Synthesis and Hydrolysis of 4-Chloro-PyMTA and 4-Iodo-PyMTA Esters and Their Oxidative Degradation with Cu(I/II) and Oxygen

DOI: 10.1055/s-0035-1561660

Source and publish data:

Synthesis p. 3773 - 3784 (2016)

Update date:2022-08-04

Topics:

Authors:

Qi, Mian

Hülsmann, Miriam

Godt, Adelheid

Read Full Text PDF DownLoad Join now for total 90,000,000 free articles

Article abstract of DOI:10.1055/s-0035-1561660

We disclose the syntheses of ethyl and tert-butyl esters of 4-chloro-PyMTA and 4-iodo-PyMTA from the commercially available chelidamic acid monohydrate in 39-67% overall yield. Additionally, ester hydrolyses with aqueous NaOH (ethyl esters) or trifluoroacetic acid (tert-butyl esters) are reported. The resulting materials contain 4-halo-PyMTA in mixture with partially deprotonated or partially protonated 4-halo-PyMTA. The ligand content expressed as the content of the common structural motifs of the present species, namely [PyMTA - 4 H⁺^4- (basic hydrolysis) and PyMTA (acidic hydrolysis), was determined to be 90-94 wt % by¹H NMR spectroscopy using maleic acid as an internal standard. The tert-butyl esters were easily hydrolyzed with aqueous alkali hydroxide, with a decreasing rate in the series NaOH, KOH, LiOH. This finding indicates a Lewis acid assisted ester cleavage with the Na⁺ ion fitting best to the multidentate ligand. Unexpectedly, PyMTA esters are incompatible with Cu(I/II) salts in the presence of oxygen. Under these conditions, one of the two aminomethyl groups is converted into a formyl group. This reaction not only limits the application of Cu(I/II)-catalyzed reactions but also necessitates trapping of any copper ions (e.g., with a metal ion scavenger) before the material is exposed to oxygen.

Full text of DOI:10.1055/s-0035-1561660

SYNTHESIS

2016, 48, A–L

paper

M. Qi et al.

Paper

Synthesis

Synthesis and Hydrolysis of 4-Chloro-PyMTA and 4-Iodo-PyMTA

Esters and Their Oxidative Degradation with Cu(I/II) and Oxygen

Mian Qi

Miriam Hülsmann

H₂O chelidamic acid monohydrate

Adelheid Godt*

Faculty of Chemistry and Center for Molecular Materials (CM₂),

Bielefeld University, Universitätsstraße 25, 33615 Bielefeld, Ger-

many

godt@uni-bielefeld.de

X = Cl: 4 steps; X = I: 5 steps

CO₂R

CO₂H

Cu(I/II)

O₂

CO₂R

ester-aldehyde

4-halo-PyMTA ester

4-halo-PyMTA

X = Cl, I

X = Cl, I; R = Et, ^tBu

yield: 39–67%

incompatibility of PyMTA

ester with Cu(I/II) in air

yield: 85–95%

Received: 03.03.2016

Accepted after revision: 02.05.2016

Published online: 29.06.2016

Dy(III), Tm(III), and Yb(III) are used as paramagnetic labels

for nuclear magnetic resonance spectroscopy,^22,23and its

complexes with Eu(III), Tb(III), Sm(III), Nd(III), and Yb(III)

are widely used because of their fluorescence^1–13for time-

resolved fluorescence spectroscopy,^8,26fluorescence reso-

nance energy transfer,^30–35quenching resonance energy

transfer,²⁵immunoassays,^8,36–39protein quantification,^28,34

minisequencing,⁴⁰ligand binding assays,²⁶and DNA analy-

sis.^41,42

DOI: 10.1055/s-0035-1561660; Art ID: ss-2016-z0156-op

Abstract We disclose the syntheses of ethyl and tert-butyl esters of 4-

chloro-PyMTA and 4-iodo-PyMTA from the commercially available che-

lidamic acid monohydrate in 39–67% overall yield. Additionally, ester

hydrolyses with aqueous NaOH (ethyl esters) or trifluoroacetic acid

(tert-butyl esters) are reported. The resulting materials contain 4-halo-

PyMTA in mixture with partially deprotonated or partially protonated 4-

halo-PyMTA. The ligand content expressed as the content of the com-

mon structural motifs of the present species, namely [PyMTA – 4 H⁺]^4–

(basic hydrolysis) and PyMTA (acidic hydrolysis), was determined to be

90–94 wt % by ¹H NMR spectroscopy using maleic acid as an internal

standard. The tert-butyl esters were easily hydrolyzed with aqueous al-

kali hydroxide, with a decreasing rate in the series NaOH, KOH, LiOH.

This finding indicates a Lewis acid assisted ester cleavage with the Na⁺

ion fitting best to the multidentate ligand. Unexpectedly, PyMTA esters

are incompatible with Cu(I/II) salts in the presence of oxygen. Under

these conditions, one of the two aminomethyl groups is converted into

a formyl group. This reaction not only limits the application of Cu(I/II)-

catalyzed reactions but also necessitates trapping of any copper ions

(e.g., with a metal ion scavenger) before the material is exposed to oxy-

gen.

Most of the functionalized PyMTAs used in the above-

mentioned studies are prepared from 4-halo-PyMTA esters

by employing either an aromatic nucleophilic substitution

(S_NAr) reaction^13,22,27or

Pd-catalyzed cross-cou-

pling^{6,9,10,16,18–21,23,41–46}(Scheme 1). The most-often found

order of reactivity is aryl chloride > aryl bromide > aryl io-

dide in S_NAr reactions⁴⁷and aryl iodide > aryl bromide >

aryl chloride in cross-couplings.⁴⁸In agreement with this,

when 4-azido-PyMTA ester was prepared via S_NAr reaction

of 4-chloro-PyMTA ester or 4-bromo-PyMTA ester with

NaN₃under comparable conditions, the yields were 100%²⁷

and 48%,¹³respectively, and in the cross-coupling with an

alkyne 4-iodo-PyMTA ester was more reactive than 4-bro-

mo-PyMTA ester, as reported by Takalo and co-workers⁴⁶

and confirmed by our own experiments.⁴⁹Although the 4-

chloro-PyMTA ester and the 4-iodo-PyMTA ester clearly are

of advantage when it comes to reactivity, the 4-bromo-

PyMTA ester has so far been used in most cases because of

its much better accessibility via the procedure disclosed by

Takalo and co-workers. Their well-established route starts

from chelidamic acid and provides 4-bromo-PyMTA ester in

up to 64% yield in four simple steps.⁴⁶The situation is very

different for 4-chloro-PyMTA ester and 4-iodo-PyMTA es-

ter. The synthesis of 4-chloro-PyMTA ester starting from

2,6-bis(bromomethyl)-4-chloropyridine was reported to

Key words dipicolinates, Ln(III) complexes, PyMTA, pyridine-based li-

gands, spin label

2,2′,2′′,2′′′-[(Pyridine-2,6-diyl)bis(methylenenitri-

lo)]tetrakis(acetic acid) (PyMTA) is one of the few ligands

whose Ln(III) complexes possess concurrently attractive

optical^1–13and magnetic^12–23properties, excellent chemical

stability,^12,13,24and biocompatibility.^{5,8,13,16,20,22,25–29}It is the

only ligand we know of whose Ln(III) complexes have al-

ready been applied in a wide scope: Its Gd(III) complex is

used as a spin label for electron paramagnetic resonance

spectroscopy^14–18and as a contrast agent for magnetic reso-

nance imaging.^{12,13,19–21}PyMTA complexes with Tb(III),

M. Qi et al.

Paper

Synthesis

give a yield of 96%;²⁷however, the starting material is very

difficult to obtain commercially and the only report on its

preparation states a yield of 8% with 4-chloro-2,6-dimeth-

ylpyridine as the starting material.⁵⁰The synthesis of 4-

iodo-PyMTA ester was reported as a seven-step route start-

ing from chelidamic acid, giving an overall yield of ca.

21%.^46,51,52Two steps lack a detailed description; moreover,

when we tried to repeat the synthesis, at the stage of con-

verting the hydroxyl groups of (4-iodopyridine-2,6-diyl)di-

methanol (8) into bromo substituents inseparable byprod-

ucts formed, which resulted in inseparable contaminants in

the products of subsequent reactions.

linate (7) via treatment with NaI, acetyl chloride, and ultra-

sound.⁵³Iodo diester 7 was reduced to iodo diol 8, as per

the corresponding chloro diol 3. The hydroxyl groups of

iodo diol 8 were converted into mesylate groups, which

were exchanged for dialkyl nitrilodiacetate groups. In the

following, essential details of the individual synthetic steps

are outlined.

Dimethyl 4-chlorodipicolinate (dimethyl 4-chloropyri-

dine-2,6-dicarboxylate, 2) is a very important compound

not only for the synthesis of 4-halo-PyMTA esters but also

for the synthesis of other pyridine-based compounds.^54–66

Its first reported preparation started from chelidamic acid

with PCl₅and MeOH as the reagents.⁶⁶This procedure was

frequently used and gave yields between 26% and 71%.^59–65,67

Recently, SOCl₂was introduced as an alternative chlorina-

tion reagent for chelidamic acid.^54–57,68Using SOCl₂is bene-

ficial, because the two side products SO₂and HCl easily es-

cape as gases from the reaction mixture. However, the re-

ports are contradictory: Müller and co-workers⁵⁶and

Hamada and co-workers⁵⁵reported to have prepared

chloro diester 2 by treatment of chelidamic acid with SOCl₂

followed by addition of MeOH, whereas Kupai and co-work-

ers⁵⁴and Bradshaw and co-workers⁶⁸reported that dimeth-

yl 4-hydroxydipicolinate formed under these conditions.

Kupai and co-workers converted the dimethyl 4-hydroxy-

dipicolinate into chloro diester 2 in a second, separate step

by treatment with SOCl₂in the presence of a catalytic

amount (‘2 drops’) of DMF.⁵⁴Bourdolle and co-workers⁵⁷

described a one-pot procedure for the synthesis of chloro

diester 2 from chelidamic acid by treatment with SOCl₂and

DMF, followed by the addition of MeOH. Neither experi-

mental details nor the yield was given. We treated cheli-

damic acid monohydrate (1) with SOCl₂in the absence and

in the presence of DMF, distilled off the residual SOCl₂, and

then added MeOH. Isolation of chloro diester 2 was very

simple: it precipitated as pure material from the methano-

lic reaction mixture, whereas dimethyl 4-hydroxydipico-

linate and chelidamic acid remained in solution. Without

DMF only dimethyl 4-hydroxydipicolinate was formed,

whereas chloro diester 2 was formed in the presence of

DMF. Clearly, the Vilsmeier reagent, generated in situ from

SOCl₂and DMF, is needed for the substitution of the hy-

droxyl group at the pyridine ring by a chlorine atom. Fur-

thermore, we found that the amount of DMF played an im-

portant role. Experiments with 0.05–0.2 equivalents of

DMF (relative to chelidamic acid) and a reaction time of 4

hours provided chloro diester 2 in 46–67% yield. The miss-

ing 54–33% was a mixture of dimethyl 4-hydroxydipico-

linate and chelidamic acid. Increasing the reaction time to

21 hours had no effect; however, increasing the amount of

DMF to 2 equivalents raised the yield to 85–87% (3 runs).

For the regular reaction, in which DMF acts as a catalyst, a

small amount of DMF is sufficient. Therefore, our results in-

dicate that DMF decomposes during the reaction. A further

CO₂R

S_NAr

cross-

coupling

4-halo-PyMTA ester

FG = functional group

X = Cl, Br, I

R = ^tBu, Et, Me

Scheme 1 Synthesis of substituted PyMTA esters from 4-halo-PyMTA

esters via S_NAr reaction or cross-coupling

Herein, we report an easy access to 4-chloro-PyMTA es-

ters and 4-iodo-PyMTA esters starting from commercially

available chelidamic acid monohydrate in overall yields of

39–42% and 60–67%, respectively. In addition, we closely

examine the ester hydrolysis. The PyMTA tert-butyl esters

are surprisingly labile towards alkali hydroxides, especially

NaOH. Moreover, basic hydrolysis of the ethyl esters gives

partially deprotonated PyMTA and acidic hydrolysis of the

tert-butyl esters gives partially protonated PyMTA. We dis-

close procedures for the ester hydrolysis that give materials

with a ligand content of a minimum 90 wt % expressed in

terms of the content of the common structural motifs

[PyMTA – 4 H⁺]^4–(basic hydrolysis) and PyMTA (acidic hy-

drolysis), and report a procedure for the quantification. Fi-

nally, we document the surprising lability of PyMTA esters

towards oxidative degradation in the presence of Cu(I/II)

salts and oxygen, and outline countermeasures.

Our routes to 4-chloro-PyMTA esters 6 and 4-iodo-

PyMTA esters 10 are depicted in Scheme 2A. Briefly, cheli-

damic acid monohydrate (1) was treated with thionyl chlo-

ride followed by the addition of MeOH to afford dimethyl 4-

chlorodipicolinate (2), the common intermediate for the

preparation of all the esters. To obtain the 4-chloro-PyMTA

esters 6, chloro diester 2 was reduced with NaBH₄in the

presence of CaCl₂to chloro diol 3. Then, the two hydroxyl

groups of 3 were replaced by chlorine atoms via reaction

with phosphoryl chloride and DMF, and these were subse-

quently exchanged for dialkyl nitrilodiacetate groups. For

preparing the corresponding 4-iodo-PyMTA esters 10,

chloro diester 2 was converted into dimethyl 4-iododipico-

M. Qi et al.

Paper

Synthesis

increase in the amount of DMF did not improve the yield.

Using anhydrous chelidamic acid, which is more expensive

than the hydrate, gave about the same yield.

which precipitated from the solution. The NMR spectra of

the precipitates show only the signals of the diol. However,

the masses of the materials were more than those expected

for a 100% yield. Clearly inorganic matter, possibly Ca(OH)₂,

which is not detectable by NMR spectroscopy, also precipi-

tated. This inorganic matter did not interfere with the sub-

sequent reactions.

To introduce the dialkyl nitrilodiacetate groups, the hy-

droxyl groups of the diols needed to be converted into a

good leaving group. In the case of chloro diol 3, the hydrox-

yl groups were exchanged for chloro substituents: with

POCl₃and DMF, trichloride 4 was obtained in 63% yield

(over 2 steps from 2). The reaction of chloro diol 3 with

PCl₃, in analogy to the conversion of the corresponding

bromo diol into tribromide,⁴⁶gave trichloride 4 in ca. 35%

yield as a mixture with an unidentified product. The report-

ed synthesis of trichloride 4 from chloro diol 3 using mesyl

chloride⁷²was not reproducible in our hands. The conver-

sion of chloro diol 3 was far from complete, even at higher

temperature and for a longer reaction time than given in

the reported procedure. After 6 days at room temperature,

The reduction of dimethyl 4-chlorodipicolinate (2) to

chloro diol 3 and of dimethyl 4-iododipicolinate (7) to

iodo diol 8 have been reported:^46,69,703–5 equivalents of

NaBH₄were used and the reaction mixtures were refluxed

for 15–18 hours; the workup was laborious as a continuous

liquid–liquid extraction was run for several days. NaBH₄in

combination with CaCl₂in MeOH allowed us to decrease

the amount of NaBH₄to 2 equivalents and, additionally, the

reactions were complete after 4–5 hours at room tempera-

ture. The products were simply isolated via precipitation:

The reaction mixture, a suspension, was filtered through

silica gel and the solvent of the filtrate, i.e., MeOH, was re-

moved. 1 M HCl was added to the residue to convert the diol

into the corresponding hydrochloride salt; thereby, the hy-

drochloride salt of the diol as well as the inorganic compo-

nents went into solution. The pH of the solution was raised

to pH 12 by the addition of NaOH, and consequently the hy-

drochloride salt of the diol was converted back into the diol

CO₂R

OMe

CO₂R

(b)

(c)

(d)

CO₂R

(a)

H₂O

(e)

CO₂R

OMs

OMe

CO₂R

(f)

(g)

(h)

CO₂R

^tBu

CO₂R

(i)

OMs

(j)

OMs

Scheme 2 Reagents and conditions: A: Preparation of 4-halo-PyMTA esters. (a) 1) SOCl₂, DMF, reflux, 22 h, 2) MeOH, 87%; (b) NaBH₄, CaCl₂, MeOH, r.t.,

4 h; (c) POCl₃, DMF, 50 °C, 40 min, 63% over 2 steps; (d) Na₂CO₃, KI, MeCN, r.t., 23 h, 71–76%; (e) NaI, AcCl, MeCN, ultrasound, r.t., 30 min, 94%; (f)

NaBH₄, CaCl₂, MeOH, r.t., 5 h; (g) MsCl, Et₃N, CH₂Cl₂, 0 °C, 10 min, then r.t., 15 min (a mixture of iodo dimesylate 9, monochloride 12, and

iodo dichloride 13 was obtained; see Scheme 2C); (h) Na₂CO₃, KI, MeCN, r.t., 20 h, 10a (73% over 3 steps), 10b (82% over 3 steps). B: Preparation of

iodo dibromide 11 from iodo diol 8. (i) PBr₃, CHCl₃, reflux, 8 h. C: Mesylation of iodo diol 8. (j) MsCl, Et₃N, CH₂Cl₂, 0 °C, 10 min, then r.t., 15 min, 9 (69%),

12 (5%), and 13 (2%), yields over 2 steps from 7.

M. Qi et al.

Paper

Synthesis

only 17–38% (4 runs) of chloro diol 3 had reacted and the

products were the chloro dimesylate, the dichloride, and

trichloride 4. Raising the temperature to 40 °C and extend-

ing the reaction time to 9 days gave a conversion of only

25%. Most probably, the low solubility of chloro diol 3 in

CH₂Cl₂hindered the reaction; supporting this assumption,

mesylation of the much better soluble iodo diol 8 was pos-

sible (see below). The precipitating triethylamine hydro-

chloride may also have contributed to the low reactivity by

coating the undissolved chloro diol 3.

change resin.^{4,13,16,23,42,73}The combination of LiOH with HCl

was applied in the preparation of PyMTA-labeled lipids,

which are insoluble in water, so that the water-soluble by-

product LiCl could be easily removed by washing with wa-

ter.⁴The combination of Ba(OH)₂with H₂SO₄was applied in

the preparation of water-soluble derivatives of PyMTA. In

this case, the water-insoluble byproduct BaSO₄was re-

moved by filtration.⁷³It is very doubtful that the Ba²⁺ions

could be completely removed because chelate ligands with

aminoacetate groups [e.g., diethylenetriaminepentaacetic

acid (DTPA) and ethylenediaminetetraacetic acid (EDTA)]

are strong complexing agents for the Ba²⁺ion and their

aqueous solutions can even dissolve BaSO₄.⁷⁴PyMTA is

structurally very similar to EDTA and therefore we expect

its aqueous solution to dissolve BaSO₄. PyMTA contaminat-

ed with Ba²⁺ions may be problematic in the case of biologi-

cal applications considering the high toxicity of Ba²⁺ions.

The combination of NaOH with proton-exchange resin has

been recently used.^13,16,23The advantage of this combina-

tion is that the only byproduct, the Na⁺-loaded resin, is re-

moved simply by filtration. We applied this procedure,

namely treatment with NaOH, then acidification with a

proton-exchange resin, to 4-chloro-PyMTA ethyl ester 6a

and 4-iodo-PyMTA ethyl ester 10a in order to obtain 4-

chloro-PyMTA and 4-iodo-PyMTA (Scheme 3A). The first

step, the basic ester hydrolysis, proceeded smoothly,

whereas the second step, the protonation with the proton-

exchange resin, was not without pitfalls. We found that the

amount of the proton-exchange resin needs to be carefully

controlled. We added just enough needed to reduce the pH

of the solution to pH 3; adding more resin barely changed

the pH of the solution, but drastically decreased the yield.

Obviously, protonation of the amino groups of 4-halo-PyM-

TA occurs and the corresponding protonated 4-halo-PyMTA

is trapped as a counterion of the resin. The adhesion of 4-

halo-PyMTA to the resin was reversible upon addition of

NaOH. Minimizing the loss of 4-halo-PyMTA resulted in an

incomplete exchange of Na⁺for H⁺, such that mixtures of 4-

halo-PyMTA and partially deprotonated 4-halo-PyMTA as

sodium salts, {[4-halo-PyMTA – 4 H⁺]^4–·n H⁺·m Na⁺} 14 and

15, were obtained. Quantitative ¹H NMR spectroscopy⁷⁵re-

vealed the ligand content, which is expressed as the

amount of the motif [4-halo-PyMTA – 4 H⁺]^4–that is com-

mon to all species, to be 90–92 wt% (for details, see Sup-

porting Information). The content varied slightly from

batch to batch. Therefore, whenever uncomplexed Ln(III)

ions are to be avoided, e.g., because of their high toxicity⁷⁶

and interference with physical measurements, it is neces-

sary to quantify the amount of ligand in the individual ma-

terials, so that the stoichiometric amount of Ln(III) can be

added.

For the conversion of iodo diol 8 into the iodo-PyMTA

esters 10, we tried to prepare iodo dibromide 11 by reflux-

ing iodo diol 8 with PBr₃in CHCl₃according to a published

procedure.⁴⁶Unfortunately, the halogen atom at the pyri-

dine ring and the two benzylic halogen atoms interchanged

during the reaction and a mixture of all possible halides

was the result (Scheme 2B; for details, see Supporting In-

formation). The desired iodo dibromide 11 was the major

product, but the other products could not be removed by

either column chromatography or crystallization. Using this

mixture as the starting material to prepare the PyMTA ester

gave an inseparable mixture of 4-iodo-PyMTA ester and 4-

bromo-PyMTA ester. Additionally, the compounds with an

iodomethyl group decompose slowly in solution, a decom-

position which starts with oxidation of the iodomethyl

group to a formyl group. Having failed with a clean forma-

tion of iodo dibromide 11, we turned to the mesylation of

iodo diol 8 (Scheme 2C). Reactions of iodo diol 8 with mesyl

chloride in CH₂Cl₂at room temperature gave mixtures of

iodo dimesylate 9, monochloride 12, and iodo dichloride

13, with an iodo dimesylate 9 content of 69 to 93 mol% (3

runs). Lowering the temperature to ice bath temperature in

order to gain higher selectivity resulted in an incomplete

conversion. The three products are easily separable via

chromatography, but when 4-iodo-PyMTA esters 10 are the

goal, there is no need for a separation (see below).

To finally obtain 4-chloro-PyMTA esters 6 and 4-iodo-

PyMTA esters 10, we treated trichloride

and

iodo dimesylate 9, or the mixture of iodo dimesylate 9,

monochloride 12, and iodo dichloride 13, with dialkyl imi-

nodiacetates 5. The reactivities of the benzylic chloride and

of the benzylic mesylate groups were quite different. Sub-

stitution of the mesylate group in MeCN using Na₂CO₃as

base went smoothly at room temperature. The benzylic

chloride is inert under these conditions, and a reaction at

room temperature was accomplished only upon the addi-

tion of KI.

The reported procedures for the hydrolyses of PyMTA

ethyl ester and PyMTA methyl ester consist of an ester

cleavage under basic conditions with LiOH, Ba(OH)₂, and

NaOH as the reagents and a subsequent protonation of the

carboxylates by the addition of HCl, H₂SO₄, or a proton-ex-

M. Qi et al.

Paper

Synthesis

side of the starting compound 18. Adding anisole as a trap

for the intermediately occurring tert-butyl carbenium ion

makes the reaction irreversible, but this comes at the ex-

pense of the removal of the trapping agent and its product,

such as anisole and 4-tert-butylanisole. A simple solution is

to apply the process repeatedly: thus, the PyMTA tert-butyl

ester was treated with 50 equivalents of TFA for 1 hour,

then all volatiles, TFA and trifluoroacetate 20, were evapo-

rated. This treatment was applied again, allover for four

times. This way, a complete ester hydrolysis was achieved. If

100 equivalents of TFA per run were used, two runs were

sufficient. The materials obtained from this process were

highly hygroscopic and well soluble in MeOH. They con-

tained not only 4-chloro-PyMTA or 4-iodo-PyMTA but also

a large amount of TFA which tenaciously resisted removal

under reduced pressure. Dissolving the materials in MeOH

and dropping this solution into a mixture of pentane and

Et₂O gave {4-halo-PyMTA·n TFA} 16 and 17 with a content

of the structural motif 4-halo-PyMTA common to all spe-

cies of around 94 wt% as colorless precipitates (for details,

see Supporting Information). These materials have a very

limited solubility in MeOH.

Surprisingly, the PyMTA tert-butyl ester is also easily

hydrolyzed under basic conditions, as shown in Figure 1 for

4-iodo-PyMTA tert-butyl ester 10b. With NaOD, 78% of the

ester groups were hydrolyzed within 7 hours at room tem-

perature and after 24 hours the hydrolysis was nearly com-

plete. Such an unusual reactivity of a tert-butyl ester has

also been reported for 1,4,7,10-tetraazacyclododecane-

1,4,7,10-tetraacetic acid (DOTA) tert-butyl ester.⁸⁰We as-

sume complexation of the metal ion by the PyMTA ester

causing an increase in electrophilicity of the ester groups

and consequently the unusual lability of a tert-butyl ester

group towards hydroxide. In line with this assumption, we

found that the rate of the hydrolysis depends on the type of

the alkali metal ion M⁺of MOD but does not follow the ba-

sicity of MOD (basicity: KOH > NaOH > LiOH). Only one of

the three ions will best meet the demand of the ligand with

respect to ion radius. The hydrolysis rate decreases in the

order Na⁺, K⁺, Li⁺(Figure 1); clearly, Na⁺is a better fit than

Li⁺or K⁺.

CO₂

CO₂Et

(a)

n H⁺

m Na⁺

X = Cl 14

X = Cl 6a

X = I

10a

CO₂^tBu

CO₂H

(b)

n TFA

X = Cl 16

X = I 17

X = Cl 6b

X = I

10b

CF₃

TFA

CF₃

Scheme 3 Reagents and conditions: A: Basic hydrolysis of 4-halo-PyMTA

ethyl esters 6a and 10a. (a) 1) NaOH, H₂O, EtOH, r.t., 4 h, 2) proton-

exchange resin, 85–95%. B: Hydrolysis of 4-halo-PyMTA tert-butyl esters

6b and 10b with TFA. (b) TFA (50 equiv), r.t., 1 h (treatment applied

four times), 89–93%. C: Reversibility of the hydrolysis of tert-butyl es-

ters with TFA.

All the reports on the hydrolysis of PyMTA tert-butyl es-

ters^{1,6,8,10,18,20,21,27,44–46,77,78}have applied the same procedure:

the PyMTA tert-butyl ester is dissolved in ca. 400 equiva-

lents of trifluoroacetic acid (TFA), the solution is stirred at

room temperature for 1.5 hours, and finally all volatiles are

removed to obtain PyMTA as the trifluoroacetate salt. We

applied this protocol to hydrolyze 4-iodo-PyMTA tert-butyl

ester 10b (Scheme 3B) and found that ca. 1% of the ester

groups had not been hydrolyzed. Taking into account that

the PyMTA ester has four ester groups, this corresponds to

ca. 4% of incompletely hydrolyzed PyMTA ester. Even after a

reaction time of 22 days, 0.2% of the ester groups remained.

The reaction with less TFA provided a product with more

residual ester groups. These findings are in agreement with

the reversibility of the hydrolysis of tert-butyl esters with

TFA. Reaction of a tert-butyl ester 18 with TFA results in the

formation of the corresponding acid 19 and isobutene

(Scheme 3C). Isobutene reacts immediately with TFA giving

tert-butyl trifluoroacetate (20), which can transfer the tert-

butyl group back to the acid 19.⁷⁹Upon removal of the vola-

tiles, the ratio of TFA and trifluoroacetate 20 changes in fa-

vor of trifluoroacetate 20 because of the higher volatility of

TFA.⁷⁹Therefore, the overall equilibrium shifts back to the

When working with PyMTA esters, we noticed that one

of the two aminomethyl substituents was slowly converted

into a formyl group when the material was in contact with

Cu(I/II), Pd(II), and air (for details, see Supporting Informa-

tion).⁴⁹To study this oxidation more closely, PdCl₂or Pd-

Cl₂(PPh₃)₂or CuI or a mixture of these salts was added to a

solution of 4-iodo-PyMTA ethyl ester 10a in THF and the re-

action mixture (first a suspension, after a few hours a solu-

tion) was stirred in air for 240 hours (Table 1). No oxidation

occurred in the presence of only PdCl₂and/or PdCl₂(PPh₃)₂

(Table 1, entries 2–4). When CuI was added, 12 mol% of 4-

iodo-PyMTA ethyl ester 10a was converted into PyMTA

ester aldehyde 21 (Table 1, entry 5). Additional introduction

M. Qi et al.

Paper

Synthesis

matches a Cu(II)–PyMTA ester complex with one chloride

ion as counterion. We repeated some of these experiments

using CH₂Cl₂instead of THF as solvent or exchanging ethyl

4-iodo-PyMTA ethyl ester 10a for 4-iodo-PyMTA tert-butyl

ester 10b. The results were comparable to the results with

4-iodo-PyMTA ester 10a in THF, which indicates that the

oxidation is neither specific to the solvent nor to the type of

ester group.

If there is a need to bring Cu(I/II) in contact with PyMTA

esters, according to our experience the following procedure

inhibits the oxidation: The reaction is performed under ox-

ygen-free conditions. After the reaction but before expo-

sure to air during standard workup, metal scavenger Quad-

raPure™-TU is added to remove Cu(I/II). Note that column

chromatography on silica gel gave material that became

slowly oxidized, which means that this measure is insuffi-

cient to remove Cu(I/II) completely. Though the oxidation is

very slow, Cu(I/II) causes oxidation during storage of the

materials. Material that had been treated with metal scav-

enger was storable without change in air.

In summary, we have developed procedures for the syn-

theses of 4-chloro-PyMTA esters and 4-iodo-PyMTA esters

from commercially available chelidamic acid monohydrate

in overall yields of 39–67%. Hydrolysis of the 4-halo-PyMTA

ethyl esters with aqueous NaOH and subsequent addition of

proton-exchange resin in an amount that is just sufficient

to adjust the pH to pH 3 gave materials with a content of

Figure 1 Hydrolysis of 4-iodo-PyMTA tert-butyl ester 10b with 0.35 M

alkali deuteroxide in D₂O/CD₃OD at r.t., as studied by ¹H NMR spectros-

copy. The data points are connected purely as a visual aid.

of PdCl₂or PdCl₂(PPh₃)₂raised the conversion to 35 mol%

and 42 mol%, respectively (Table 1, entries 6 and 7). When

air was excluded by working under argon and degassing the

solution, no oxidation occurred (Table 1, entries 8–10).

Clearly, oxygen is the oxidant, Cu(I/II) the catalyst, and

Pd(II) an accelerator. We assume that the complexation of

Cu(II) triggers the oxidation. This assumption gains support

from an ESI mass spectrum of the reaction solution of entry

6 (Table 1) which shows a signal at m/z = 705.0 which

Table 1 Oxidative Degradation of 4-Iodo-PyMTA Ethyl Ester 10a to PyMTA Ester Aldehyde 21^a

CO₂Et

THF, in air

CO₂Et

10a

Entry

Ester and catalysts (μmol)

Conditions

Color of the solution Product composition (%)

10a 21

10a

CuI

PdCl₂

PdCl₂(PPh₃)₂

–

air

colorless

yellow

100

5.1

–

air

100

5.0

5.1

–

air

5.1

–

air

yellow

5.3

air

yellow

5.5

5.0

–

air

yellow

5.0

–

air

yellow

–

argon

colorless

brown

yellow

100

5.0

–

5.0

^aReaction conditions: Entries 1–7: 10a (20 μmol) was dissolved in THF (5 mL) in a 10-mL glass vial with a snap-on cap, which was punctured with a cannula. The

catalysts were added and the reaction mixture was stirred at r.t. for 240 h. The solvent was removed under reduced pressure. Entries 8–10: 10a (20 μmol) was

dissolved in THF (5 mL). The colorless solution was degassed. Under argon, catalysts were added and the reaction mixture was stirred at r.t. for 240 h. Metal

scavenger QuadraPure™-TU (10 mg for 1 μmol metal) was added. The suspension was stirred for 5 h, filtered, and the solvents of the filtrate were removed. For

all entries, the ratio of 10a and 21 in the residue was determined by ¹H NMR spectroscopy.

M. Qi et al.

Paper

Synthesis

90–92 wt% of the common motif [PyMTA – 4 H⁺]^4–. Adding

more resin resulted in adhesion of PyMTA to the resin and a

drastically decreased yield. The PyMTA tert-butyl esters

were hydrolyzed by reaction with TFA. For complete hydro-

lysis, a twofold treatment with 100 equivalents of TFA is

necessary, and some TFA remains in the final product. The

content of the common motif PyMTA was determined to be

around 94 wt%. In addition to the syntheses, we have dis-

closed the surprising lability of PyMTA tert-butyl esters to-

wards alkali hydroxide and the oxidative degradation of

PyMTA esters in the presence of Cu(I/II) and oxygen, two is-

sues of importance when working with PyMTA esters.

this temperature for 22 h (caution: vigorous gas evolution). During

this time the suspension turned into a pale yellow solution. The ex-

cess SOCl₂was distilled off under reduced pressure. To the cooled (ice

bath) residual orange-red solid, MeOH (500 mL) was added slowly

(caution: exothermic reaction and vigorous gas evolution). The re-

sulting suspension was heated to reflux (oil bath temperature 70 °C)

and MeOH was added until all of the solid was dissolved. Upon slowly

cooling to room temperature, colorless needles formed. The suspen-

sion was cooled with an ice-water bath and the needles were collect-

ed by filtration and washed with cold MeOH to obtain chloro diester 2

(35.6 g, 87%); mp 139–140 °C.

¹H NMR (500 MHz, CDCl₃): δ = 8.28 (s, 2 H, ArH), 4.02 (s, 6 H, Me).

¹³C NMR (125 MHz, CDCl₃): δ = 163.9 (C=O), 149.3 (C_Armeta to Cl),

146.6 (C_ArCl), 128.1 (C_ArH), 53.3 (CH₃).

MS (ESI): m/z = 480.9 [2 M + Na]⁺, 251.9 [M + Na]⁺, 229.9 [M + H]⁺.

HRMS (EI, 70 eV): m/z [M]^+•calcd for C₉H₈ClNO₄^+•: 229.01364; found:

229.01373.

Unless otherwise stated, reactions were performed under ambient at-

mosphere, using commercial solvents and reagents. The solvents used

for extraction and chromatography were of technical grade and were

distilled prior to use. The proton-exchange resin (Dowex^®50WX4 hy-

drogen form, Sigma-Aldrich, 91 g) was sequentially washed with THF

(3 × 200 mL), EtOH (2 × 100 mL), H₂O (2 × 150 mL), and EtOH (200

mL), and then dried over P₄O₁₀at 0.05 mbar for 5 d to obtain a pure

and dry proton-exchange resin (30 g). The temperature given for the

reactions refers to the bath temperature. Solvents were removed un-

der reduced pressure at a bath temperature of ca. 40 °C. The products

were dried at room temperature at ca. 0.05 mbar. The pH values of

solutions were determined using pH indicator strips (resolution: 0.3

pH, Merck). Column chromatography was carried out on silica gel

60 M (Acros), applying slight pressure. The size of columns is given as

diameter × length. Material was loaded onto the column dissolved in

a small quantity of the eluent. TLC was performed on silica gel coated

aluminum foil (Merck, 60 F254). The spots were detected with UV

light (λ = 254 nm). The compositions of solvent mixtures are given in

volume ratios. For centrifugation, a VWR CompactStar CS4 centrifuge

(centrifuging radius: 8.5 cm, relative centrifugal force: 4000g, rotor

angle: 30°) and VWR metal-free centrifuge tubes (15 mL, overall

length: 118 mm, inside diameter: 15 mm, outside diameter: 17 mm,

polypropylene, sterile) were used. NMR spectra were recorded at

room temperature on a Bruker Avance 500 500-MHz instrument.

The spectra were calibrated using the solvent signal as an internal

standard [CDCl₃: δ(¹H) = 7.25, δ(¹³C) = 77.0; CD₂Cl₂: δ(¹H) = 5.32,

δ(¹³C) = 53.8; DMSO-d₆: δ(¹H) = 2.49, δ(¹³C) = 39.5; CD₃OD: δ(¹H) =

3.31, δ(¹³C) = 49.0; D₂O: δ(¹H) = 4.79]. For ¹³C NMR experiments in

Anal. Calcd for C₉H₈ClNO₄: C, 47.08; H, 3.51; N, 6.10. Found: C, 47.18;

H, 3.51; N, 6.07.

Chloro Diol 3

This reaction was performed under argon. NaBH₄(3.28 g, 86.7 mmol)

was added portionwise within 35 min to a suspension of chloro di-

ester 2 (10.0 g, 43.6 mmol) and finely powdered CaCl₂(4.84 g, 43.6

mmol) in MeOH (250 mL) under cooling with a water bath (about

room temperature). The colorless suspension was stirred at room

temperature and the reaction was followed by TLC [R_f(2) = 0.40, R_f(3)

= 0.10 (Et₂O)]. The reaction was complete after 4 h and the suspension

was filtered through silica gel (3 cm × 6 cm, rinsing with MeOH). Re-

moval of the solvents from the filtrate afforded a yellow solid (21.4 g)

which was dissolved in 1 M HCl (180 mL). The pH of the solution was

raised to pH 12 by addition of 1 M aqueous NaOH, whereupon a col-

orless solid precipitated. The precipitate was isolated by filtration and

dried under reduced pressure over P₄O₁₀for 3 d and in air for 7 d until

the weight of the product stayed constant. A colorless powder (8.01 g,

consisting of chloro diol 3 and inorganic salts) was obtained.

¹H NMR (500 MHz, CDCl₃): δ = 7.24 (s, 2 H, ArH), 4.76 (d, ³J = 5.4 Hz, 4

H, CH₂), 2.99 (t, ³J = 5.4 Hz, 2 H, OH).

¹H NMR (500 MHz, DMSO-d₆): δ = 7.35 (s, 2 H, ArH), 5.28 (t, J = 5.7

Hz, 2 H, OH), 4.52 (d, ³J = 5.7 Hz, 4 H, CH₂).

¹³C NMR (125 MHz, DMSO-d₆): δ = 163.5 (C_Armeta to Cl), 144.0 (C_ArCl),

118.0 (C_ArH), 63.7 (CH₂).

MS (ESI): m/z = 195.9 [M + Na]⁺, 174.0 [M + H]⁺.

HRMS (ESI): m/z [M + Na]⁺calcd for C₇H₈ClNO₂Na⁺: 196.0136; found:

D₂O, a drop of MeOH was added as the internal standard [δ(¹³C)_MeOH

49.5]. Signal assignments are supported by DEPT-135, COSY, HMBC,

and HMQC experiments. ESI mass spectra were recorded using an Es-

quire 3000 ion trap mass spectrometer (Bruker Daltonik) equipped

with a standard ESI source. EI mass spectra were recorded using an

Autospec X magnetic sector mass spectrometer with EBE geometry

(Vacuum Generators, Manchester, UK) equipped with a standard EI

source. HRMS analyses were performed using an FT-ICR mass spec-

trometer (APEX III, Bruker Daltonik) interfaced to an external ESI ion

source. The ratio of components in a mixture was determined by ¹H

NMR spectroscopy and is given as the molar ratio.

196.0138.

Trichloride 4

The material obtained from the reduction of chloro diester 2 (2.00 g,

consisting of chloro diol 3 and inorganic salts) was suspended in DMF

(50 mL). POCl₃(2.3 mL, 24.7 mmol) was added slowly to the suspen-

sion (caution: slightly exothermic reaction), whereupon the precipi-

tate dissolved completely. The pale yellow solution was stirred at

50 °C for 40 min. During this time the color of the solution changed

from pale yellow to orange. The solution was cooled to room tem-

perature and saturated aqueous NaHCO₃(50 mL) was added (caution:

exothermic reaction and vigorous gas evolution). Et₂O (50 mL) was

added, the organic phase was separated, and the aqueous phase was

Chloro Diester (Dimethyl 4-Chlorodipicolinate, 2)

Anhydrous DMF (27.6 mL, 358 mmol) was added slowly to a suspen-

sion of chelidamic acid monohydrate (1) (36.0 g, 179 mmol) in SOCl₂

(230 mL, 3.17 mol) (caution: exothermic reaction and vigorous gas

evolution). After 1 h of stirring at room temperature, the suspension

was slowly heated to reflux (oil bath temperature 95 °C) and stirred at

M. Qi et al.

Paper

Synthesis

extracted with Et₂O (2 ×). The combined organic phases were washed

with H₂O (3 ×) and dried over MgSO₄. Removal of the solvents gave

trichloride 4 (1.54 g, 63% over 2 steps) as a pale yellow solid.

¹H NMR (500 MHz, CDCl₃): δ = 7.46 (s, 2 H, ArH), 4.62 (s, 4 H, CH₂).

¹³C NMR (125 MHz, CDCl₃): δ = 157.8 (C_Armeta to Cl), 145.9 (C_ArCl),

122.3 (C_ArH), 45.7 (CH₂).

Iodo Diester (Dimethyl 4-Iododipicolinate, 7)

The published procedure⁵³was followed. NaI (39.2 g, 261 mmol) was

added to a solution of chloro diester 2 (6.0 g, 26.1 mmol) in MeCN

(200 mL). The suspension was sonicated for 20 min at room tempera-

ture. AcCl (6.05 mL, 78.2 mmol) was added slowly to the colorless

suspension, whereupon the color of the suspension changed to dark

red. The dark red suspension was sonicated for 30 min at room tem-

perature. CH₂Cl₂(200 mL) and saturated aqueous Na₂CO₃(100 mL)

were added. The organic phase was separated and washed firstly

with saturated aqueous Na₂S₂O₃until the color of the organic phase

had changed from red to colorless and then with saturated aqueous

NaCl (2 × 50 mL) and H₂O (50 mL). The organic phase was dried over

Na₂SO₄and filtered. Removal of the solvents gave pale yellow crystals

which were recrystallized in MeOH (280 mL) giving iodo diester 7 as

colorless needles (7.85 g, 94%); mp 170–171 °C.

¹H NMR (500 MHz, CDCl₃): δ = 8.65 (s, 2 H, ArH), 4.01 (s, 6 H, Me).

¹³C NMR (125 MHz, CDCl₃): δ = 163.7 (C=O), 148.2 (C_Armeta to I),

137.0 (C_ArH), 106.9 (C_ArI), 53.3 (CH₃).

MS (ESI): m/z = 664.7 [2 M + Na]⁺, 343.9 [M + Na]⁺, 321.9 [M + H]⁺.

HRMS (EI, 70 eV): m/z [M]^+•calcd for C₉H₈INO₄^+•: 320.94925; found:

320.94928.

MS (ESI): m/z = 209.8 [M + H]⁺.

HRMS (EI, 70 eV): m/z [M]^+•calcd for C₇H₆Cl₃N^+•: 208.95603; found:

208.95683.

4-Chloro-PyMTA Ethyl Ester (6a)

Trichloride 4 (749 mg, 3.56 mmol) was dissolved in MeCN (30 mL).

Na₂CO₃(3.78 g, 35.7 mmol), KI (1.18 g, 7.13 mmol), and diethyl imi-

nodiacetate (5a) (2.2 mL, 12.6 mmol) were added successively. The

suspension was sonicated in an ultrasound bath at room temperature

for 35 min and then stirred at room temperature for 23 h. Et₂O and

H₂O were added. The organic phase was separated and the aqueous

phase was extracted with Et₂O (2 ×). The solvents of the combined or-

ganic phases were removed. Chromatography (4.5 cm × 28 cm; pen-

tane–Et₂O, 1:2) of the residual yellow oil (2.09 g) gave ethyl ester 6a

(1.30 g, 71%) as a colorless oil; R_f= 0.35.

¹H NMR (500 MHz, CDCl₃): δ = 7.52 (s, 2 H, ArH), 4.13 (q, ³J = 7.2 Hz, 8

H, CH₂CH₃), 3.98 (s, 4 H, ArCH₂), 3.56 (s, 8 H, CH₂CO), 1.23 (t, J = 7.2

Anal. Calcd for C₉H₈INO₄: C, 33.67; H, 2.51; N, 4.36. Found: C, 33.74;

H, 2.54; N, 4.23.

Hz, 12 H, CH₃).

Iodo Diol 8

¹³C NMR (125 MHz, CDCl₃): δ = 171.0 (CO), 160.3 (C_Armeta to Cl),

145.6 (C_ArCl), 121.4 (C_ArH), 60.5 (CH₂CH₃), 59.5 (ArCH₂), 54.9 (CH₂CO),

14.2 (CH₃).

This reaction was performed under argon. NaBH₄(946 mg, 25.0

mmol) was added portionwise within 10 min to a suspension of iodo

diester 7 (4.03 g, 12.5 mmol) and finely powdered CaCl₂(1.39 g, 12.5

mmol) in anhyd MeOH (100 mL) at room temperature, whereupon

the color of the suspension changed to orange. The orange suspension

was stirred at room temperature and the reaction was followed by

TLC [R_f(7) = 0.53, R_f(8) = 0.1 (Et₂O)]. The reaction was complete after 5

h and the colorless suspension was filtered through silica gel (3 cm ×

2 cm, rinsing with MeOH). The solvents of the filtrate were removed

and the residual colorless solid was dissolved in 1 M HCl (60 mL). The

pH of the solution was raised to pH 12 by addition of 1.0 M aqueous

NaOH. Thereupon, a colorless solid precipitated. The precipitate was

isolated by filtration and dried under reduced pressure over P₄O₁₀for

3 d. A white powder (3.95 g, consisting of iodo diol 8 and inorganic

salts) was obtained.

MS (ESI): m/z = 538.3 [M + Na]⁺, 516.3 [M + H]⁺.

HRMS (ESI): m/z [M + Na]⁺calcd for C₂₃H₃₄ClN₃O₈Na⁺: 538.1927;

found: 538.1928.

Anal. Calcd for C₂₃H₃₄ClN₃O₈: C, 53.54; H, 6.64; N, 8.14. Found: C,

53.51; H, 6.75; N, 8.07.

4-Chloro-PyMTA tert-Butyl Ester (6b)

Trichloride 4 (765 mg, 3.63 mmol) was dissolved in MeCN (30 mL).

Na₂CO₃(3.85 g, 36.4 mmol), KI (1.21 g, 7.30 mmol), and di-tert-butyl

iminodiacetate (5b) (3.05 g, 12.4 mmol) were added successively. The

suspension was sonicated in an ultrasound bath at room temperature

for 35 min and then stirred at room temperature for 23 h. The suspen-

sion was filtered through silica gel (3.0 cm × 4.0 cm, rinsing with

Et₂O). The solvents of the filtrate were removed. Chromatography (4.5

cm × 28 cm; pentane–Et₂O, 1:1) of the residual yellow oil gave tert-

butyl ester 6b (1.74 g, 76%) as a colorless oil; R_f= 0.45.

¹H NMR (500 MHz, CDCl₃): δ = 7.54 (s, 2 H, ArH), 3.96 (s, 4 H, ArCH₂),

3.43 (s, 8 H, CH₂CO), 1.43 (s, 36 H, CH₃).

¹³C NMR (125 MHz, CDCl₃): δ = 170.3 (CO), 160.7 (C_Armeta to Cl),

145.6 (C_ArCl), 121.1 (C_ArH), 81.0 (CMe₃), 59.4 (ArCH₂), 55.8 (CH₂CO),

28.1 (CH₃).

¹H NMR (500 MHz, DMSO-d₆): δ = 7.69 (s, 2 H, ArH), 5.48 (t, J = 5.9

Hz, 2 H, OH), 4.47 (d, ³J = 5.9 Hz, 4 H, CH₂).

¹³C NMR (125 MHz, DMSO-d₆): δ = 162.2 (C_Armeta to I), 127.0 (C_ArH),

107.4 (C_ArI), 63.6 (CH₂).

MS (ESI): m/z = 552.8 [2 M + Na]⁺, 287.9 [M + Na]⁺, 265.9 [M + H]⁺.

HRMS (ESI): m/z [2 M + Na]⁺calcd for C₁₄H₁₆I₂N₂O₄Na⁺: 552.90916;

found: 552.90706.

Iodo Dimesylate 9; Experiment A

MS (ESI): m/z = 650.4 [M + Na]⁺, 628.4 [M + H]⁺.

HRMS (ESI): m/z [M + Na]⁺calcd for C₃₁H₅₀ClN₃O₈Na⁺: 650.3179;

The material obtained from the reduction of iodo diester 7 (2.52 g,

consisting of iodo diol 8 and inorganic salts) was suspended in CH₂Cl₂

(25 mL) and Et₃N (3.5 mL, 25.1 mmol). MsCl (1.85 mL, 23.9 mmol)

was added within 5 min to the cooled (ice bath) suspension. After the

suspension was stirred for 10 min under cooling, the ice bath was re-

moved and the suspension was stirred for 15 min at room tempera-

ture. H₂O (15 mL) and saturated aqueous NaHCO₃(15 mL) were add-

ed. The suspension was filtered. The filtrate consisted of two liquid

phases. The organic phase was separated and the aqueous phase was

extracted with CH₂Cl₂(6 × 15 mL). The organic phases were com-

found: 650.3162.

Anal. Calcd for C₃₁H₅₀ClN₃O₈: C, 59.27; H, 8.02; N, 6.69. Found: C,

59.14; H, 8.09; N, 6.56.

M. Qi et al.

Paper

Synthesis

bined, dried over MgSO₄, and the solvents were removed. A white sol-

id consisting of iodo dimesylate 9, monochloride 12, and

iodo dichloride 13 in a molar ratio of 90:7:3 was obtained. The com-

ponents of the colorless solid were separated by column chromatog-

raphy (5 cm × 46 cm; Et₂O–CH₂Cl₂, 1:4). Iodo dimesylate 9 (2.34 g,

69% over 2 steps; R_f= 0.42), monochloride 12 (130 mg, 5% over 2

steps; R_f= 0.65), and iodo dichloride 13 (61 mg, 2% over 2 steps; R_f=

0.80) were obtained as colorless solids.

4-Iodo-PyMTA Ethyl Ester 10a; Experiment A

A suspension of iodo dimesylate 9 (5.22 g, 12.4 mmol), diethyl imino-

diacetate (5a) (8.23 g, 43.5 mmol), and Na₂CO₃(15.67 g, 147.8 mmol)

in MeCN (130 mL) was stirred at room temperature for 23 h. The reac-

tion mixture was filtered through silica gel (4 cm × 4 cm, rinsing with

Et₂O). The solvents of the filtrate were removed. Chromatography (6

cm × 35 cm; pentane–Et₂O, 1:2) of the residual pale yellow oil yielded

4-iodo-PyMTA ethyl ester 10a (5.88 g, 78%) as a colorless oil; R_f= 0.60.

For analytical data, see experiment B.

Iodo Dimesylate 9

Mp 129–130 °C.

4-Iodo-PyMTA Ethyl Ester 10a; Experiment B

¹H NMR (500 MHz, CDCl₃): δ = 7.83 (s, 2 H, ArH), 5.24 (s, 4 H, CH₂),

3.11 (s, 6 H, CH₃).

¹H NMR (500 MHz, CD₂Cl₂): δ = 7.86 (s, 2 H, ArH), 5.24 (s, 4 H, CH₂),

3.10 (s, 6 H, CH₃).

¹³C NMR (125 MHz, CD₂Cl₂): δ = 154.9 (C_Armeta to I), 131.4 (C_ArH),

107.4 (C_ArI), 70.5 (CH₂), 38.3 (CH₃); recorded in CD₂Cl₂because of the

poor solubility of 9 in CDCl₃.

MS (ESI): m/z = 459.8 [M + K]⁺, 443.9 [M + Na]⁺, 421.9 [M + H]⁺.

HRMS (ESI): m/z [M + Na]⁺calcd for C₉H₁₂INO₆S₂Na⁺: 443.90429;

The material obtained in experiment B for the preparation of

iodo dimesylate 9 [1.17 g, consisting of iodo dimesylate 9 (2.49

mmol), monochloride 12 (0.31 mmol), and iodo dichloride 13 (0.03

mmol)] was dissolved in MeCN (30 mL). Na₂CO₃(3.00 g, 28.4 mmol),

KI (936 mg, 5.64 mmol), and diethyl iminodiacetate (5a) (1.8 mL, 10.3

mmol) were added successively. The suspension was sonicated in an

ultrasound bath at room temperature for 1 h and then stirred at room

temperature for 20 h. Et₂O and H₂O were added successively to the

suspension. The organic phase was separated and the aqueous phase

was extracted with Et₂O (2 ×). The combined organic phases were

washed with H₂O (2 ×). The solvents were removed. Chromatography

(4.5 cm × 28 cm; pentane–Et₂O, 1:2) of the residual orange oil gave 4-

iodo-PyMTA ethyl ester 10a (1.25 g, 73% over 3 steps) as a colorless

oil; R_f= 0.60.

found: 443.90533.

Anal. Calcd for C₉H₁₂INO₆S₂: C, 25.66; H, 2.87; N, 3.33; S, 15.22. Found:

C, 25.71; H, 2.81; N, 3.46; S, 14.94.

¹H NMR (500 MHz, CDCl₃): δ = 7.91 (s, 2 H, ArH), 4.16 (q, ³J = 7.1 Hz, 8

H, CH₂CH₃), 3.97 (s, 4 H, ArCH₂), 3.58 (s, 8 H, CH₂CO), 1.26 (t, J = 7.1

Hz, 12 H, CH₃).

¹³C NMR (125 MHz, CDCl₃): δ = 170.9 (CO), 159.3 (C_Armeta to I), 130.3

(C_ArH), 107.3 (C_ArI), 60.5 (CH₂CH₃), 59.2 (ArCH₂), 54.9 (CH₂CO), 14.2

(CH₃).

Monochloride 12

¹H NMR (500 MHz, CDCl₃): δ = 7.85 and 7.78 (2 s, 1 H each, ArH), 5.24

(s, 2 H, CH₂O), 4.57 (s, 2 H, CH₂Cl), 3.11 (s, 3 H, CH₃).

MS (ESI): m/z = 744.7 [2 M + Na]⁺, 399.8 [M + K]⁺, 383.9 [M + Na]⁺,

361.9 [M + H]⁺.

MS (ESI): m/z = 630.2 [M + Na]⁺, 608.2 [M + H]⁺.

HRMS (ESI): m/z [M + Na]⁺calcd for C₂₃H₃₄IN₃O₈Na⁺: 630.12828;

Iodo Dichloride 13

¹H NMR (500 MHz, CDCl₃): δ = 7.82 (s, 2 H, ArH), 4.58 (s, 4 H, CH₂).

MS (EI, 70 eV): m/z = 300.8 (100.0) [M]^+•, 265.9 (38.6) [M − Cl]⁺, 173.9

(6.5) [M − I]⁺, 139.0 (11.5) [M − I − Cl]^+•, 104.0 (18.6) [M − I − 2Cl]⁺.

found: 630.12686.

Anal. Calcd for C₂₃H₃₄IN₃O₈: C, 45.48; H, 5.64; N, 6.92. Found: C, 45.62;

H, 5.72; N, 6.91.

Iodo Dimesylate 9; Experiment B

4-Iodo-PyMTA tert-Butyl Ester 10b; Experiment A

The material obtained from the reduction of iodo diester 7 (1.83 g,

consisting of iodo diol 8 and inorganic salts) was suspended in CH₂Cl₂

(80 mL). This colorless suspension was sonicated in an ultrasound

bath at room temperature for 20 min. Ice was added to cool the ultra-

sound bath to 3 °C. Et₃N (2.4 mL, 17.2 mmol) and MsCl (1.30 mL, 16.8

mmol) were added at 3 °C successively to the suspension. The suspen-

sion was stirred and sonicated in the cold water bath for 100 min,

whereupon the bath temperature increased slowly to 10 °C. H₂O (100

mL) and saturated aqueous NaHCO₃(20 mL) were added. The suspen-

sion was filtered. The filtrate consisted of two liquid phases. The or-

ganic phase was separated and the aqueous phase was extracted with

CH₂Cl₂(3 × 40 mL). The combined organic phases were filtered

through silica gel (4 cm × 4 cm; rinsing with CH₂Cl₂–Et₂O, 4:1). Re-

moval of the solvents from the filtrate gave a colorless solid (2.15 g)

consisting of iodo dimesylate 9 (79% yield over 2 steps), monochloride

12 (10% yield over 2 steps), and iodo dichloride 13 (1% yield over 2

steps) in a molar ratio of 88:11:1. For analytical data, see experiment

A suspension of iodo dimesylate 9 (1.50 g, 3.56 mmol), di-tert-butyl

iminodiacetate (5b) (2.59 g, 10.6 mmol), and Na₂CO₃(3.78 g, 35.6

mmol) in MeCN (50 mL) was stirred at room temperature for 24 h.

Et₂O (25 mL) and H₂O (25 mL) were added successively. The organic

phase was separated and the aqueous phase was extracted with Et₂O

(3 × 15 mL). The solvents of the combined organic phases were re-

moved. Chromatography (5 cm × 32 cm; pentane–Et₂O, 1:1) of the re-

sidual colorless viscous oil provided 4-iodo-PyMTA tert-butyl ester

10b (2.30 g, 90%) as a colorless viscous oil; R_f= 0.50. For analytical da-

ta, see experiment B.

4-Iodo-PyMTA tert-Butyl Ester 10b; Experiment B

The material obtained in experiment B for the preparation of

iodo dimesylate 9 [959 mg, consisting of iodo dimesylate 9 (2.04

mmol), monochloride 12 (0.26 mmol), and iodo dichloride 13 (0.02

mmol)] was dissolved in MeCN (30 mL). Na₂CO₃(2.46 g, 23.2 mmol),

KI (769 mg, 4.63 mmol), and di-tert-butyl iminodiacetate (5b) (2.11 g,

8.61 mmol) were added successively. The suspension was too viscous

to be stirred. Therefore, additional MeCN (10 mL) was added. The col-

orless suspension was sonicated in an ultrasound bath at room tem-

perature for 1 h and then stirred at room temperature for 20 h. Et₂O

and H₂O were added successively. The organic phase was separated

M. Qi et al.

Paper

Synthesis

and the aqueous phase was extracted with Et₂O (2 ×). The combined

organic phases were washed with H₂O (2 ×). The solvents were re-

moved. Chromatography (4.5 cm × 29 cm; pentane–Et₂O, 1:2) of the

residual orange oil gave 4-iodo-PyMTA tert-butyl ester 10b (1.37 g,

82% over 3 steps) as a colorless oil; R_f= 0.50.

¹H NMR (500 MHz, CDCl₃): δ = 7.89 (s, 2 H, ArH), 3.94 (s, 4 H, ArCH₂),

3.44 (s, 8 H, CH₂CO), 1.44 (s, 36 H, CH₃).

¹³C NMR (125 MHz, CDCl₃): δ = 170.3 (CO), 159.8 (C_Armeta to I), 130.1

(C_ArH), 107.5 (C_ArI), 81.1 (CMe₃), 59.2 (ArCH₂), 55.8 (CH₂CO), 28.2

(CH₃).

MS (ESI): m/z = 742.3 [M + Na]⁺, 720.3 [M + H]⁺.

HRMS (ESI): m/z [M + Na]⁺calcd for C₃₁H₅₀IN₃O₈Na⁺: 742.25348;

¹H NMR (500 MHz, CD₃OD): δ = 7.72 (s, 2 H, ArH), 4.27 (s, 4 H, ArCH₂),

3.73 (s, 8 H, CH₂CO).

¹H NMR (500 MHz, D₂O): δ = 7.68 (s, 2 H, ArH), 4.74 (s, 4 H, ArCH₂),

4.14 (s, 8 H, CH₂CO).

¹³C NMR (125 MHz, D₂O): δ = 169.3 (CO), 151.2 (C_Armeta to Cl), 147.5

(C_ArCl), 125.7 (C_ArH), 58.1 (ArCH₂), 56.0 (CH₂CO).

¹⁹F NMR (470 MHz, D₂O): 75.39 (F₃CCO₂^–).

MS (ESI): m/z = 401.9 [M – H]^–, 200.3 [M – 2 H]^2–

Anal. Calcd for {4-chloro-PyMTA·n TFA} with 94.2 wt% of 4-chloro-

PyMTA (C₁₅H₁₈ClN₃O₈) not taking TFA into account: C, 42.03; H, 4.22;

N, 9.81. Found: C, 43.24*; H, 4.65*; N, 9.80. *TFA contributes C and H.

found: 742.25518.

{[4-Iodo-PyMTA – 4 H⁺]^4–·n H⁺·m Na⁺} (15)

Anal. Calcd for C₃₁H₅₀IN₃O₈: C, 51.74; H, 7.00; N, 5.84. Found: C, 51.85;

H, 7.04; N, 5.82.

4-Iodo-PyMTA ethyl ester 10a (200 mg, 329 μmol) was dissolved in

EtOH (2 mL). A solution of NaOH (104 mg, 2.60 mmol) in H₂O (2 mL)

was added whereupon a mixture of two liquid phases formed, which

turned into one colorless phase within the next 10 min of stirring.

This was stirred at room temperature for 4 h. Then, proton-exchange

resin was added in an amount needed to reduce the pH of the solution

to pH 3. The solution was separated from the proton-exchange resin

by filtration, and the resin was washed with H₂O (5 × 2 mL). The sol-

vents of the combined filtrates were removed. {[4-Iodo-PyMTA – 4

H⁺]^4–·n H⁺·m Na⁺} (15) (168 mg, 95%) was obtained as a colorless solid.

The content of the structural motif [4-iodo-PyMTA – 4 H⁺]^4–was de-

termined by quantitative ¹H NMR spectroscopy to be 92.2 wt% (for

details, see Supporting Information).

¹H NMR (500 MHz, CD₃OD): δ = 7.90 (s, 2 H, ArH), 4.25 (s, 4 H, ArCH₂),

3.67 (s, 8 H, CH₂CO).

¹³C NMR (125 MHz, CD₃OD): δ = 173.4 (CO), 157.1 (C_Armeta to I),

133.6 (C_ArH), 108.8 (C_ArI), 59.6 (ArCH₂), 57.6 (CH₂CO).

MS (ESI): m/z = 493.8 [M – H]^–, 246.3 [M – 2 H]^2–

{[4-Chloro-PyMTA – 4 H⁺]^4–·n H⁺·m Na⁺} (14)

4-Chloro-PyMTA ethyl ester (6a) (199 mg, 386 μmol) was dissolved in

EtOH (2 mL). A solution of NaOH (125 mg, 3.12 mmol) in H₂O (2 mL)

was added whereupon a mixture of two liquid phases formed, which

turned into one colorless phase within the next 10 min of stirring.

This was stirred at room temperature for 4 h. Then, proton-exchange

resin was added in an amount needed to reduce the pH of the solu-

tion to pH 3. The solution was separated from the proton-exchange

resin by filtration, and the resin was washed with a 1:1 mixture of

EtOH and H₂O (3 × 1.5 mL). The solvents of the combined filtrates

were removed. {[4-Chloro-PyMTA – 4 H⁺]^4–·n H⁺·m Na⁺} (14) (148 mg,

85%) was obtained as a colorless solid. The content of the structural

motif [4-chloro-PyMTA – 4 H⁺]^4–was determined by quantitative ¹H

NMR spectroscopy to be 89.5 wt% (for details, see Supporting Infor-

mation).

¹H NMR (500 MHz, CD₃OD): δ = 7.52 (s, 2 H, ArH), 4.25 (s, 4 H, ArCH₂),

3.62 (s, 8 H, CH₂CO).

HRMS (ESI): m/z [M – H]^–calcd for C₁₅H₁₇IN₃O₈^–: 494.00658; found:

494.00558.

¹³C NMR (125 MHz, CD₃OD): δ = 173.5 (CO), 158.1 (C_Armeta to Cl),

147.3 (C_ArCl), 124.6 (C_ArH), 59.9 (ArCH₂), 57.8 (CH₂CO).

Anal. Calcd for {[4-iodo-PyMTA – 4 H⁺]^4–·n H⁺·m Na⁺} with 92.2 wt% of

[4-iodo-PyMTA – 4 H⁺]^4–(C₁₅H₁₄IN₃O₈): C, 33.82; H, 2.65; N, 7.89.

Found: C, 34.00; H, 3.66*; N, 7.93. *H comes from [4-iodo-PyMTA – 4

H⁺]^4–as well as from ‘·n H⁺’.

MS (ESI): m/z = 401.9 [M – H]^–, 200.4 [M – 2 H]^2–

HRMS (ESI): m/z [M – H]^–calcd for C₁₅H₁₇ClN₃O₈^–: 402.07097; found:

402.06984.

Anal. Calcd for {[4-chloro-PyMTA – 4 H⁺]^4–·n H⁺·m Na⁺} with 89.5 wt%

of [4-chloro-PyMTA – 4 H⁺]^4–(C₁₅H₁₄ClN₃O₈): C, 40.33; H, 3.16; N,

9.40. Found: C, 40.61; H, 4.32*; N, 9.53. *H comes from [4-chloro-

PyMTA – 4 H⁺]^4–as well as from ‘·n H⁺’.

{4-Iodo-PyMTA·n TFA} (17)

4-Iodo-PyMTA tert-butyl ester (10b) (100 mg, 139 μmol) was dis-

solved in TFA (535 μL, 7.0 mmol) and the solution was stirred at room

temperature for 1 h. The volatile components were removed at room

temperature by lowering the pressure to 10 mbar, giving a pale yellow

solid. This procedure was applied all over four times. Then, the resid-

ual pale yellow solid was dissolved in MeOH (1 mL) and the solution

was dropped into a mixture of Et₂O (1 mL) and pentane (2 mL),

whereupon a colorless solid precipitated. Additional pentane was

added slowly to the supernatant until no more precipitate formed.

The precipitate was separated via centrifugation (5000 rpm, 1 min)

and dried under reduced pressure over P₄O₁₀. {4-Iodo-PyMTA·n TFA}

(17) (68 mg, 93%) was obtained as a colorless solid. The content of the

structural motif 4-iodo-PyMTA was determined by quantitative ¹H

NMR spectroscopy to be 93.9 wt% (for details, see Supporting Infor-

mation).

{4-Chloro-PyMTA·n TFA} (16)

4-Chloro-PyMTA tert-butyl ester (6b) (99 mg, 158 μmol) was dis-

solved in TFA (613 μL, 8.0 mmol) and the solution was stirred at room