weinreb amide 3 (10.71 g, 42.6 mmol) was dissolved

in dry CH₂Cl₂(100 mL) and the resulting solution cooled to -78 ^◦C

with a bath of dry ice in acetone. A 1.0 M solution of DIBAL-H

in dry toluene (100 mL, 100 mmol) was added dropwise over a

period of 80 min. The reaction was checked for completion by

TLC (cyclohexane–AcOEt, 7 : 3, v/v) and quenched by sequential

addition of methanol (40 mL) and aq. sat. NH₄Cl (40 mL). The

resulting solution was allowed to warm to room temperature for

R

30 min. The mixture was then ﬁltered trough a Celite^ꢀ545 pad

1

1643 (broad), 2359, 2926, 3416 (broad); H NMR (300 MHz,

(to remove aluminium salts) and washed with a large amount

of Et₂O (300 mL). The combined organic layers were washed

with aq. 2.0 N HCl (200 mL), dried over MgSO₄, ﬁltered and

concentrated under vacuum to give intermediate aldehyde as

an orange oil that was used in the next step without further

puriﬁcation. R_f(cyclohexane–AcOEt, 7 : 3, v/v) 0.63; [a] -3.6

CDCl₃): d 1.29 (d, J = 7.5 Hz, 3H, CH₃), 2.71–2.86 (m, 2H,

CH₂), 3.04–3.09 (dd, J = 11.3 Hz, 1.9 Hz, 1H), 3.67–3.92 (m, 1H,

CH), 4.18–4.23 (m, 1H, CH), 4.53–4.61 (qd, 6.8 Hz, 3.0 Hz, 1H,

CH), 6.17 (d, 1H), 7.21–7.35 (m, 10H, Ph); ¹³C NMR (75.5 MHz,

CDCl₃): d 10.8, 27.0, 36.7, 40.7, 43.0, 70.1, 73.4, 125.4, 126.8,

128.7, 129.2, 129.5, 178.0; MS (ESI+): m/z 372.20 [M + H]⁺, calcd

for C₂₀H₂₁NO₂S₂: 371.52. (b) Trans-amidiﬁcation reaction: Aldol 2

(37.1 g, 100 mmol) was dissolved in dry CH₂Cl₂(150 mL) and

N,O-dimethylhydroxylamine hydrochloride (14.6 g, 150 mmol)

and imidazole (27.2 g, 400 mmol) were sequentially added. The

resulting reaction mixture was stirred at room temperature for 12 h.

The reaction was checked for completion by TLC (cyclohexane–

AcOEt, 7 : 3, v/v). The solution was then washed with deionised

water (3 ¥ 300 mL), the organic layer was dried over Na₂SO₄,

ﬁltered and concentrated under vacuum. The resulting residue

was puriﬁed by chromatography on a silica gel column (600 g)

with a step gradient of AcOEt (30–50%) in cyclohexane as the

mobile phase, to give the intermediate Weinreb amide as a yellow

oil (15.1 g, 64 mmol). R_f(cyclohexane–AcOEt, 7 : 3, v/v) 0.12;

[a] -25.7 (c 1.3 in CH₃OH); ¹H NMR (300 MHz, CDCl₃): d 1.25

(d, J = 7.5 Hz, 3H), 2.66–2.90 (m, 3H, CH₂& CH), 3.13 (s, 3H,

1

(c 1.1 in CH₃OH); H NMR (300 MHz, CDCl₃): d 1.16 (d, J =

7.2 Hz, 3H, CH₃), 2.66–2.99 (m, 2H, CH₂), 3.28 (s, 3H, CH₃), 3.84–

3.90 (m, 1H, CH), 7.14–7.31 (m, 5H, Ph), 9.66 (s, 1H, C(O)H); ¹³

C

NMR (75.5 MHz, CDCl₃): d 7.6, 37.5, 49.0, 58.1, 81.7, 126.6,

129.3, 138.1. (b) Wittig reaction: The freshly prepared aldehyde

(8.48 g) was dissolved in dry toluene (275 mL), stabilised phos-

phorus ylide (1-ethoxycarbonylethylidene)triphenylphosphorane

(18.52 g, 51.1 mmol) was added and the resulting mixture was

stirred under reﬂux for 12 h. The reaction was checked for

completion by TLC (cyclohexane–AcOEt, 8 : 2, v/v). The reaction

mixture was concentrated under reduced pressure and the resulting

residue was puriﬁed by chromatography on a silica gel column

(250 g) with a step gradient of AcOEt (0–10%) in cyclohexane

as the mobile phase, to give the targeted alkene 4 as a yellow

oil. R_f(cyclohexane–AcOEt, 7 : 3, v/v) 0.59; [a] -6.78 (c 0.6 in

1

CH₃OH); H NMR (300 MHz, CDCl₃): d 1.07 (d, J = 6.8 Hz,

684 | Org. Biomol. Chem., 2010, 8, 676–690

This journal is

The Royal Society of Chemistry 2010

©

View Article Online

3H, CH₃), 1.28 (d, J = 6.1 Hz, 3H, CH₃), 1.74 (s, 3H, CH₃), 2.57–

2.85 (m, 2H, CH₂), 3.26 (s, 3H, CH₃), 4.16–4.23 (q, J = 7.2 Hz,

2H), 6.68 (d, J = 10.1 Hz, 1H, CH), 7.14–7.74 (m, 5H, Ph). (c)

Reduction: After its puriﬁcation, alkene 4 (7.59 g, 27.5 mmol)

was dissolved in dry THF (55 mL) and the resulting solution was

cooled to -78 ^◦C with a bath of dry ice in acetone. A 1.0 M

solution of DIBAL-H in dry toluene (68 mL, 68 mmol) was added

dropwise over a period of 1 h. The resulting reaction mixture was

stirred at this temperature for 90 min and checked for completion

by TLC (cyclohexane–AcOEt, 8 : 2, v/v). Thereafter, the reaction

was quenched by sequential addition of methanol (30 mL) and aq.

sat. NH₄Cl (30 mL). The resulting solution was allowed to warm

to room temperature for 30 min. The mixture was then ﬁltered

mobile phase to give the ester 7 as a yellow oil (3.14 g, 10.4 mmol,

yield 72%). R_f(cyclohexane–AcOEt, 8 : 2, v/v) 0.76; [a] -45.2

1

(c 0.5 in CHCl₃); H NMR (300 MHz, CDCl₃): d 1.07 (d, J =

6.4 Hz, 3H, CH₃), 1.29 (t, J = 3 Hz, 3H, CH₃), 1.66 (s, 3H, CH₃),

2.60–2.84 (m, 3H, CH₂& CH), 3.20–3.25 (m, 4H, CH₃& CH),

4.18–4.25 (q, J = 7.1 Hz, 2H, CH₂), 5.78–5.83 (m, 2H, 2 ¥ CH),

7.17–7.27 (m, 5H, Ph); ¹³C NMR (75.5 MHz, CDCl₃): d 12.8, 14.8,

16.2, 37.5, 38.6, 59.1, 60.7, 87.0, 116.7, 126.6, 128.7, 129.9, 132.9,

139.4, 144.8, 150.0, 167.9.

(2S,3R,4E,6E,8S,9S)-1-O-(tert-Butoxycarbamoyl)-3-hydroxy-9-

methoxy-2,6,8-trimethyl-10-phenyl-4,6-decadienoate (11)

R

trough a Celite^ꢀ545 pad (to remove aluminium salts) and washed

(a) Reduction: Ester 7 (1.0 g, 3.3 mmol) was dissolved in dry

THF (12 mL), and the resulting solution was cooled to -78 ^◦C

with a bath of dry ice in acetone. A 1.0 M solution of DIBAL-

H in dry toluene (12 mL, 12 mmol) was added dropwis_◦e over

15 min. The resulting reaction mixture was stirred at -78 C for

90 min and checked for completion by TLC (cyclohexane–AcOEt,

8 : 2, v/v). Thereafter, the reaction was quenched at -78 ^◦C by slow

addition of methanol (5 mL) and aq. sat. NH₄Cl (5 mL). Then the

solution was allowed to warm to room temperature for 30 min.

The organic compounds were extracted by AcOEt (3 ¥ 50 mL)

after the addition of 1.0 N aq. HCl (50 mL). The organic layer was

dried over MgSO₄, ﬁltered and concentrated under vacuum. The

residue was then puriﬁed by chromatography on a silica gel column

(30 g) with a step gradient of AcOEt (0–20%) in cyclohexane as

the mobile phase, giving the intermediate alcohol as a yellow oil

that was used in the next step without further puriﬁcation. R_f

(cyclohexane–AcOEt, 8 : 2, v/v) 0.26; [a] -20.2 (c 1.7 in CHCl₃);

¹H NMR (300 MHz, CDCl₃): d 1.04 (d, J = 6.8 Hz, 3H), 1.66

(s, 3H), 2.19 (bs, 1H, OH), 2.56–2.85 (m, 3H, CH₂& CH), 3.16–

3.23 (s, 4H, CH₃& CH), 4.18 (d, J = 6.0 Hz, 2H, CH₂), 5.40 (d,

J = 9.8 Hz, 1H), 5.68–5.77 (dt, 1H, J = 15 Hz, 6.3 Hz, CH),

7.16–7.20 (m, 5H, Ph); ¹³C NMR (75.5 MHz, CDCl₃): d 12.3,

15.9, 36.3, 37.8, 58.3, 63.3, 86.6, 125.5, 125.6, 127.8, 129.0, 132.3,

135.4, 136.1, 138.9. (b) Oxidation: The alcohol was dissolved in

dry toluene (35 mL) and the resulting solution was cooled to

with CH₂Cl₂(150 mL). The organic layer was washed with aq.

1.0 N HCl (200 mL); re-extraction of this latter aq. layer with

Et₂O (280 mL) was needed for a complete recovery of the desired

alcohol. The combined organic layers were dried over MgSO₄,

ﬁltered and concentrated under vacuum. The resulting brown oily

residue was puriﬁed by chromatography on a silica gel column (400

g) with a step gradient of AcOEt (0–20%) in cyclohexane as the

mobile phase, to give allylic alcohol 5 as a pale yellow oil (4.76 g,

20.3 mmol, overall yield for the three steps 48%). R_f(cyclohexane–

AcOEt, 8 : 2, v/v) 0.19; ¹H NMR (300 MHz, CDCl₃): d 1.02 (d, J =

6.8 Hz, 3H, CH₃), 1.57 (s, 3H, CH₃), 2.37 (bs, 1H, OH), 2.37–2.84

(m, 3H, CH₂& CH), 3.19 (s, 4H, CH₃& CH), 3.97 (s, 2H, CH₂),

5.32 (d, J = 9.8 Hz, 1H, CH), 7.18–7.26 (m, 5H, Ph); ¹³C NMR

(75.5 MHz, CDCl₃): d = 13.9, 16.4, 35.9, 38.0, 58.1, 68.6, 87.1,

126.2, 128.3, 128.4, 129.3, 135.0, 139.6. All other spectroscopic

data are identical to those reported by Pearson et al.

(6S,7S,2E,4E)-Ethyl 4,6-dimethyl-7-methoxy-8-phenylocta-

2,4-dienoate (7)

(a) Oxidation: Allylic alcohol 5 (3.4 g, 14.5 mmol) was dissolved

in dry toluene (100 mL) and the resulting solution was cooled to

0

^◦C. Activated MnO₂(tech., Mn 58% min, 13.6 g, 156 mmol)

was added and the resulting suspension vigorously stirred at

◦

55 C for 2 h. The reaction was checked for completion by TLC

4

^◦C. Activated MnO₂(tech., Mn 58% min, 4.0 g, 26.8 mmol)

(cyclohexane–AcOEt, 8 : 2, v/v). The resulting suspension was

was added and the reaction mixture was stirred at 50 ^◦C for 1 h.

The reaction was checked for completion by TLC (cyclohexane–

R

ﬁltered on a Celite^ꢀ545 pad and washed with a large amount

R

of CH₂Cl₂. The resulting ﬁltrate was concentrated under reduced

pressure to give aldehyde 6 as a yellow oil (3.39 g, 14.5 mmol, quan-

titative yield). R_f(cyclohexane–AcOEt, 7 : 3, v/v) 0.68; ¹H NMR

(300 MHz, CDCl₃): d 1.13 (d, J = 6.8 Hz, 3H, CH₃), 1.62 (s, 3H,

CH₃), 2.80 (bm, 3H, CH₃), 3.29–3.24 (m, 4H, CH₃& CH), 6.41 (d,

J = 9.8 Hz, 1H, CH), 7.16–7.31 (m, 5H, Ph), 9.38 (s, 1H, C(O)H);

¹³C NMR (75.5 MHz, CDCl₃): d 9.5, 14.8, 37.1, 38.0, 58.7, 85.8,

126.5, 128.4, 128.5, 129.5, 138.0, 138.6, 138.9, 156.7, 195.6. (b) Wit-

tig reaction: The freshly prepared aldehyde 6 (3.39 g, 14.5 mmol)

was dissolved in dry toluene (80 mL), stabilised phospho-

rus ylide (1-ethoxycarbonylmethylidene)triphenylphosphorane

(6.06 g, 17.4 mmol) was added and the resulting reaction mixture

was stirred under reﬂux overnight. The reaction was checked for

completion by TLC (cyclohexane–AcOEt, 8 : 2, v/v). Thereafter,

AcOEt, 7 : 3, v/v) and the suspension was ﬁltered on a Celite^ꢀ545

pad and washed with a large amount of CH₂Cl₂. The ﬁltrate was

concentrated under reduced pressure to give aldehyde 8 as a yellow

oil that was used in the next step without further puriﬁcation. R_f

(cyclohexane–AcOEt, 7 : 3, v/v) 0.74; [a] -6.8 (c 0.6 in CHCl₃); ¹H

NMR (300 MHz, CDCl₃): d 1.10 (d, J = 6.8 Hz, 3H, CH₃), 1.70

(s, 3H, CH₃), 2.76–2.80 (m, 3H, CH₃), 3.28 (s, 4H, CH₃& CH),

5.95 (d, J = 9.8 Hz, 1H, CH), 6.07–6.15 (dd, J = 15.5 Hz, 7.5 Hz,

1 H, CH), 7.09–7.31 (m, 6H, Ph & CH), 9.57 (d, J = 7.5 Hz, 1H,

C(O)H); ¹³C NMR (75.5 MHz, CDCl₃): d 12.6, 15.5, 37.2, 38.1,

58.7, 86.4, 126.4, 127.4, 128.4, 129.2, 133.1, 138.9, 146.8, 158.0. (c)

Crimmins aldolisation: Chiral N-propylthiazolidinethione 9 (1.3 g,

4.95 mmol) was dissolved in_◦dry CH₂Cl₂(35 mL) and the resulting

solution was cooled to -78 C with a bath of dry ice in acetone.

TiCl₄(592 mL, 5.4 mmol) was added dropwise and the resulting

reaction mixture was stirred at room temperature for 20 min

in order to form the titanium complex. Then, freshly distilled

DIEA (891 mL, 5.4 mmol) was added at -78 ^◦C and the mixture

R

the reaction mixture was ﬁltered on a Celite^ꢀ545 pad, washed with

CH₂Cl₂and concentrated under reduced pressure. The resulting

oily residue was puriﬁed by chromatography on a silica gel column

with a step gradient of AcOEt (0-20%) in cyclohexane as the

This journal is

The Royal Society of Chemistry 2010

Org. Biomol. Chem., 2010, 8, 676–690 | 685

©

View Article Online

was stirred for a further 40 min at room temperature to form

the titanium enolate. Thereafter, freshly distilled NMP (962 mL,

10 mmol) was added at -78 ^◦C. 10 min later, a solution of freshly

prepared aldehyde 8 in dry CH₂Cl₂(5 mL) was added dropwise.

The reaction mixture was stirred at -78 ^◦C for 30 min, then allowed

All spectroscopic data of this compound are identical to that

previously reported in the literature.

Dipeptide H-D-Glu(MeAla-OMe)-OMe (12)

◦

to warm at 4 C and stirred for a further 1 h. The reaction was

(a) Peptide coupling: Boc-D-Glu-OMe (386 mg, 1.47 mmol), H-

MeAla-OMe (250 mg, 1.63 mmol) and PyBrOP reagent (685.3 mg,

1.47 mmol) were dissolved in dry CH₃CN (2 mL) and the resulting

mixture was cooled to 4 ^◦C. Dry DIEA (768 mL, 4.41 mmol) was

added dropwise and the resulting reaction mixture was stirred

at room temperature for 12 h. The reaction was checked for

completion by TLC (CH₂Cl₂–CH₃OH, 9 : 1, v/v). The mixture

was evaporated to dryness and the resulting residue was taken

up with AcOEt (25 mL), washed by aq. 10% citric acid (25 mL),

aq. sat. NaHCO₃(25 mL) and brine (25 mL). The organic layer

was dried over Na₂SO₄, ﬁltrated and evaporated to dryness. The

resulting residue was puriﬁed by chromatography on a silica gel

column with a step gradient of CH₃OH (0–4%) in CH₂Cl₂as the

mobile phase, giving Boc-D-Glu(MeAla-OMe)-OMe as a yellow

oil (530 mg, 1.47 mmol, quantitative yield). R_f(CH₂Cl₂–CH₃OH,

9 : 1, v/v) 0.65; IR (KBr): n_max756, 1028, 1052, 1084, 1168, 1214,

1366, 1408, 1455, 1519, 1644, 1714, 1743, 2979, 3340 (broad); ¹H

NMR (300 MHz, CDCl₃): d 1.39 (d, J = 7.3 Hz, 3H, CH₃), 1.44 (s,

9H, tBu), 1.94–2.52 (m, 4H, CH₂b Glu, CH₂g Glu), 2.92 (s, 3H,

CH₃), 3.71 (s, 3H, CH₃), 3.74 (s, 3H, CH₃), 4.29 (bs, 1H, CH a Glu),

5.22 (q, J = 7.2 Hz, 1H, CH a MeAla); MS (ESI+): m/z 383.07

[M + Na]⁺, 399.00 [M + K]⁺, calcd for C₁₆H₂₈N₂O₇: 360.41. (b)

Boc removal: Dipeptide Boc-D-Glu(MeAla-OMe)-OMe (404 mg,

1.12 mmol) was dissolved in dry THF (12 mL). TFA (6.7 mL,

89.7 mmol) was added dropwise and the resulting reaction mixture

was heated under reﬂux for 2 h. The reaction was checked for

completion by TLC (CH₂Cl₂–CH₃OH, 9 : 1, v/v) and the reaction

mixture was evaporated to dryness. The resulting oily residue was

puriﬁed by chromatography on a silica gel column with a step

gradient of CH₃OH (0–10%) in CH₂Cl₂as the mobile phase, to

give the TFA salt of H-Glu(MeAla-OMe)-OMe 12 as a yellow oil

(244 mg, 0.65 mmol, yield 58%). R_f(CH₂Cl₂–CH₃OH, 9 : 1, v/v)

checked for completion by TLC (cyclohexane–AcOEt, 7 : 3, v/v)

and quenched by the sequential addition of aq. sat. NH₄Cl (15 mL)

and aq. 1.0 N HCl (15 mL). This mixture was then washed twice by

aq. 1.0 N HCl (2 ¥ 20 mL) and brine (20 mL) before drying over

Na₂SO₄. The organic layer was ﬁltered and concentrated under

vacuum. The resulting residue was puriﬁed by chromatography on

a silica gel column (60 g) with a step gradient of AcOEt (0–20%)

in cyclohexane as the mobile phase, to give the aldol 10 as a yellow

oil. R_f(cyclohexane–AcOEt, 7 : 3, v/v) 0.49; [a] +121.7 (c 0.6 in

CHCl₃); ¹H NMR (300 MHz, CDCl₃): d 0.97 (d, J = 6.6 Hz, 3H,

CH₃), 1.32 (d, J = 5.1 Hz, 3H, CH₃), 1.64 (s, 3H, CH₃), 2.49–3.33

(m, 12H, CH₃, 3 ¥ CH₂, CH₂, CH₂& 3 ¥ CH), 4.43–4.46 (t, J =

5.9 Hz, 1H, CH), 4.59 (t, J = 5.3 Hz, 1H, CH), 5.17–5.24 (m, 1H,

CH), 5.42 (d, J = 9.6 Hz, 1H, CH), 5.52–5.60 (dd, J = 6.7 Hz,

15.6 Hz, 1H, CH), 6.32 (d, J = 15.8 Hz, 1H, CH), 7.17–7.25

(m, 10H); ¹³C NMR (75.5 MHz, CDCl₃): d 11.8, 12.8, 16.3, 32.4,

36.7, 36.8, 38.3, 44.9, 58.8, 69.2, 74.5, 86.9, 126.1, 126.3, 127.4,

128.3, 129.0, 129.5, 132.6, 136.5, 136.8, 137.0, 139.4, 177.6, 201.4;

MS (ESI+): m/z 506.53 [M - H₂O + H]⁺, dehydration reaction

occurred within the ESI probe under our ionisation conditions,

calcd for C₃₀H₃₇NO₃S₂: 523.76; elemental analysis: calcd (%) for

C₃₀H₃₇NO₃S₂: C, 68.53; H, 7.48; N, 2.66; found: C, 68.78; H, 7.22;

N, 2.75. (d) Trans-esteriﬁcation reaction: Aldol 10 was dissolved

in dry THF (20 mL), tert-butyl-N-hydroxycarbamate (1.31 g,

3.9 mmol) and imidazole (674 mg, 9.9 mmol) were sequentially

added and the resulting reaction mixture was stirred at room

temperature for 12 h. The reaction was checked for completion by

TLC (cyclohexane–AcOEt, 8 : 2, v/v) and volatiles were removed

by evaporation under vacuum. The resulting residue was taken up

with AcOEt (50 mL) and sequentially washed by aq. 10% citric

acid (50 mL), aq. sat. NaHCO₃(50 mL) and brine (50 mL). The

organic layer was dried over Na₂SO₄, ﬁltrated and evaporated to

dryness. The resulting residue was puriﬁed by chromatography

on a silica gel column with a mixture of cyclohexane–AcOEt

(9 : 1, v/v) as the mobile phase, to give the compound 11 as an

opaque white oil (1.03 g, 2.3 mmol, overall yield for the four

1

0.24; H NMR (300 MHz, CD₃OD): d 1.37 (d, J = 7.1 Hz, 3H,

CH₃), 1.85–2.03 (m, 2H, CH₂b Glu), 2.49–2.55 (m, 2H, CH₂g

Glu), 3.01 (s, 3H, CH₃), 3.62 (t, J = 6.7 Hz, 1H, CH a Glu), 3.70

(s, 3H, CH₃), 3.73 (s, 3H, CH₃), 4.69 (q, J = 7.3 Hz, 1H, CH a

MeAla); MS (ESI+): m/z 261.07 [M + H]⁺, calcd for C₁₁H₂₀N₂O₅:

260.29.

1

steps 70%). R_f(cyclohexane–AcOEt, 8 : 2, v/v) 0.32; H NMR

(300 MHz, CDCl₃): d 1.03 (d, J = 6.7 Hz, 3H, CH₃), 1.24 (d, J =

6.7 Hz, 3H, CH₃), 1.49 (s, 3H, CH₃), 1.64 (s, 3H, CH₃), 2.56–2.99

(m, 4H, 4 ¥ CH), 3.18–3.24 (m, 4H, CH₃& CH), 4.58 (t, J =

4.6 Hz, 1H, CH), 5.43 (d, J = 9.7 Hz, 1H, CH), 5.51–5.58 (dd, J =

6.8 Hz, 8.8 Hz, 1H, CH), 6.32 (d, J = 15.6 Hz, 1H, CH), 7.17–7.30

(m, 5H), 7.89 (s, 1H); ¹³C NMR (75.5 MHz, CDCl₃): d 11.0, 12.8,

16.3, 28.1, 29.5, 29.8, 30.4, 32.1, 36.9, 38.4, 44.2, 58.8, 73.5, 83.9,

87.0, 125.2, 126.1, 128.3, 129.4, 129.6, 132.7, 136.7, 137.4, 139.5,

156.2, 174.5.

Dipeptide H-D-Glu(Ala-OMe)-OMe (13)

(a) Peptide coupling: Boc-D-Glu-OMe (261 mg, 1.0 mmol), H-Ala-

OMe (HCl salt, 139 mg, 1.0 mmol) and BOP reagent (530 mg,

1.2 mmol) were dissolved in dry CH₃CN (2 mL) and the resulting

solution was cooled to 4 ^◦C. Dry DIEA (628 mL, 3.6 mmol) was

added dropwise and the resulting reaction mixture was stirred

at room temperature for 2 h. The reaction was checked for

completion by TLC (CH₂Cl₂–CH₃OH, 9 : 1, v/v). Thereafter, the

mixture was evaporated to dryness and the resulting residue was

taken up with AcOEt (25 mL), sequentially washed with aq. 10%

citric acid (25 mL), aq. sat. NaHCO₃(25 mL) and brine (25 mL).

The organic layer was dried over Na₂SO₄, ﬁltrated and evaporated

to dryness. The resulting residue was puriﬁed by chromatography

on a silica gel column with a step gradient of CH₃OH (0–3%) in

N-Boc-Adda

This N-protected unusual b-amino acid was ﬁnally synthesized

in 50% from compound 11 by a Mitsunobu cyclisation followed

by sodium naphthalide reduction in a “one-pot” process, under

the same experimental conditions as reported by Pearson et al.

686 | Org. Biomol. Chem., 2010, 8, 676–690

This journal is

The Royal Society of Chemistry 2010

©

View Article Online

CH₂Cl₂as the mobile phase, to give Boc-D-Glu(Ala-OMe)-OMe as

a yellow oil (310 mg, 0.9 mmol, yield 90%). R_f(CH₂Cl₂–CH₃OH,

9 : 1, v/v) 0.39; ¹H NMR (300 MHz, CDCl₃): d 1.41 (s, 9H, tBu),

1.87–2.35 (m, 4H, CH₂b Glu, CH₂g Glu), 3.72 (s, 6H, 2 ¥ CH₃),

4.34–4.40 (m, 1H, CH a Glu), 4.50–4.60 (m, 1H, CH a Ala),

5.34 (d, J = 7.9 Hz, 1H, NH), 6.78 (d, J = 6.0 Hz, 1 H, NH).

(b) Boc removal: Dipeptide Boc-D-Glu(Ala-OMe)-OMe (310 mg,

0.9 mmol) was dissolved in THF–H₂O (10 mL, 95 : 5, v/v). TFA

(5.3 mL, 71 mmol) was added dropwise and the resulting reaction

mixture was stirred at reﬂux for 2 h. The reaction was checked for

completion by TLC (CH₂Cl₂–CH₃OH, 9 : 1, v/v) and the mixture

was evaporated to dryness. The resulting oily residue was dissolved

in deionised water and lyophilised to give H-D-Glu(Ala-OMe)-

OMe 13 as a yellow oil (320 mg, 0.89 mmol, quantitative yield).

This compound was used in the next reaction step without further

puriﬁcation.

1 : 1 mixture of two diastereomers). ¹H NMR (300 MHz, CD₃OD):

d 0.82–0.86 (3H), 1.00–1.07 (3H), 1.35–1.39 (3H), 1.45–1.47 (3H),

1.80–2.90 (8H), 2.90–2.92 (3H), 3.19–3.26 (4H), 4.11–4.16 (1H),

4.25–4.39 (2H), 4.90 (1H, partially masked by water peak), 5.34–

5.47 (1H), 5.54–5.72 (1H), 6.19–6.25 (1H), 7.13–7.24 (5H). The

complexity of the spectrum for this mixture of two diastereomers

makes a complete assignment impossible. MS (ESI+): m/z 546.87

[M + H]⁺, 569.00 [M + Na]⁺, calcd for C₂₉H₄₂N₂O₈: 546.67; HPLC

(system A): t_R= 13.7 min and 14.0 min (two diastereomers), purity

>95%.

Acyclic microcystin analogue (15)

(a) Trans-amidiﬁcation: Aldol 10 (50 mg, 0.95 mmol) and TFA salt

of dipeptide H-Glu(Ala-OMe)-OMe 13 (38 mg, 0.7 mmol) were

dissolved in dry THF (3 mL). Imidazole (15 mg, 0.2 mmol) and

DMAP (6.1 mg, 50 mmol) were sequentially added. The resulting

mixture was stirred at room temperature for 12 h. The reaction was

checked for completion by TLC (cyclohexane–AcOEt, 6 : 4, v/v).

Thereafter, the solvent was evaporated under reduced pressure.

The resulting residue was dissolved in a mixture of CH₂Cl₂–Et₂O

(20 mL, 1 : 1, v/v) and washed with aq. 10% citric acid (25 mL) and

brine (25 mL). The organic layer was dried over Na₂SO₄, ﬁltrated

and evaporated to dryness. The resulting residue was puriﬁed

by chromatography on a silica gel column with a mixture of

cyclohexane–AcOEt (6 : 4, v/v), then CH₂Cl₂–CH₃OH (96 : 4, v/v)

as the mobile phase to give the methyl diester of the targeted MC

analogue as a white amorphous powder (21 mg, 0.37 mmol, yield

39%). R_f(cyclohexane–AcOEt, 6 : 4, v/v) 0.15. (b) Saponiﬁcation:

Methyl diester (21 mg, 37 mmol) was dissolved in CH₃OH (2 mL).

Aq. 1.0 M LiOH (0.55 mL, 550 mmol) was added dropwise. The

resulting reaction mixture was stirred at room temperature for 12 h.

The reaction was checked for completion by RP-HPLC (system

A) and quenched by addition of acetic acid (0.6 mL, 696 mmol).

Thereafter, the reaction mixture was evaporated to dryness. The

resulting residue was dissolved with a mixture of 0.1% aq. TFA

and CH₃CN (1 : 1, v/v, 4 mL) and puriﬁed by semi-preparative RP-

HPLC (system C, 2 injections, t_R= 16.6-20.4 min). The product-

containing fractions were lyophilised to give microcystin analogue

15 as a white amorphous powder (15.7 mg, 29.5 mmol, yield 80%).

¹H NMR (300 MHz, CD₃OD): d 1.01 (d, J = 6.6 Hz, 3H), 1.20 (d,

J = 7.0 Hz, 3H), 1.38 (d, J = 7.3 Hz, 3H), 1.63 (s, 3H), 1.82–2.88

(m, 8H), 3.20–3.25 (m, 4H), 4.16 (t, J = 7.5 Hz, 1H, CH), 4.30–

4.44 (m, 2H), 4.87 (1H, partially masked by water peak), 5.41 (d,

J = 9.9 Hz, 1H), 5.57–5.64 (dd, J = 7.3 Hz, 8.3 Hz, 1H), 6.24

(d, J = 15.7 Hz, 1H), 7.17–7.25 (m, 5H); MS (ESI-): m/z 531.47

[M - H]^-, 1063.20 [2M - H]^-, calcd for C₂₈H₄₀N₂O₈: 532.64; HPLC

(system A): t_R= 17.6 min, purity 94%.

Acyclic microcystin analogue (14)

(a) Trans-amidiﬁcation: Aldol 10 (206 mg, 0.55 mmol) and

TFA salt of dipeptide H-D-Glu(MeAla-OMe)-OMe 12 (145 mg,

0.39 mmol) were dissolved in dry THF (12 mL). Imidazole (68 mg,

0.86 mmol) and DMAP (4.8 mg, 40 mmol) were sequentially added.

The resulting reaction mixture was stirred at room temperature

for 12 h. The reaction was checked for completion by TLC

(cyclohexane–AcOEt, 6 : 4, v/v). Thereafter, the solvent was

evaporated under reduced pressure. The resulting residue was

dissolved in a mixture CH₂Cl₂–Et₂O (20 mL, 1 : 1, v/v). This

organic layer was washed with aq. 10% citric acid (25 mL) and

brine (25 mL), dried over Na₂SO₄, ﬁltered and evaporated to

dryness. The resulting residue was puriﬁed by chromatography

on a silica gel column with a mixture of cyclohexane–AcOEt

(6 : 4, v/v), then CH₂Cl₂–CH₃OH (96 : 4, v/v) as the mobile

phase to give the methyl diester of the targeted MC analogue

(165 mg, 0.29 mmol, yield 52%) as a white amorphous powder.

R_f(cyclohexane–AcOEt, 6 : 4, v/v) 0.15; IR (KBr): n_max666, 701,

752, 968, 1030, 1083, 1214, 1374, 1455, 1495, 1539, 1645, 1742,

2932, 3389 (broad); ¹H NMR (300 MHz, CDCl₃): d 1.03 (d, J =

6.8 Hz, 3H), 1.15 (d, J = 7.0 Hz, 3H), 1.39 (d, J = 7.3 Hz, 3H),

1.64 (s, 3 H), 1.83–1.88 (m, 2H), 2.08–2.84 (m, 6H), 2.93 (s, 3H),

3.17–3.23 (m, 4H), 3.71 (s, 3H), 3.77 (s, 3H), 4.41–4.54 (m, 2H),

5.14 (q, J = 7.3 Hz, 1 H), 5.41 (d, J = 9.9 Hz, 1H), 5.50–5.58 (dd,

J = 6.6 Hz, 9.0 Hz, 1H), 6.29 (d, J = 15.5 Hz, 1H), 7.17–7.29

(m, 5H); ¹³C NMR (75.5 MHz, CDCl₃): d 11.8, 13.2, 15.0, 16.7,

26.1, 26.7, 30.6, 32.2, 37.2, 38.8, 45.9, 52.8, 52.9, 53.0, 53.1, 59.2,

74.2, 87.4, 126.4, 126.5, 128.7, 130.0, 133.2, 136.6, 137.3, 139.9,

172.6, 172.9, 173.3, 176.6; MS (ESI+): m/z 575.00 [M + H]⁺,

calcd for C₃₁H₄₆N₂O₈: 574.72. (b) Saponiﬁcation: Methyl diester

(50 mg, 87 mmol) was dissolved in CH₃OH (3 mL). Aq. 1.0 M

LiOH (1.3 mL, 1.3 mmol) was added dropwise. The resulting

reaction mixture was stirred at room temperature for 30 min. The

reaction was checked for completion by RP-HPLC (system A).

The reaction was quenched by addition of acetic acid (0.6 mL,

696 mmol). Thereafter, the mixture was evaporated to dryness. The

resulting residue was dissolved with a mixture of 0.1% aq. TFA

and CH₃CN (1 : 1, v/v, 4 mL) and puriﬁed by semi-preparative RP-

HPLC (system C, 3 injections, t_R= 18.6–21.2 min). The product-

containing fractions were lyophilised to give microcystin analogue

14 as a white amorphous powder (39 mg, 71.6 mmol, yield 82%, a

Acyclic microcystin analogue (16)

Aldol 10 (17 mg, 33 mmol) was dissolved in CH₃OH (2 mL). Aq.

1.0 M LiOH (0.36 mL, 360 mmol) was added and the resulting

reaction mixture was stirred at room temperature for 1 h. The

reaction was checked for completion by RP-HPLC (system A)

and acidiﬁed to pH ~3 by addition of acetic acid. Thereafter,

volatiles were removed by evaporation under reduced pressure.

The resulting residue was dissolved with a mixture of 0.1% aq. TFA

and CH₃CN (1 : 1, v/v, 4 mL) and puriﬁed by semi-preparative

This journal is

The Royal Society of Chemistry 2010

Org. Biomol. Chem., 2010, 8, 676–690 | 687

©

View Article Online

RP-HPLC (system D, 2 injections, t_R= 32.9-33.7 min). The

product-containing fractions were lyophilised to give microcystin

analogue 16 as a white amorphous powder (7.6 mg, 23.1 mmol,

yield 70%). ¹H NMR (300 MHz, CDCl₃): d 1.04 (d, J = 6.8 Hz,

3H), 1.20 (d, J = 7.1 Hz, 3H), 1.64 (s, 3H), 2.59–2.84 (m, 4H),

3.18–3.24 (m, 4H), 4.47–4.51 (m, 1H), 5.45 (d, J = 9.8 Hz, 1H),

5.51–5.58 (dd, J = 15.7 Hz, 7.1 Hz, 1H), 6.31 (d, J = 15.5 Hz,

1H), 7.18–7.30 (m, 5 H); MS (ESI-): m/z 331.47 [M - H]^-, 663.27

Adda

N-Boc-Adda (18 mg, 42 mmol) was dissolved in CH₂Cl₂(2 mL).

TFA (124 mL, 1.7 mmol) was added dropwise and the resulting

reaction mixture was stirred at room temperature for 2 h. The

reaction was checked for completion by RP-HPLC (system A) and

the reaction mixture was evaporated to dryness. The resulting oily

residue was puriﬁed by semi-preparative RP-HPLC (system G, 2

injections, t_R= 18.0-23.9 min). The product-containing fractions

were lyophilised to give Adda as a TFA salt (15.6 mg, 35 mmol,

yield 84%). ¹H NMR (300 MHz, CD₃CN): d 1.00 (d, J = 6.8 Hz,

3H), 1.18 (d, J = 7.1 Hz, 3H), 1.63 (s, 3H), 2.64–2.83 (m, 4H),

2.51–2.75 (m, 4H), 3.20 (s, 3H), 3.90 (t, J = 9.0 Hz, 1H), 5.53–5.56

(m, 2H), 6.40 (d, J = 15.6 Hz, 1H), 7.19–7.27 (m, 5H); ¹³C NMR

(75.5 MHz, CD₃CN): d 12.4, 14.5, 15.1, 16.0, 29.6, 36.7, 38.2, 58.6,

86.9, 126.2, 128.4, 129.5, 132.2, 139.3, 139.7, 145.4; MS (ESI+):

m/z 332.00 [M + H]⁺, calcd for C₂₀H₂₉NO₃: 331.46; HPLC (system

A): t_R= 17.2 min, purity 89%.

[2M - H]^-, calcd for C₂₀H₂₈O₄: 332.44; HPLC (system A): t_R

21.6 min, purity >95%.

=

TFA salt of tripeptide H-Adda-iso-D-Glu(Ala)-OH (17)

(a) Peptide coupling: N-Boc-Adda (9.0 mg, 21 mmol) was dissolved

in dry DMF (800 mL). Uronium salt-based coupling reagent

HATU (9.5 mg, 25 mmol) and 2,4,6-collidine (5 mL, 63 mmol)

were sequentially added and the resulting reaction mixture was

stirred at room temperature. The complete conversion of N-Boc-

Adda into the corresponding HOAt active esters was checked

by RP-HPLC (system A). Then, TFA salt of H-Glu(Ala-OMe)-

OMe 13 (9.1 mg, 63 mmol) was added and the reaction mixture

was stirred at room temperature. The reaction was checked for

completion by RP-HPLC (system A), the disappearance of HOAt

active esters was observed but the coupling product seemed to

remain adsorbed on the C₁₈stationary phase. Thereafter, AcOEt

(5 mL) was added and the crude reaction mixture was sequentially

washed with aq. 10% citric acid (5 mL), aq. sat. NaHCO₃(25 mL)

and brine (5 mL). The organic layer was dried over Na₂SO₄,

ﬁltrated and evaporated to dryness. The crude product was used

in the next step without further puriﬁcation. (b) Boc removal: The

crude fully-protected tripeptide Boc-Adda-iso-D-Glu(Ala-OMe)-

OMe was dissolved in CH₂Cl₂(1 mL). TFA (74 mL, 100 mmol) was

added dropwise and the resulting reaction mixture was stirred at

reﬂux for 2 h. The reaction was checked for completion by ESI

mass spectrometry and the mixture was evaporated to dryness.

The resulting oily residue was dissolved in deionised water and

lyophilised to give TFA salt of H-Adda-D-Glu(Ala-OMe)-OMe.

This product was used in the next step without further puriﬁcation.

MS (ESI+): m/z 560.07[M+ H]⁺, calcd for C₃₅H₅₃N₃O₉:559.71. (c)

Saponiﬁcation: Crude tripeptide H-Adda-iso-D-Glu(Ala-OMe)-

OMe was dissolved in CH₃OH (1 mL). Aq. 1.0 M LiOH (200

mL) was added and the resulting reaction mixture was stirred

at room temperature for 1 h. The reaction was checked for

completion by ESI mass spectrometry and acidiﬁed to pH ~3

by addition of acetic acid. Thereafter, volatiles were removed by

evaporation under reduced pressure. The resulting residue was

dissolved with a mixture of CH₃CN and 0.1% aq. TFA (1 : 2, v/v,

3 mL) and puriﬁed by semi-preparative RP-HPLC (system E, 2

injections, t_R= 19.8–20.9 min). The product-containing fractions

were lyophilised to give the microcystin analogue 17 as a white

amorphous powder (3.3 mg, 5 mmol, overall yield for the three

steps 24%). ¹H NMR (300 MHz, CD₃OD): d 0.93 (d, J = 6.8 Hz,

3H), 1.11 (d, J = 7.0 Hz, 3H), 1.19–1.29 (m, 9H), 1.57 (s, 3H),

1.89–1.96 (m, 2H), 2.20–2.27 (m, 2H), 2.51–2.75 (m, 4H), 3.15 (s,

3H), 3.79 (t, J = 8.9 Hz, 1H), 4.16–4.25 (m, 2H), 5.36–5.52 (dd,

J = 15.6 Hz, 8.9 Hz, 1H), 5.50 (d, J = 9.8 Hz, 1H), 6.38 (d, J =

15.6 Hz, 1H), 7.08-7.10 (m, 5H); MS (ESI+): m/z 532.13 [M + H]⁺,

calcd for C₂₈H₄₁N₃O₇: 531.65; HPLC (system A): t_R= 15.3 min,

purity 100%.

Adda-tripod conjugate (19)

N-Hydroxysuccinimidyl carbamate cross-linking reagent 18

(5.0 mg, 5.4 mmol) was dissolved in dry NMP (200 mL). A solution

of Adda (TFA salt, 3.1 mg, 7 mmol) and dry DIEA (1.8 mL,

11 mmol) was prepared in dry NMP (400 mL) and added to the

NMP solution containing active carbamate. The resulting reaction

mixture was stirred at room temperature for 2 h. The reaction

was checked for completion by RP-HPLC (system A). A mixture

of 0.1% aq. TFA and CH₃CN (9 : 3, v/v, 2 mL) was added and

the resulting solution was puriﬁed by semi-preparative RP-HPLC

(system E, 2 injections) The product-containing fractions were

lyophilised to give peptide conjugate 19 as a white amorphous

powder (3.0 mg, 2.6 mmol, yield 48%). MS (ESI+): m/z 1193.13

[M + O + H₂O + H]⁺, calcd for C₅₄H₇₈N₈O₁₆S₂1158.49. Under

our ionisation conditions, both oxidation of one double bond of

Adda moiety (C4–C5) or (C6–C7) and hydration of phthalimide

moiety were observed. HPLC (system A): t_R= 22.1 min, purity

97%.

R

FluoProbes^ꢀ532A labelled Adda-tripod conjugate (20)

(a) Removal of the phthaloyl group: Peptide conjugate 19 (3 mg,

2.6 mmol) was dissolved in CH₃OH (500 mL) and a solution of

hydrazine monohydrate (3.9 mL) in CH₃OH (45 mL) was added.

The reaction was stirred at room temperature for 2 h and checked

for completion by RP-HPLC (system A). The reaction ﬂask was

stored at -20 ^◦C overnight. Thereafter, 0.1% aq. TFA (pH 2.2,

2 mL) was added and the resulting solution was puriﬁed by

semi-preparative RP-HPLC (system F, 2 injections). The product-

containing fractions were lyophilised to give the free aminooxy

tripod as a white amorphous powder. MS: see ESI†. (b) Removal

of the SEt group: Free aminooxy tripod (2.3 mg, 2 mmol) was

dissolved in aq. 0.1 M NaHCO₃buffer (pH 8.5, 300 mL). A

solution of DTT (12 mg, 78 mmol) in aq. 0.1 M NaHCO₃buffer

(pH 8.5, 200 mL) was added and the resulting reaction mixture

was stirred at room temperature for 1 h. The reaction was checked

for completion by RP-HPLC (system A). 0.1% aq. TFA (pH 2.2,

2 mL) was added and the resulting aq. solution was puriﬁed by

semi-preparative RP-HPLC (system H, 1 injection). The product-

containing fractions were lyophilised to give the free sulfhydryl

688 | Org. Biomol. Chem., 2010, 8, 676–690

This journal is

The Royal Society of Chemistry 2010

©

View Article Online

tripod as a white amorphous powder. MS: see ESI†. (c) Fluorescent

labeling: Free sulfhydryl tripod (0.5 mg, 0.6 mmol) was dissolved

in aq. 0.1 M NaHCO₃buffer (pH 8.5, 200 mL). A solution of

2 R. M. Dawson, Toxicon, 1998, 36, 953.

3 R. Jayaraj and P. V. L. Rao, Toxicon, 2006, 48, 272.

4 (a) O. Allis, J. Dauphard, B. Hamilton, A. NiShuilleabhain, M.

Lehane, K. J. James and A. Furey, Anal. Chem., 2007, 79,

3436; (b) J. Meriluoto, Anal. Chim. Acta, 1997, 352, 277; (c) J.

Meriluoto, L. Lawton and K. Harada, Methods Mol. Biol., 2000,

145, 65.

5 (a) I. R. Falconer, in Algal Toxins in Seafood Drinking Water, ed.

I. R. Falconer, Academic Press, London, 1993, p. 165; (b) G.-B. Trogen,

A. Annila, J. Eriksson, M. Kontteli, J. Meriluoto, I. Sethson, J. Zdunek

and U. Edlund, Biochemistry, 1996, 35, 3197; (c) L. H. Xu, P. K. S.

Lam, J. P. Chen, J. M. Xu, B. S. F. Wong, Y. Y. Zhang, R. S. S. Wu and

K. I. Harada, Chemosphere, 2000, 41, 53.

6 (a) C. Rivasseau, P. Racaud, A. Deguin and M.-C. Hennion, Anal.

Chim. Acta, 1999, 394, 243; (b) A. Zeck, M. G. Weller and R. Niessner,

Anal. Chem., 2001, 73, 5509.

7 (a) J. S. Metcalf, S. G. Bell and G. A. Codd, Water Res., 2000, 34, 2761;

(b) P. Lindner, R. Molz, E. Yacoub-George, A. Du¨rkop and H. Wolf,

Anal. Chim. Acta, 2004, 521, 37.

R

FluoProbes 532A^ꢀmaleimide labeling reagent (0.5 mg, 0.7 mmol)

in dry CH₃CN (200 mL) was added and the resulting reaction

mixture was protected from light and stirred at room temperature

for 1 h. The labeling reaction was checked for completion by RP-

HPLC (system B). Further amount of aq. 0.1 M NaHCO₃buffer

(pH 8.5, 200 mL) was added. Finally, 0.1% aq. TFA (pH 2.2,

2 mL) was added and the resulting aq. solution was puriﬁed by

semi-preprative RP-HPLC (system H, 1 injection). The product-

containing fractions were lyophilised to give ﬂuorescent tripod

20 as a pink amorphous powder. Quantiﬁcation was achieved

by UV-visible measurements at l_max= 532 nm of FluoProbes

R

532A^ꢀby using the e value 115 000 M^-1cm^-1(overall yield

for the three steps estimated after RP-HPLC puriﬁcation 5%).

MS (ESI+): m/z 1769.40 [M + O + H₂O + H]⁺. Expected

molecular ion [M + H]⁺of 20 at m/z 1737.49 was not observed.

Under our ionisation conditions, both oxidation of one double

bond of Adda moiety (C4–C5) or (C6–C7) and hydration of

maleimide moiety were observed. HPLC (system B): t_R= 29.4 min,

purity 81%.

8 J. McElhiney, M. Drever, L. A. Lawton and A. J. Porter, Appl. Environ.

Microbiol., 2002, 68, 5288.

9 (a) I. Chianella, M. Lotierzo, S. A. Piletsky, I. E. Tothill, B. Chen,

K. Karim and A. P. F. Turner, Anal. Chem., 2002, 74, 1288; (b) I.

Chianella, S. A. Piletsky, I. E. Tothill, B. Chen and A. P. F. Turner,

Biosens. Bioelectron., 2003, 18, 119.

10 J. McElhiney and L. A. Lawton, Toxicol. Appl. Pharmacol., 2005, 203,

219.

11 (a) Y. M. Kim, S. W. Oh, S. Y. Jeong, D. J. Pyo and E. Y. Choi,

Environ. Sci. Technol., 2003, 37, 1899; (b) N. Khreich, P. Lamourette,

B. Lagoutte, C. Ronco, X. Franck, C. Cre´minon, H. Volland, Anal.

Bioanal. Chem., 2010, in press.

Functionalisation and peptide immobilisation on silica surfaces

12 H. Volland, L.-M. Neuburger, E. Schultz, J. Grassi, F. Perraut and C.

Cre´minon, Anal. Chem., 2005, 77, 1896.

13 G. Clave´, H. Boutal, A. Hoang, F. Perraut, H. Volland,

P.-Y. Renard and A. Romieu, Org. Biomol. Chem., 2008, 6,

3065.

14 (a) M. Namikoshi, K. L. Rinehart, A. M. Dahlem, V. R. Beasley

and W. W. Carmichael, Tetrahedron Lett., 1989, 30, 4349; (b) M. F.

Beatty, C. Jennings-White and M. A. Avery, J. Chem. Soc., Perkin

Trans. 1, 1992, 1637; (c) F. D’Aniello, A. Mann and M. Taddei, J. Org.

Chem., 1996, 61, 4870; (d) J. M. Humphrey, J. B. Aggen and A. R.

Chamberlin, J. Am. Chem. Soc., 1996, 118, 11759; (e) H. Y. Kim and

P. L. Toogood, Tetrahedron Lett., 1996, 37, 2349; (f) N. Sin and J.

Kallmerten, Tetrahedron Lett., 1996, 37, 5645; (g) T. Hu and J. S. Panek,

J. Am. Chem. Soc., 2002, 124, 11368; (h) S. Meiries and R. Marquez,

J. Org. Chem., 2008, 73, 5015; (i) S. Meiries, A. Parkin and R. Marquez,

Tetrahedron, 2009, 65, 2951.

(a) Preparation of the aldehyde-functionalised silica surface: This

was achieved from a silicon substrate (75 ¥ 25 ¥ 1 mm) doped with

a thick SiO₂layer (thickness 500 nm) by using experimental condi-

tions already reported by us.¹³(b) Immobilisation of ﬂuorescent la-

belled “microcystin-tripod” 20: This ﬂuorescent aminooxy pseudo-

peptide was prepared as a 10 mM solution in immobilisation buffer

(potassium phosphate 0.1 M, pH 7.4) and 1 mL of the resulting

solution was put over the freshly prepared aldehydic surface. After

4 h of incubation in the dark and in a humid atmosphere at

room temperature, the reaction solution was carefully removed

and the surface was washed for 10 min with phosphate buffer

(10 mM + 0.05% Tween, pH 7.4). Finally, slide was stored with EIA

buffer (0.1 M phosphate buffer (pH 7.4) containing 0.15 M NaCl,

0.1% BSA and 0.01% sodium azide) overnight before observation.

Fluorescence scanning was performed with an Olympus inverted

microscope model IX71 (4X objective) equipped with a camera

PCO 1600, and the resulting images were analysed using ImageJ

software (http://rsbweb.nih.gov/ij/).

15 C. Pearson, K. L. Rinehart, M. Sugano and J. R. Costerison, Org. Lett.,

2000, 2, 2901.

16 N. Khreich, P. Lamourette, P.-Y. Renard, G. Clave´, F. Fenaille, C.

Cre´minon and H. Volland, Toxicon, 2009, 53, 551.

17 J. Patel, G. Clave´, P.-Y. Renard and X. Franck, Angew. Chem., Int. Ed.,

2008, 47, 4224.

18 Y. Wu, Y.-P. Sun, Y.-Q. Yang, Q. Hu and Q. Zhang, J. Org. Chem.,

2004, 69, 6141.

19 M. Crimmins and J. She, Synlett, 2004, 1371.

Acknowledgements

20 D. Delaunay, L. Toupet and M. L. Corre, J. Org. Chem., 1995, 60,

6604.

This work was supported by the Agence Nationale de la Recherche

(Programme Emergence et Maturation 2005, ANR-2005-EMPB-

027-04), Commissariat a` l’Energie Atomique, and Institut Uni-

versitaire de France (IUF). We thank the “Re´gion Haute-

Normandie” for ﬁnancial support via the CRUNCh program

(CPER 2007-2013). We thank Elisabeth Roger (INSA de Rouen)

for IR measurements, Annick Leboisselier (INSA de Rouen) for

the determination of elemental analyses and QUIDD company

for the open access to their HPLC instruments.

21 (a) Y. Nagao, T. Miyasaka, K. Seno, M. Yagi and E. Fujita, Chem.

Lett., 1981, 463; (b) C. H. Li, Y. H. Yieh, Y. Lin, Y. J. Lu, A. H. Chi

and C. Y. Hsing, Tetrahedron Lett., 1981, 22, 3467.

22 For a recent publication about the use of novel thiazolidine-2-thione

coupling chemistry in the preparation of protein-polymer conjugates,

see: L. Tao, J. Liu, J. M. Xu and T. P. Davis, Org. Biomol. Chem., 2009,

7, 3481.

23 J. Coste, E. Frerot and P. Jouin, J. Org. Chem., 1994, 59, 2437.

24 For

a comprehensive review on peptide coupling reagents,

see: E. Valeur and M. Bradley, Chem. Soc. Rev., 2009, 38,

606.

25 (a) J. R. McDermott and N. L. Benoiton, Can. J. Chem., 1973,

51, 2555; (b) S. T. Cheung and N. L. Benoiton, Can. J. Chem.,

1977, 55, 916; E. Biron and H. Kessler, J. Org. Chem., 2005, 70,

5183.

Notes and references

1 M. Welker, M. Brunke, K. Preussel, I. Lippert and H. von Dohren,

Microbiology, 2004, 150, 1785.

26 (a) B. M. Gulledge, J. B. Aggen and A. R. Chamberlin, Bioorg. Med.

Chem. Lett., 2003, 13, 2903; (b) B. M. Gulledge, J. B. Aggen, H. Eng,

This journal is

The Royal Society of Chemistry 2010

Org. Biomol. Chem., 2010, 8, 676–690 | 689

©

View Article Online

K. Sweimeh and A. R. Chamberlin, Bioorg. Med. Chem. Lett., 2003,

13, 2907.

27 K. R. Shreder, Y. Liu, T. Nomanhboy, S. R. Fuller, M. S. Wong, W. Z.

Gai, J. Wu, P. S. Leventhal, J. R. Lill and S. Corral, Bioconjugate Chem.,

2004, 15, 790.

30 D. J. Cundy, A. C. Donohue and T. D. McCarthy, Tetrahedron Lett.,

1998, 39, 5125.

31 For recent reviews about ABPs, see: (a) M. Fonovic and M. Bogyo,

Curr. Pharm. Des., 2007, 13, 253; (b) M. C. Hagenstein and N. Sewald,

J. Biotechnol., 2006, 124, 56.

28 E. J. A. Oliveira, S. P. V. Nova, S. Alves, Jr., P. Santa-Cruz, R. J. R.

Molica, A. Teixeira, E. Malageno and J. L. Lima Filho, J. Braz. Chem.

Soc., 2006, 17, 243.

32 M. J. McKennon, A. I. Meyers, K. Drauz and M. Schwarm, J. Org.

Chem., 1993, 58, 3568.

33 H. E. Gottlieb, V. Kotlyar and A. Nudelman, J. Org. Chem., 1997, 62,

7512.

29 O. A. Sadik and F. Yan, Chem. Commun., 2004, 1136.

690 | Org. Biomol. Chem., 2010, 8, 676–690

This journal is

The Royal Society of Chemistry 2010

©

Article Doi

Facile and rapid access to linear and truncated microcystin analogues for the implementation of immunoassays

DOI: 10.1039/b920193a

Source and publish data:

Authors:

Article abstract of DOI:10.1039/b920193a

Full text of DOI:10.1039/b920193a

Products guided by the article

R&D Labs maybe for 1217320-41-6

Relevant to this article

Hot Product