Tetrakisphosphates and Bispyrophosphates of myo-Inositol Derivatives as Allosteric Effectors of Human Hemoglobin: Synthesis, Molecular Recognition, and Oxygen Release

Article abstract of DOI:10.1002/cmdc.201000421

Various 2,5- and 1,4-substituted and unsubstituted myo-inositol tetrakisphosphates and bispyrophosphates were prepared following a general synthetic pathway. All final compounds were tested for their capability to induce oxygen release from human hemoglob

Full text of DOI:10.1002/cmdc.201000421

DOI: 10.1002/cmdc.201000421
Tetrakisphosphates and Bispyrophosphates of myo-Inositol
Derivatives as Allosteric Effectors of Human Hemoglobin:
Synthesis, Molecular Recognition, and Oxygen Release
Alexandros E. Koumbis,[a, d] Carolina D. Duarte,[a, b] Claude Nicolau,*[a, b, c] and
Jean-Marie Lehn*[a]
Dedicated to the memory of Professor Antonio Tamburro
Various 2,5- and 1,4-substituted and unsubstituted myo-inositol
tetrakisphosphates and bispyrophosphates were prepared fol-
lowing a general synthetic pathway. All final compounds were
tested for their capability to induce oxygen release from
human hemoglobin. Most of these proved to be efficient allo-
steric effectors, with similar affinities for hemoglobin to that of
myo-inositol hexakisphosphate, which is one of the best
known allosteric effectors of hemoglobin. The efficacy was
found to be higher for free phosphates than pyrophosphates.
As allosteric Hb effectors, these compounds enable enhanced
oxygen release. These effects increase with the strength of Hb
binding and correspond primarily to electrostatic interactions.
Stereochemical and steric factors also play a significant but
secondary role in molecular recognition. In view of the central
role played by hypoxia in numerous types of diseases, the ex-
ploration of myo-inositol phosphate derivatives represents an
important avenue in the search for substances which act on
the oxygenation status of tissues and may have significant po-
tential in the discovery and development of novel drug candi-
dates.
Introduction
myo-Inositol phosphates play key roles in biological systems,
functioning as secondary messengers in important cell signal-
ing pathways.[1] Among these compounds, several inositol tet-
raphosphates (ITPs), including [Ins(1,3,4,5)P4, Ins(1,3,4,6)P4,
Ins(1,2,4,5)P4, Ins(1,4,5,6)P4 and Ins(3,4,5,6)P4], have been
shown to mobilize calcium ions from internal storage sites
through interactions with the inositol 1,4,5-trisphosphate re-
ceptor.[2] Some of these phosphates [Ins(1,3,4,5)P4,
Ins(1,3,4,6)P4, and Ins(3,4,5,6)P4] have also been implicated in
the regulation of calcium influx across the plasma mem-
brane.[3,4] Additionally,
ing, upon entrance into circulating RBCs, to increased and
regulated oxygen release upon tissue demand.[10] Other types
of synthetic allosteric effectors of Hb have previously been re-
ported, including members of the fibrate family, such as antili-
pidemic agents clofibrate,[11] bezafibrate (BZF),[12] and urea[13]
and amide[14,15] analogues of BZF, with BZF amide RSR13[16] as
the best lead compound. As numerous ailments involve hypo-
xia, including cardiovascular diseases and cancer, achieving in-
creased oxygen release is expected to restore normoxia. There-
fore, the discovery and synthesis of allosteric effectors of he-
moglobin that are able to increase oxygen release and delivery
by RBCs represent a research field which possesses major ther-
apeutic potential.
(ꢀ)-Ins(1,2,4,5)P4 dramatically increases iron accumulation in
Pseudomonas aeruginosa,[5] and Ins(3,4,5,6)P4 has been identi-
fied as an important intracellular regulator of chloride ion cur-
rent in epithelial cells.[6]
Delivery to all tissues of oxygen bound to hemoglobin (Hb)
in red blood cells (RBCs) is one of the most fundamental phys-
iological processes in aerobic organisms. Oxygen delivery is
regulated by allosteric effectors which bind to hemoglobin
and decrease its oxygen binding affinity. In humans, allosteric
regulation is effected by 2,3-bisphosphoglycerate (BPG, 1,
Figure 1),[7] whose binding to the allosteric pocket of the Hb
tetramer has been well characterized.[8] Adenosine trisphos-
phate (ATP, 2, Figure 1) and myo-inositol hexakisphosphate
(IHP, 3, Figure 1), also act as allosteric Hb effectors.[7] In fact,
IHP, a widely available natural product, is the strongest alloste-
ric effector identified to date.[7a] It displaces Hb-bound 2,3-BPG
by binding to the allosteric pocket with higher affinity,[7b,9] trig-
gering a decrease in oxygen/Hb affinity and subsequently lead-
[a] Prof. A. E. Koumbis, Dr. C. D. Duarte, Prof. C. Nicolau, Prof. J.-M. Lehn
Institut de Science et d’Ingꢀnierie Supramolꢀculaires
Universitꢀ de Strasbourg
8 Allꢀe Gaspard Monge, 67000 Strasbourg (France)
Fax: (+33)368-855140
E-mail: lehn@isis.u-strasbg.fr
[b] Dr. C. D. Duarte, Prof. C. Nicolau
NormOxys, Inc., 200 Boston Avenue, Medford, MA 02155 (USA)
[c] Prof. C. Nicolau
Friedman School of Nutrition Science and Policy
Tufts University, Boston, MA 02115 (USA)
[d] Prof. A. E. Koumbis
Present address: Laboratory of Organic Chemistry
Aristotle University of Thessaloniki, 54124 Thessaloniki (Greece)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cmdc.201000421.
ChemMedChem 2011, 6, 169 – 180
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
169

J.-M. Lehn et al.
MED
Figure 2. General structures of targeted tetrasodium ITPs and IBPPs.
ships in allosteric Hb regulation. To the best of our knowledge,
no InsP4 has ever been examined as an allosteric effector of
Hb.
Results and Discussion
Design of synthetic pathways
Several structural features were taken into account during the
synthetic design of suitable ITPs. First, only ITPs bearing two
pairs of vicinal phosphate groups were selected, in order to be
able to construct the corresponding IBPPs. With this character-
istic in mind, two major libraries emerged, one derived from
2,5-di-O-substituted myo-inositol (numbering corresponds to
the position of substituents X, Figure 2), leading to meso com-
pounds, and the other from 1,4-di-O-substituted myo-inositol,
leading to racemic mixtures of the targeted structures. Addi-
tionally, determination of a parameter regarding the nature of
substituents R was pursued. For both libraries, candidates with
either hydroxy groups (R=H) or alkyl ethers (R=Me, Et, Bu) or
esters [R=Ac, butyryl (Bt)] were selected.
Figure 1. Structures of allosteric Hb effectors 1–5 (counter cations are omit-
ted for simplicity).
We have previously shown that an IHP derivative containing
three seven-membered cyclic pyrophosphate rings, myo-inosi-
tol trispyrophosphate (ITPP, 4, Figure 1), is a novel membrane-
permeating allosteric effector of Hb. ITPP was found to strong-
ly increase oxygen release in vitro, both from free Hb and from
RBCs in whole blood, in a concentration-dependent manner.[17]
As a result, ITPP was identified as a novel and highly effective
anti-angiogenic agent with anticancer potential, able to coun-
teract the effects of hypoxia. It caused suppression of HIF-1a
and strong decrease of VEGF in cells, thus blocking the route
leading to angiogenesis.[18] Furthermore, it was found that ITTP
was capable of increasing exercise capacity in both normal
and transgenic mice with severe heart failure.[19] Therefore, it
appears that the biological effects resulting from the enhanced
oxygen release induced by ITPP may have general utility in a
variety of disease states where tissues suffer from low oxygen
tension, such as ischemic insult, heart attacks, stroke, cardio-
vascular disease, and tumor progression.
This synthetic plan was designed to enable identification of
key structural features effecting biological activity, such as:
a) the number of phosphates or pyrophosphates, which result
in different numbers of charges in comparison with IHP or
ITPP; b) the pattern of phosphate or pyrophosphate groups
(position–activity relationship); and finally c) the X group effect
(lipophilicity–activity relationship), which is fundamental for
the development of membrane-permeable drug candidates. In
general, studies of the binding interaction between synthe-
sized compounds and human Hb and the subsequent effect
on oxygen release is expected to provide insight into molecu-
lar recognition features and structure–activity relationships for
Hb allosteric regulators.
High effector activity toward Hb, similar to that of IHP, has
recently been observed for several structurally related perphos-
phorylated hexapyranoses and pentapyranoses (e.g. 5,
Figure 1) and their corresponding pyrophosphates, which were
synthesized in our laboratories.[20] The remarkable effects of
IHP and ITPP,[10,17–19] in particular, prompted us to extend our
investigations in this field of chemistry by including ITPs and
inositol bispyrophosphates (IBPPs) in our search for allosteric
effectors of Hb (6 and 7, respectively; Figure 2).[21] The synthe-
sis of these compounds is reported herein. Furthermore, the
binding of these molecules to human Hb and their effects on
oxygen release were investigated in order to gain insight into
molecular recognition features and structure–activity relation-
Synthesis of ITPs
Substituted and unsubstituted Ins(1,3,4,6)P4 (12a–f) and (ꢀ)-
Ins(2,3,5,6)P4 (17a–f), as well as unsubstituted (ꢀ)-Ins(1,2,3,4)P4
(22) were prepared as depicted in Schemes 1, 2, and 3, respec-
tively. Various salts (or the corresponding acidic forms) of un-
substituted Ins(1,3,4,6)P4 (12 f),[2a,22] Ins(2,3,5,6)P4 (17 f, racemic
or pure enantiomers),[2b,23,24] and Ins(1,2,3,4)P4 (22, racemic or
pure enantiomers)[2a,24b] have been reported previously. The
preparation of their tetrasodium salts is described herein.
170
www.chemmedchem.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2011, 6, 169 – 180

Allosteric Effectors of Human Hemoglobin
For the construction of symmetrical Ins(1,3,4,6)P4, an easily
accessible derivative of myo-inositol, butane 2,3-bisacetal 8,[25]
was used as starting material (Scheme 1). Alkylation of 8
through the corresponding sodium alkoxides yielded dialkyl
derivatives 9a–c and f, whereas diacyl compounds 9d and e
were obtained through reactions with the corresponding anhy-
drides in a mixture of triethylamine (Et3N), N,N-dimethylforma-
mide (DMF), and dichloromethane (CH2Cl2). Addition of DMF
was critical to facilitate dissolution of starting material and im-
prove the rate of acylation reactions, all of which proceeded in
excellent yields (93–99%).
strates. In all cases, the obtained crude tetrols 10 (94–100%)
were determined to be sufficiently pure by 1H and 13C NMR
and were used without further purification in the subsequent
phosphorylation reaction.
Crude tetrols 10 were suspended in a 1H-tetrazole solution
in acetonitrile (CH3CN), and dibenzyl N,N-diisopropylphosphor-
amidate was added at ambient temperature. The resulting
phosphite intermediate was directly oxidized with m-chloro-
perbenzoic acid (m-CPBA) in CH2Cl2 at low temperature to
yield benzyl tetrakisphosphates 11 in good yields (80–91%).
These perphosphorylated derivatives were then subjected to
hydrogenolysis in ethanol/water in the presence of Pd/C and
sodium bicarbonate (NaHCO3; exactly four equivalents). Pure
tetrasodium salts 12 were isolated in nearly quantitative yields
following removal of catalyst and solvents. Decabenzyl deriva-
tive 11 f required higher pressure (3 atm) and a longer reaction
time (4–5 days) for complete deprotection, as one of the inosi-
tol ring benzyl groups remained intact under milder condi-
tions.
The next step involved removal of the bisacetal protecting
groups upon treatment with aqueous trifluoroacetic acid (TFA).
While the reaction went to completion in less than 2 h for alkyl
derivatives 9a–c, extended times were required for consump-
tion of starting compounds 9d–f and resulted in the formation
of a number of side products. Addition of CH2Cl2 as a co-sol-
vent proved necessary for clean deprotection of the latter sub-
2,3:5,6-Bicyclohexylidene ketal 13, prepared in one step
from myo-inositol,[26] was used as the starting material for
preparation of the racemic Ins(2,3,5,6)P4 compounds
(Scheme 2). Thus, 13 was alkylated or acylated, following ap-
proaches similar to those described for the Ins(1,3,4,6)P4 library,
Scheme 1. Synthesis of 1,3,4,6-myo-inositol tetrakisphosphates. Reagents and
conditions: a) NaH, RI, DMF, 08C!RT, 12 h (for a, b, c); b) R2O, Et3N, DMAP,
DMF, CH2Cl2, 08C!RT, 16 h (for d, e); c) NaH, BnBr, DMF, 08C!408C, 20 h
(for f); d) TFA, H2O, RT, 2 h (for a, b, c); e) TFA, CH2Cl2, H2O, RT, 75–180 min
(for d, e, f); f) 1. (BnO)2PN(iPr)2, 1H-tetrazole, CH3CN, RT, 24 h; 2. m-CPBA,
CH2Cl2, ꢁ408C!08C, 5 h; g) Pd/C, H2 (1 or 3 atm), NaHCO3, EtOH/H2O (1:1),
RT, 24–120 h.
Scheme 2. Synthesis of 2,3,5,6-myo-inositol tetrakisphosphates. Reagents and
conditions: a) NaH, RI, DMF, 08C!RT, 12 h (for a, b, c); b) R2O, Et3N, DMAP,
DMF, CH2Cl2, 08C!RT, 16 h (for d, e); c) NaH, BnBr, DMF, 08C!408C, 20 h
(for f); d) CSA, CH2Cl2/MeOH (2:1), RT, 24 h (for a, b, c, f); e) TFA, CH2Cl2, H2O,
RT, 3 h (for d, e); f) 1. (BnO)2PN(iPr)2, 1H-tetrazole, CH3CN, RT, 24 h; 2. m-
CPBA, CH2Cl2, ꢁ408C!08C, 5 h; g) Pd/C, H2 (1 or 3 atm), NaHCO3, EtOH/H2O
(1:1), RT, 16–48 h.
ChemMedChem 2011, 6, 169 – 180
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
171

J.-M. Lehn et al.
MED
to obtain fully protected inositols 14 (87–100% yields). A clas-
sic deprotection protocol for cyclohexylidene ketals using cam-
phorosulfonic acid (CSA) in CH2Cl2/methanol was then em-
ployed for all alkyl derivatives (14a–c and f) to obtain the cor-
responding desired tetrols 15a–c and f in very good yields
(80–100%). However, this reaction failed with the correspond-
ing acylated compounds (14d and e), resulting in complicated
mixtures, an indication of acyl migration and/or transesterifica-
tion. The latter were smoothly deprotected to 15d and e (92–
99%) upon treatment with aqueous TFA in CH2Cl2. Crude te-
trols obtained by either route were used without further purifi-
cation in the perphosphorylation step, with the exception of
benzyl derivative 15 f, which was purified by column chroma-
tography before use.
genolysis protocol, to cleanly furnish tetrasodium tetrakisphos-
phate 22.
Synthesis of IBPPs
With the exception of publications regarding ITPP synthe-
sis,[11,27] only one other myo-inositol cyclic pyrophosphate has
previously been targeted: Ins(1,4,5)P3(PP).[28] Synthesis of IBPPs
requires the preparation of the corresponding tetrakis(triethyl-
ammonium) phosphate salts, which were previously proven
the most suitable starting materials for the IHP pyrophosphate
ring closures, in order to obtain ITPP.[17,19] Triethylammonium
salts 23, required for construction of the Ins(1,3,4,6)P4 series
pyrophosphates, were able to be prepared from the corre-
sponding sodium salts 12 (Scheme 4). Indeed, after cation ex-
change (DOWEX H+, then excess Et3N), 23a–c were obtained
Sequential phosphorylation and deprotection of tetrols 15
was achieved as described above for isomeric analogues 10.
Initially, benzyl tetrakisphosphates 16 were formed in 85–95%
yields and were subsequently hydrogenolyzed. Pure tetrasodi-
um tetrakisphosphates 17 were quantitatively obtained as
well, following filtration of the catalyst and evaporation of the
solvents. Interestingly, complete deprotection of decabenzyl
derivative 17 f was easily achieved (1 atm, 48 h).
Another myo-inositol bicyclohexylidene ketal, 2,3:4,5-pro-
tected compound 18,[26] was used for the preparation of the
final ITP target, racemic Ins(2,3,4,5)P4 (Scheme 3). Benzylation
afforded fully protected derivative 19 (90%). Cyclohexylidene
ketal removal was achieved by adopting a slightly modified lit-
erature protocol (acetyl chloride in CH2Cl2/MeOH),[2a] and the
resulting tetrol 20 was employed in the perphosphorylation re-
action following purification by column chromatography. Phos-
phorylation of 20, upon treatment with dibenzyl N,N-diisopro-
pylphosphoramidate and 1H-tetrazole in CH3CN, followed by
oxidation with m-CPBA, yielded tetrakisphosphate 21 (80%).
The latter was globally deprotected, following the usual hydro-
Scheme 4. Synthesis of 1,6:3,4-myo-inositol bispyrophosphates. Reagents
and conditions: a) 1. DOWEX H+, H2O, RT, 10 min; 2. Et3N, H2O, RT, 10 min;
b) Pd/C, H2 (1 or 3 atm), Et3N, EtOH/H2O (1:1), RT, 20–120 h; c) 1. DCC,
CH3CN/H2O (2:1), 708C, 2–24 h; 2) DOWEX H+, H2O, RT, 10 min; 3. DOWEX
Na+, H2O, RT, 1 h.
from 12a–c in near quantitative yields. On the other hand, acyl
derivatives 23d and e and diol 23 f were not found to be sub-
stantially stable under these conditions. Therefore, they were
prepared directly from benzyl tetrakisphosphates 11 d–e upon
hydrogenolysis in the presence of excess Et3N. Although time
consuming, this procedure results in pure desired salts 23d–f.
Similarly, triethylammonium phosphates 25 were synthesized
from either 17a–c via cation exchange (DOWEX H+, Et3N) or
from 17d–e through catalytic hydrogenolysis in the presence
of Et3N (Scheme 5).
For the intramolecular phosphate couplings, our efforts
mainly focused on obtaining a single and pure product by effi-
cient transformation, without laborious and troublesome purifi-
cations. The best conditions found used a 2:1 mixture of
MeCN/H2O, ensuring maximum solubilization of both substrate
(triethylammonium salt) and coupling reagent (N,N’-dicyclohex-
ylcarbodiimide, DCC) upon heating.[17,19] 1-[3-(dimethylamino)-
propyl]-3-ethylcarbodiimide (EDC) was also tested but did not
result in full consumption of the starting materials. Reactions
were generally monitored by 31P NMR spectra, with consump-
tion of phosphates (at ~0 ppm) giving rise to peaks corre-
sponding to pyrophosphates (at ~ꢁ10 ppm, most often with
Scheme 3. Synthesis of 2,3,4,5-myo-inositol tetrakisphosphate 22. Reagents
and conditions: a) NaH, BnBr, DMF, 08C!408C, 20 h; b) AcCl, CH2Cl2/MeOH
(1:1), RT, 12 h; c) 1. (BnO)2PN(iPr)2, 1H-tetrazole, CH3CN, RT, 24 h; 2. m-CPBA,
CH2Cl2, ꢁ408C!08C, 5 h; d) Pd/C, H2 (1 atm), NaHCO3, EtOH/H2O (1:1), RT,
48 h.
2
an observed JPP). Completion was usually achieved in a reac-
tion time of a few hours to a day.
172
www.chemmedchem.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2011, 6, 169 – 180

Allosteric Effectors of Human Hemoglobin
The reaction of triethylammonium salt 25 f was much more
complicated. Formation of five-membered cyclic phosphates
was again observed, along with pyrophosphates, but it was
not possible to determine the composition of the mixture after
heating for 2, 12, or 24 h. Purification of these complicated
mixtures was not pursued; we also decided not to attempt
preparations of the triethylammonium salt of 22 and its corre-
sponding bispyroposphate.
Binding affinity for hemoglobin and oxygen release
As indicated above, the phosphorylated compounds BPG (1)
and IHP (3), as well as the trispyrophosphate ITPP (4, Figure 1),
bind to Hb and markedly enhance its oxygen releasing capaci-
ty. After successfully synthesizing five series of parent and sub-
stituted myo-inositol tetrakisphosphate (ITP) and myo-inositol
bispyrophosphate (IBPP) derivatives, all final compounds (12a–
f, 17a–f, 22, 24a–f and 26a–e) were evaluated for their effect
on stripped human Hb. In order to determine the abilities of
the compounds to shift the oxygen saturation curve to higher
pO2 values, we assessed the partial pressure of oxygen for half-
saturation (P50) of Hb in the presence of the compounds and
their corresponding dissociation constants (Kd) to Hb. The pro-
cedures for determination of the P50 and Kd values are indicat-
ed in the footnotes of Table 1 and described in the Experimen-
tal Section. The results are listed below in Table 1 and repre-
sented in Figures 3–6.
Scheme 5. Synthesis of 2,3:5,6-myo-inositol bispyrophosphates. Reagents
and conditions: a) 1. DOWEX H+, H2O, RT, 10 min; 2. Et3N, H2O, RT, 10 min;
b) Pd/C, H2 (1 or 3 atm), Et3N, EtOH/H2O (1:1), RT, 36–96 h; c) 1. DCC, CH3CN/
H2O (2:1), 708C, 3–36 h; 2. DOWEX H+, H2O, RT, 10 min; 3. DOWEX Na+, H2O,
RT, 1 h.
The DCC protocol was successfully applied to salts 23
(Scheme 4) and 25 (Scheme 5). Initially formed triethylammoni-
um bispyrophosphates were transformed, without isolation, to
corresponding sodium salts 24 and 26 upon sequential treat-
ment with DOWEX H+ and DOWEX Na+. Special care was
taken to pass the triethylammonium bispyrophosphate salts
through the acidic resin as quickly as possible in order to pre-
vent decomposition. Combined yields for the two steps were
near quantitative. Moreover, the cyclic bispyrophosphates ob-
tained were pure enough (ꢂ98%) to be used directly in bio-
logical experiments.
All compounds were able to shift the Hb oxygenation
curves, from 11% up to 382%, with the relationship between
Hb binding and oxygen release illustrated in Figure 3. In gener-
al, the ability of the compounds to lower the Hb affinity for
oxygen is directly related to their number of negative charges;
that is, a greater number of phosphates corresponds to a
higher P50 value. This trend was observed between the tetrakis-
phosphate (12a–f, 17a–f, and 22) and bispyrophosphate
However, there was a distinct exception regarding the syn-
thesis of bispyrophosphate 24 f (Scheme 4). A 31P NMR spec-
trum recorded after 2 h reflux showed not only the presence
of phosphates and pyrophosphates, but also peaks corre-
sponding to five-membered cyclic phosphates (at ~15 ppm).
Analysis of the mixture composition revealed that the fraction
of desired bispyrophosphate 24 f formed was nearly 80%,
while the rest consisted of phosphates (15%) and cyclic phos-
phates (5%). Heating the reaction for a total of 16 h enabled
us to gain further insight into this reaction. The starting materi-
al was completely consumed, and the resulting composition
was altered, corresponding to a mixture of bispyrophosphate
24 f (44%), a bicyclic phosphate-trans-monopyrophosphate
(44%) and a bicyclic phosphate-bisphosphate (12%). Elucida-
tion of the structures of side products was mainly carried out
by analysis of 31P NMR spectra. Both side products were found
to contain two five-membered cyclic phosphates (31P shift
~15 ppm). The major side product also had a narrow AB quar-
tet in the pyrophosphate region (31P shift ~ꢁ12 ppm) which is
related to a trans-fused system, as was deduced by comparison
with the spectra of ITTP[17] and the other bispyrophosphates
(24 and 26) synthesized in this work. The minor side product
also had two peaks in the phosphates region (31P shift
~0 ppm). These results support the idea that formation of
cyclic pyrophosphates by this protocol, is a reversible process,
and indicates that, in cases where other products could be
formed, it is possible to observe ring opening.
Figure 3. Relationship between Hb–oxygen binding (P50) and Hb dissociation
constants (Kd) for compounds BPG (1), IHP (3), ITPP (4), substituted and un-
substituted Ins(1,3,4,6)P4 (12a–f), Ins(2,3,5,6)P4 (17a–f), Ins(2,3,4,5)P4 (22), Ins-
(1,6:3,4)BPP (24a–f), and Ins(2,3:5,6)BPP (26a–e). The line corresponds to
the linear regression function (R2 =0.841) for all compounds studied. Com-
pounds with the same number of negative charges are grouped within each
bold rectangle in their fully ionized states. All values were extracted from
Table 1.
ChemMedChem 2011, 6, 169 – 180
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
173

J.-M. Lehn et al.
MED
to a higher affinity of the com-
pounds toward Hb, which is di-
rectly linked to their number of
charges (that is, phosphate or
pyrophosphate groups), and re-
sults in tighter electrostatic
docking of the compounds to
the allosteric pocket of Hb.
Table 1. P50 values, Hill coefficients, and dissociation constants (Kd) for stripped human Hb with allosteric effec-
tor compounds.
Compd
myo-ins type
R
P50 [Torr][a]
P50 shift [%][b]
Hill coeff.[c]
Kd [m][d]
Control
BPG (1)
IHP (3)
ITPP (4)
12a
12b
12c
12d
12e
12 f
17a
17b
17c
17d
17e
17 f
–
–
–
–
–
–
–
–
Me
Et
Bu
Ac
Bt
H
Me
Et
Bu
Ac
Bt
H
H
Me
Et
Bu
Ac
Bt
H
Me
Et
10.2ꢀ0.7
17.5ꢀ1.0
57.7ꢀ0.6
22.2ꢀ0.4
31.9ꢀ1.5
33.3ꢀ1.5
46.6ꢀ0.8
29.2ꢀ0.5
42.4ꢀ0.6
49.2ꢀ0.7
24.9ꢀ1.4
28.8ꢀ1.3
45.3ꢀ2.1
31.6ꢀ1.7
44.3ꢀ0.4
32.4ꢀ0.4
40.4ꢀ0.3
13.0ꢀ0.5
12.0ꢀ0.4
12.2ꢀ0.2
12.6ꢀ0.7
12.2ꢀ0.4
19.3ꢀ0.2
12.6ꢀ0.2
11.3ꢀ0.5
14.2ꢀ0.4
14.0ꢀ0.7
12.1ꢀ1.2
–
72
1.2ꢀ0.1
1.2ꢀ0.1
2.3ꢀ0.1
1.6ꢀ0.1
2.0ꢀ0.1
2.0ꢀ0.1
2.2ꢀ0.1
2.0ꢀ0.1
2.1ꢀ0.1
2.1ꢀ0.1
2.0ꢀ0.1
2.0ꢀ0.1
2.2ꢀ0.1
2.1ꢀ0.1
2.2ꢀ0.1
2.0ꢀ0.1
2.1ꢀ0.1
1.4ꢀ0.1
1.6ꢀ0.1
1.6ꢀ0.1
1.7ꢀ0.1
1.6ꢀ0.1
1.8ꢀ0.1
1.5ꢀ0.1
1.5ꢀ0.1
1.6ꢀ0.1
1.7ꢀ0.1
1.5ꢀ0.1
–
8.4ꢁ10ꢁ5
3.2ꢁ10ꢁ8
1.7ꢁ10ꢁ7
1.7ꢁ10ꢁ6
1.0ꢁ10ꢁ6
1.0ꢁ10ꢁ7
6.0ꢁ10ꢁ6
9.6ꢁ10ꢁ8
3.1ꢁ10ꢁ8
2.0ꢁ10ꢁ5
5.6ꢁ10ꢁ6
2.3ꢁ10ꢁ7
2.6ꢁ10ꢁ6
1.9ꢁ10ꢁ7
1.7ꢁ10ꢁ6
2.2ꢁ10ꢁ7
7.6ꢁ10ꢁ4
3.6ꢁ10ꢁ2
4.1ꢁ10ꢁ2
3.3ꢁ10ꢁ2
3.0ꢁ10ꢁ2
1.2ꢁ10ꢁ4
9.0ꢁ10ꢁ2
8.3ꢁ10ꢁ2
2.1ꢁ10ꢁ3
5.2ꢁ10ꢁ3
1.2ꢁ10ꢁ2
466
118
213
226
357
186
316
382
144
182
344
210
334
218
296
27
18
20
24
20
89
24
11
39
Ins(1,3,4,6)P4
Ins(1,3,4,6)P4
Ins(1,3,4,6)P4
Ins(1,3,4,6)P4
Ins(1,3,4,6)P4
Ins(1,3,4,6)P4
Ins(2,3,5,6)P4
Ins(2,3,5,6)P4
Ins(2,3,5,6)P4
Ins(2,3,5,6)P4
Ins(2,3,5,6)P4
Ins(2,3,5,6)P4
Ins(2,3,4,5)P4
Ins(1,6:3,4)BPP
Ins(1,6:3,4)BPP
Ins(1,6:3,4)BPP
Ins(1,6:3,4)BPP
Ins(1,6:3,4)BPP
Ins(1,6:3,4)BPP
Ins(2,3:5,6)BPP
Ins(2,3:5,6)BPP
Ins(2,3:5,6)BPP
Ins(2,3:5,6)BPP
Ins(2,3:5,6)BPP
In regard to specificities ob-
served within distinct series of
ITPs, Ins(1,3,4,6)P4 derivatives
(12a–f) were generally more ef-
fective than their corresponding
Ins(2,3,5,6)P4 analogues (17a–f),
with P50 shift ranges of 186–
382% and 144–334%, respec-
tively. In the case of unsubstitut-
ed ITPs (R=H), the following
trend regarding the positions of
the phosphate groups was ob-
served: Ins(1,3,4,6)P4 (12 f)>
Ins(2,3,4,5)P4 (22)>Ins(2,3,5,6)P4
(17 f) (Figures 4 and 5). For the
IBPP series, all compounds pre-
sented similar oxygen release ef-
fects, exhibiting low P50 shifts
between 11–39% with the ex-
ception of the unsubstituted de-
rivative 24 f, which was markedly
more effective than the other
IBPP analogues and induced a
shift of 89% (Figure 6).
22
24a
24b
24c
24d
24e
24 f
26a
26b
26c
Bu
Ac
Bt
26d
26e
37
19
[a] P50 values were assessed by linear regression analysis from data points obtained between 40 and 60%
blood oxygen saturation. For this screening evaluation, we used a 20:1 compound/Hb ratio. [b] P50 shift was
calculated from the ratio between P50 values in the presence and absence of the compounds (control). [c] Hill
coefficients were determined from the corresponding oxygen saturation curves to evaluate the effect of com-
pounds on Hb–oxygen binding cooperativity. [d] The dissociation constant (Kd) was determined from the equa-
tion:[8b,29] logP50 =constant+(1/n) log(1+C/Kd), where n is the Hill coefficient and C is the concentration of the
effector; standard error is between 5–10%.
The above results may indi-
cate that phosphate groups at
positions C1, C3, C4, and C6 of
series (24a–f and 26a–e). For example, tetrakisphos-
phate inositols 12c (357%) and 17c (344%) were
more effective than their corresponding bispyrophos-
phate analogues 24c (20%) and 26c (39%) (Table 1,
Figures 3–6). Comparison of tetrakisphosphates to
IHP and of bispyrophosphates to ITTP follows this
trend as well, and has previously been observ-
ed[8b,20,29] with regard to the relationship between the
number of negative charges of allosteric effectors
and Hb–oxygen displacement. These observations
are in agreement with the fact that the allosteric
pocket of Hb is rich in positively charged amino acid
residues, located in the b subunits at the entrance to
the Hb central cavity,[8] which favors the tight dock-
ing of polyanions by electrostatic interactions. Be-
cause the electrostatic interaction is directly propor-
tional to the number of negative charges, a higher
overall charge is expected to correspond to the for-
mation of a tighter Hb-effector complex. Indeed,
Figure 4. P50 values for stripped human Hb and corresponding Hill coefficients for com-
pounds BPG (1), IHP (3), ITPP (4), and Ins(1,3,4,6)P4 (12a–f) series. All data were extracted
from oxygen saturation curves, which were measured in triplicate. Error bars represent
there is a direct correlation between the Hb affinity
of the compounds and their oxygen releasing ability.
Figure 3 clearly shows that a larger P50 shift correlates the standard deviation.
174
www.chemmedchem.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2011, 6, 169 – 180

Allosteric Effectors of Human Hemoglobin
differences in oxygen release effects for positional an-
alogues. However, these data indicate that not only
the decreased number of negative charges induces a
significant drop in oxygen release effect; in addition,
the conformational restriction inherent to the pyro-
phosphate scaffold apparently hinders substantial
molecular recognition of IBPP compounds by the Hb
allosteric pocket (Figure 7).
Among alkyl and acyl substituents, a direct rela-
tionship between the length of the carbon chain and
the oxygen release effect was observed for molecular
recognition of the ITPs. For both the Ins(1,3,4,6)P4
(12a–f) and Ins(2,3,5,6)P4 (17a–f) series, we observed
a significant increase in effect with an increase in
substituent group chain length following the trends:
a) among alkyl substituents: butyl (12c, 357%; 17c,
344%)>ethyl(12b, 226%; 17b, 182%)>methyl
(12a, 213%; 17a, 144%); b) among acyl substituents:
butyryl (12e, 316%; 17e, 334%)>acetyl (12d, 186%;
17d, 210%).
Figure 5. P50 values for stripped human Hb and corresponding Hill coefficients for com-
pounds IHP (3), the Ins(2,3,5,6)P4 (17a–f) series, and Ins(2,3,4,5)P4 (22). All data were ex-
tracted from oxygen saturation curves, which were measured in triplicate. Error bars rep-
resent the standard deviation.
Moreover, an interesting but ambiguous effect
seems to positively contribute to the enhancement
of oxygen release within the ITP series (12a–f, 17a–f,
and 22). We have described above the effect of the
chain length of alkyl/acyl substituents, for example,
butyl (12c and 17c) or butyryl (12e and 17e); how-
ever, we have also observed that unsubstituted com-
pounds 12 f, 17 f, and 22 exhibited oxygen release
effects in the same range (382%, 218%, and 296%,
respectively). This remarkable feature may indicate
that the induced fit of long chain or unsubstituted al-
losteric effectors expose either lipophilic or hydrogen
bonding residues, respectively, from the Hb allosteric
binding site (Figure 7). Given the lower comparative
strength of hydrogen bonds and van der Waals inter-
actions, they should be considered as secondary to
electrostatic interactions. Based on the above results,
one can state that the latter constitute the main fea-
ture for tight Hb–effector complex formation, with
secondary, but not negligible, contribution of hydro-
gen bonding and lipophilicity effects.
Figure 6. P50 values for stripped human Hb and corresponding Hill coefficients for com-
pounds IHP (3), ITTP (4), Ins(1,6:3,4)BPP (24a–f), and the Ins(2,3:5,6)BPP (26a–e) series.
All data were extracted from oxygen saturation curves, which were measured in tripli-
cate. Error bars represent the standard deviation.
In accord with these considerations, one may note,
as shown in Figure 7, that within the series contain-
the myo-inositol scaffold are essential for molecular recognition
by the positively charged residues from the Hb allosteric
pocket. Specifically, the interaction of these pharmacophoric
moieties with Hb seems to result in an enhanced oxygen re-
lease effect, which is likely the reason that almost all com-
pounds of general structure 12 exhibit higher P50 shifts when
compared with the corresponding 17 isomers. In general, the
presence of phosphate groups at C2 and C5 does not seem to
contribute to the same extent to the molecular recognition of
ITPs by the Hb allosteric pocket and induces a considerably
lower activity (e.g., compound 22 and 17 f relative to 12 f). It
is important to note that other factors may also favor or disfa-
vor interaction with Hb (see below and Figure 7). In contrast to
ITPs, the IBPP series (24a–f and 26a–e) exhibited no significant
ing 12, 22, and 17, the number of contacts between phos-
phate groups and positively charged domains of the allosteric
pocket are four, two, and two, respectively, corresponding to
the decreasing effectiveness of the compounds. Furthermore,
the ortho disposition of C3 and C4 phosphates in 22 toward
the positively charged regions seems to be more favored than
the para orientation of C3 and C6 in 17. In addition, the
number of interactions with other regions, involving hydrogen
bonding and lipophilicity effects, plays a secondary but impor-
tant role and, again, supports the tighter binding of 12 to Hb.
ChemMedChem 2011, 6, 169 – 180
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
175

J.-M. Lehn et al.
MED
Experimental Section
General
All commercially available grade quality reagents were
used without further purification. All solvents were puri-
fied by standard procedures before use. Dry solvents
were obtained by literature methods and stored over
molecular sieves. DOWEX 50WX8 200 (H+) and DOWEX
MARATHON (Na+) resins were washed with deionized
H2O before use. Whenever possible, reactions were
monitored using commercially available pre-coated TLC
plates (0.25 mm layer thickness; Kiesel gel 60 F254). Com-
pounds were visualized using a UV lamp and/or p-anisal-
dehyde ethanol solution and/or phosphomolybdic acid
(PMA) stain and warming. Column chromatography was
performed in the usual way using Merck 60 silica gel
(40–60 mm). NMR spectra were recorded on a 400 MHz
spectrometer in the indicated deuterated solvents.
Chemical shifts are given in parts per million and J in
hertz, using solvent as an internal reference. Mass spec-
tra were determined by the Service Commun de Spec-
tromꢂtrie de Masse at the Institut d’Ingꢂnierie Supramo-
lꢂculaire. Representative examples following the respec-
tive general synthetic procedures are reported herein.
Details for the remainder of the compounds, including
their spectroscopic data, are given in the Supporting In-
formation.
Procedure A: General procedure for alkylation. The
starting diol (1.2 mmol) was dried under high vacuum
for 4–8 h. Dry DMF (10 mL) was added under N2 atmos-
phere, and the resulting suspension was cooled to 08C;
90% NaH (120 mg, 4.8 mmol) was added in one portion,
and the resulting slurry was stirred at the same tempera-
ture for 1 h. Alkyl iodide (3.5 mmol) was added dropwise,
and the mixture was stirred at room temperature for
12 h. MeOH (300 mL) was added slowly, and the mixture
was stirred at room temperature for an additional 1 h.
CH2Cl2 (25 mL) was added, and the reaction mixture was
Figure 7. Proposed molecular recognition of Ins(1,3,4,6)P4 (12a–f), Ins(2,3,5,6)P4 (17a–f),
Ins(2,3,4,5)P4 (22), Ins(1,6:3,4)BPP (24a–f), and Ins(2,3:5,6)BPP (26a–e) in the allosteric
pocket of Hb, showing essential pharmacophoric moieties within the myo-inositol tetra-
kisphosphate scaffold. Black dots indicate the positions of phosphate groups that inter-
act with a positively charged domain of the allosteric pocket.
Conclusions
washed with H2O (25 mL). The aqueous phase was extracted with
CH2Cl2 (25 mL), and the combined organic phases were washed
with saturated brine (25 mL) and dried with magnesium sulfate
(MgSO4). Solvents were removed under reduced pressure (30–
558C), and the residue was purified by flash column chromatogra-
phy, as indicated, to yield the corresponding dialkyl derivative.
We have presented an efficient and direct synthetic scheme
for the preparation of three series of substituted (alkylated or
acylated) and unsubstituted ITPs [Ins(1,3,4,6)P4, Ins(2,3,5,6)P4
and Ins(2,3,4,5)P4] and their corresponding cyclic bispyrophos-
phates (IBPPs) which uses readily available myo-inositol deriva-
tives as starting materials and results in excellent overall yields
of the targeted compounds. These derivatives may exhibit sig-
nificant biological activities in distinct areas of medicinal re-
search. In particular, they have been found herein to act as al-
losteric effectors of human hemoglobin, inducing enhanced
capacity for oxygen release. These effects increase with the
strength of Hb binding, which may be traced primarily to elec-
trostatic interactions, as they correspond to the number of
negative charges present. Stereochemical and steric factors
also play a significant but secondary role in molecular recogni-
tion. In view of the central role played by hypoxia in numerous
types of diseases, the exploration of myo-inositol phosphate
derivatives represents an important avenue in the search for
substances acting on the oxygenation status of tissues and
may, therefore, have significant potential in the discovery and
development of novel drug candidates.
1,6:3,4-Bis-[O-(2,3-dimethoxybutane-2,3-diyl)]-2,5-di-O-methyl-
myo-inositol (9a). Synthesized from diol 8[25] and methyl iodide,
according to Procedure A. Eluent: heptanes ! 30% ethyl acetate
(EtOAc) in heptanes, 96% yield of amorphous white solid. Rf =0.28
1
(30% EtOAc in heptanes); H NMR (CDCl3, 400 MHz): d=3.99 (dd,
J=10.2, 9.6 Hz, 2H, H-C(4) and H-C(6)), 3.63 (s, 3H, CHOCH3), 3.57
(s, 3H, CHOCH3), 3.55 (t, J=2.3 Hz, 1H, H-C(2)), 3.52 (dd, J=
10.2 Hz, 2.3 Hz, 2H, H-C(1) and H-C(3)), 3.28 (t, J=9.6 Hz, 1H, H-
C(5)), 3.28 (s, 6H, 2ꢁCquatOCH3), 3.25 (s, 6H, 2ꢁCquatOCH3), 1.30 (s,
6H, 2ꢁCquatCH3), 1.29 (s, 6H, 2ꢁCquatCH3); 13C NMR (CDCl3,
100 MHz): d=99.5 (2ꢁCquat); 98.9 (2ꢁCquat), 79.9 (C-5), 77.8 (C-2),
69.7 (C-4 and C-6), 69.4 (C-1 and C-3), 60.8 (CHOCH3), 60.4
(CHOCH3), 47.9 (2ꢁCquatOCH3), 47.7 (2ꢁCquatOCH3), 17.8 (2ꢁ
CquatCH3), 17.6 (2ꢁCquatCH3); HRMS (ESI): m/z calcd for C20H36NaO10
[M+Na]+: 459.2201, found: 459.2203.
Procedure B. General procedure for acylation. The starting diol
(1 mmol) was dried under high vacuum for 8 h. DMAP (12 mg,
176
www.chemmedchem.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2011, 6, 169 – 180

Allosteric Effectors of Human Hemoglobin
0.1 mmol), dry DMF (5 mL), dry CH2Cl2 (5 mL), and Et3N (700 mL,
5 mmol) were added sequentially under N2 atmosphere. The result-
ing solution was cooled to 08C, and the required anhydride
(2.6 mmol) was added dropwise (10 min). The mixture was allowed
to slowly warm to room temperature (2 h) and was stirred for an
additional 14 h. Volatiles were removed under reduced pressure
(30–558C), and the obtained residue was purified by flash column
chromatography, as indicated, to yield the corresponding diacyl
derivative.
under reduced pressure. This sequence was repeated three times
to yield crude desired tetrol. This material was found to be suffi-
ciently pure by H NMR.
1
2,5-Di-O-ethyl-myo-inositol (10b). Synthesized from diethyl ether
9b according to Procedure D (2 h); 99% of white solid. Rf =0.42
(20% MeOH in CH2Cl2); 1H NMR (D2O, 400 MHz): d=3.84 (q, J=
7.0 Hz, 2H, CH2), 3.79 (q, J=7.0 Hz, 2H, CH2), 3.83 (obscured, 1H,
H-C(2)), 3.65 (t, J=9.8 Hz, 2H, H-C(4) and H-C(6)), 3.55 (dd, J=10.2,
2.9 Hz, 2H, H-C(1) and H-C(3)), 3.09 (t, J=9.4 Hz, 1H, H-C(5)), 1.21
(t, J=7.0 Hz, 3H, CH3), 1.19 (t, J=7.0 Hz, 3H, CH3); 13C NMR (D2O,
100 MHz): d=82.9 (C5), 80.2 (C2), 72.1 (C4 and C6), 71.4 (C1 and
C3), 69.9 (CH2), 68.5 (CH2), 14.7 (2ꢁCH3); HRMS (ESI): m/z calcd for
C10H20NaO6 [M+Na]+: 259.1152, found: 259.1148.
2,5-Di-O-acetyl-1,6:3,4-bis-[O-(2,3-dimethoxybutane-2,3-diyl)]-
myo-inositol (9d). Synthesized from diol 8[25] and acetic anhydride
(Ac2O) according to Procedure B. Eluent: heptanes ! 30% EtOAc
in heptanes, 93% yield of light yellow foam. Rf =0.33 (50% EtOAc
1
in heptanes); H NMR (CDCl3, 400 MHz): d=5.23 (t, J=2.9 Hz, 1H,
Procedure E: General procedure for deprotection of 2,3-butane-
dione acetals in aqueous TFA/CH2Cl2. 2,3-Butanedione derivative
(1 mmol) was dissolved in CH2Cl2 (10 mL). Aqueous 90% TFA
(5 mL) was added, and the mixture was stirred at room tempera-
ture for the time indicated in each case. After the volatiles were re-
moved under reduced pressure (408C), absolute EtOH (10 mL) was
added, and the solvent was again removed under reduced pres-
sure. This sequence was repeated three times to yield the desired
crude tetrol. This material was found to be sufficiently pure by
1H NMR.
H-C(2)), 4.90 (t, J=9.9 Hz, 1H, H-C(5)), 3.77 (t, J=10.1 Hz, 2H, H-
C(4) and H-C(6)), 3.53 (dd, J=10.0, 2.9 Hz, 2H, H-C(1) and H-C(3)),
3.03 (s, 6H, 2ꢁOCH3), 3.02 (s, 6H, 2ꢁOCH3), 1.94 (s, 3H, COCH3),
1.90 (s, 3H, COCH3), 1.02 (s, 12H, 4ꢁCquatCH3); 13C NMR (CDCl3,
100 MHz): d=169.5 (CO), 168.9 (CO), 99.3 (2ꢁCquat), 98.7 (2ꢁCquat),
69.8 (C-2), 68.5 (C5), 67.2 (C1 and C3), 66.5 (C4 and C6), 47.4 (2ꢁ
OCH3), 47.0 (2ꢁOCH3), 20.5 (COCH3), 20.2 (COCH3), 17.1 (2ꢁ
CquatCH3), 17.0 (2ꢁCquatCH3); HRMS (ESI): m/z calcd for C22H36NaO12
[M+Na]+: 515.2099, found: 515.2157.
Procedure C: General procedure for benzylation. The starting
diol (3.6 mmol) was dried under high vacuum for 8 h. Dry DMF
(30 mL) was added under N2 atmosphere, and the resulting sus-
pension was cooled to 08C; 90% NaH (345 mg, 14.4 mmol) was
added in one portion. The resulting slurry was stirred at the same
temperature for 1 h, then benzyl bromide (1.5 mL, 12.6 mmol) was
added dropwise and the mixture was left to stir at 408C for 20 h. It
was cooled to room temperature, MeOH (1 mL) was slowly added,
and the mixture was left to stir at room temperature for 1 h.
CH2Cl2 (75 mL) was added, and the reaction mixture was washed
with H2O (75 mL). The aqueous phase was extracted with CH2Cl2
(75 mL), and the combined organic phases were washed with satu-
rated brine (75 mL) and dried (MgSO4). The solvents were removed
under reduced pressure (30–558C), and the residue was purified by
flash column chromatography as indicated.
2,5-Di-O-acetyl-myo-inositol (10d). Synthesized from diacetate 9d
according to Procedure E (3 h); 94% yield of white solid. Rf =0.21
(20% MeOH in CH2Cl2); H NMR (CD3OD, 400 MHz): d=5.45 (t, J=
2.6 Hz, 1H, H-C(2)), 4.78 (t, J=8.8 Hz, 1H, H-C(5)), 3.65 (t, J=9.4 Hz,
2H, H-C(4) and H-C(6)), 3.61 (dd, J=9.6, 2.9, 2H, H-C(1) and H-C(3)),
2.12 (s, 3H, CH3), 2.11 (s, 3H, CH3); 13C NMR (CD3OD, 100 MHz): d=
172.7 (CO), 172.5 (CO), 77.7 (C5), 75.4 (C2), 72.7 (C4 and C6), 71.5
(C1 and C3), 21.2 (CH3), 21.0 (CH3); HRMS (ESI): m/z calcd for
C10H16LiO8 [M+Li]+: 271.1000, found: 271.1006.
1
Procedure F: General procedure for the deprotection of cyclo-
hexylidene ketals with CSA. The starting dicyclohexylidene ketal
(0.54 mmol) was dissolved in CH2Cl2 (2 mL) and a solution of (ꢀ)-
camphor-10-sulfonic acid (32 mg, 0.25 mmol) in MeOH (1 mL) was
added. The mixture was stirred for 24 h at 258C. The resulting pre-
cipitate was collected by filtration. The filtrate was evaporated to
dryness, dissolved in the minimum amount of MeOH (0.3 mL), and
CH2Cl2 (6 mL) was added. This mixture was allowed to stand for
24 h at 258C, and the resulting precipitate was again collected by
filtration. The combined batches were found to be pure product
2,5-Di-O-benzyl-1,6:3,4-bis-[O-(2,3-dimethoxybutane-2,3-diyl)]-
myo-inositol (9 f). Synthesized from diol 8[25] following Proce-
dure C. Eluent: heptanes ! 10% EtOAc in heptanes, 95% yield of
white solid. Rf =0.41 (30% EtOAc in heptanes); 1H NMR (CDCl3,
400 MHz): d=7.51 (d, J=7.9 Hz, 2H, ArH), 7.41 (d, J=7.9 Hz, 2H,
ArH), 7.33–7.28 (m, 4H, ArH), 7.25–7.22 (m, 2H, ArH), 4.86 (s, 2H,
CH2Ph), 4.85 (s, 2H, CH2Ph), 4.19 (t, J=9.8 Hz, 2H, H-C(4) and H-
C(6)), 3.80 (t, J=2.5 Hz, 1H, H-C(2)), 3.57 (dd, J=10.2, 2.5 Hz, 2H,
H-C(1) and H-C(3)), 3.54 (t, J=9.4 Hz, 1H, H-C(5)), 3.26 (s, 6H, 2ꢁ
OCH3), 3.23 (s, 6H, 2ꢁOCH3), 1.33 (s, 6H, 2ꢁCquatCH3), 1.31 (s, 6H,
2ꢁCquatCH3); 13C NMR (CDCl3, 100 MHz): d=139.5 (2ꢁC-iAr), 127.9
(C-oAr), 127.7 (C-oAr), 127.5 (C-mAr), 127.4 (C-mAr), 127.1 (C-pAr),
126.8 (C-pAr), 99.5 (2ꢁCquat), 98.9 (2ꢁCquat), 78.8 (C-5), 76.0 (C-2),
74.9 (CH2Ph), 73.7 (CH2Ph), 69.9 (C4 and C6), 69.2 (C1 and C3), 47.8
(2ꢁCquatOCH3), 47.7 (2ꢁCquatOCH3), 17.8 (2ꢁCquatCH3), 17.6 (2ꢁ
CquatCH3); HRMS (ESI): m/z calcd for C32H44NaO10 [M+Na]+:
611.2827, found: 611.2824.
1
by H NMR unless otherwise mentioned.
(ꢀ)-1,4-Di-O-ethyl-myo-inositol (15b). Synthesized from dicyclo-
hexylidene ketal 14b according to Procedure F; 100% yield of
white solid. Rf =0.39 (15% MeOH in CH2Cl2); 1H NMR (D2O,
400 MHz): d=4.26 (t, J=2.8 Hz, 1H, H-C(2)), 3.85 (dq, J=2.6,
7.0 Hz, 2H, CH2), 3.77 (dq, J=9.6, 7.0 Hz, 1H, CHH), 3.66 (t, J=
9.7 Hz, 1H, H-C(6)), 3.58–3.54 (m, 2H, H-C(3) and CHH), 3.47 (t, J=
9.6 Hz, 1H, H-C(4)), 3.33 (t, J=9.4 Hz, 1H, H-C(5)), 3.28 (dd, J=9.9,
2.6 Hz, 1H, H-C(1)), 1.23 (t, J=7.0 Hz, 3H, CH3), 1.22 (t, J=7.0 Hz,
3H, CH3); 13C NMR (D2O, 100 MHz): d=80.7 (C4), 78.6 (C1), 73.9
(C5), 71.5 (C6), 70.6 (C3), 68.7 (C2), 68.5 (CH2), 65.2 (CH2), 14.7 (CH3),
14.4 (CH3); HRMS (ESI): m/z calcd for C10H20LiO6 [M+Li]+: 243.1415,
found: 243.1406.
Procedure D: General procedure for deprotection of 2,3-butane-
dione acetals in aqueous TFA. 2,3-Butanedione derivative
(1 mmol) was dissolved in aqueous 90% TFA (5 mL), and the mix-
ture was stirred at room temperature for the time indicated. After
volatiles were removed under reduced pressure (408C), absolute
EtOH (10 mL) was added, and the solvent was again removed
Procedure G: General procedure for the deprotection of cyclo-
hexylidene ketals with TFA. The starting dicyclohexylidene ketal
(1 mmol) was dissolved in CH2Cl2 (11 mL); 90% TFA (4.5 mL) was
added, and the mixture was stirred at room temperature for 3 h.
1
At this point, a H NMR spectrum of the sample showed complete
ChemMedChem 2011, 6, 169 – 180
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
177

J.-M. Lehn et al.
MED
conversion into the desired product. After volatiles were removed
under reduced pressure (408C), toluene (10 mL) was added, and
the solvent was again removed under reduced pressure. This se-
quence was repeated two times, a mixture of MeOH/toluene
(20 mL, 1:1) was added, and the solvents were removed under re-
duced pressure. The residue was purified by flash column chroma-
tography (CH2Cl2 ! 7.5% MeOH in CH2Cl2).
CH3), 3.23 (dd, J=9.6, 1.8 Hz, 1H, H-C(1)); 13C NMR (CDCl3,
100 MHz): d=135.8–135.4 (m, 8C, 8ꢁC-iAr), 128.3, 128.24, 128.20,
128.16, 128.124, 128.08, 127.99, 127.9, 127.7, 127.61, 127.60,
127.56, 127.4, 127.3 (C-oAr, C-mAr, C-pAr), 79.4 (brs, C4), 78.0 (t,
J
CP =5.7 Hz, C5), 77.9 (C1), 77.0 (obscured, C6), 75.6 (t, JCP =5.0 Hz,
2
2
C3), 73.5 (d, JCP =6.1 Hz, C2), 69.5 (d, JCP =6.1 Hz, 1ꢁCH2Ph), 69.34
2
2
(d, JCP =5.3 Hz, 1ꢁCH2Ph), 69.29 (d, JCP =5.3 Hz, 2ꢁCH2Ph), 69.12
(d, 2JCP =6.1 Hz, 1ꢁCH2Ph), 69.07 (d, 2JCP =5.3 Hz, 1ꢁCH2Ph), 69.0
(d, 2JCP =5.3 Hz, 1ꢁCH2Ph), 68.8 (d, 2JCP =5.3 Hz, 1ꢁCH2Ph), 60.9
(CH3), 57.6 (CH3); 31P NMR (CDCl3, 162 MHz, 31P-1H decoupled): d=
ꢁ0.79, ꢁ1.39, ꢁ1.48, ꢁ1.96; HRMS (ESI): m/z calcd for
C64H68NaO18P4 [M+Na]+: 1271.3248, found: 1271.3338.
(ꢀ)-1,4-Di-O-acetyl-myo-inositol (15d). Synthesized from dicyclo-
hexylidene ketal 14d according to Procedure G; 99% of white
1
solid. Rf =0.24 (20% MeOH in CH2Cl2); H NMR (CD3OD, 400 MHz):
d=5.18 (t, J=9.8 Hz, 1H, H-C(4)), 4.62 (dd, J=10.0, 2.6 Hz, 1H, H-
C(1)), 4.07 (t, J=2.6 Hz, 1H, H-C(2)), 3.90 (t, J=9.8 Hz, 1H, H-C(6)),
3.59 (dd, J=10.2, 2.9 Hz, 1H, H-C(3)), 3.38 (t, J=9.5 Hz, 1H, H-C(5)),
2.12 (s, 3H, CH3), 2.11 (s, 3H, CH3); 13C NMR (CD3OD, 100 MHz): d=
172.9 (CO), 172.6 (CO), 76.2 (C4), 75.8 (C1), 74.3 (C5), 71.8 (C2), 71.6
(C6), 71.1 (C3), 21.2 (1ꢁCH3), 21.0 (1ꢁCH3); HRMS (ESI): m/z calcd
for C10H16NaO8 [M+Na]+: 287.0737, found: 287.0757.
Procedure I: General procedure for the preparation of sodium
salts by catalytic hydrogenolysis. A starting polybenzyl derivative
(0.8 mmol) was dissolved in a 1:1 mixture of EtOH and deionized
H2O (40 mL). NaHCO3 (270 mg, 3.2 mmol) was added to the result-
ing emulsion, followed by 10% Pd/C (800 mg). This mixture was
stirred vigorously under H2 atmosphere at room temperature (H2
pressure and reaction time are given in each case). The catalyst
was removed by filtration through an LCR/PTFE hydrophilic mem-
brane (0.5 mm); the membrane was washed with a 1:1 mixture of
EtOH and deionized H2O (3ꢁ25 mL). The combined filtrates were
evaporated under reduced pressure (558C), and the resulting resi-
due was dried under high vacuum for 24 h to yield the desired tet-
rasodium salt.
Procedure H: General procedure for phosphorylation. The start-
ing tetrol (1 mmol) was dried under high vacuum for 24 h. A
0.45m solution of 1H-tetrazole in CH3CN (26.7 mL, 12 mmol) was
added under N2 atmosphere at room temperature, followed by
dropwise addition (30 min) of dibenzyl N,N-diisopropylphosphora-
midite (2.2 mL, 6.4 mmol) to the resulting suspension. The ob-
tained slurry was vigorously stirred at room temperature for 24 h.
CH2Cl2 (10 mL) was added, and the mixture was cooled to ꢁ408C.
A solution of 70% m-CPBA (2.07 g, 8.4 mmol) in CH2Cl2 (14 mL)
was added dropwise, and the mixture was stirred at 08C for 5 h.
The mixture was then diluted with CH2Cl2 (120 mL) and successive-
ly washed with a 10% aqueous solution of sodium sulfite (2ꢁ
80 mL), a saturated aqueous solution of NaHCO3 (2ꢁ60 mL), H2O
(60 mL), and saturated brine (60 mL). The organic phase was dried
(MgSO4), and the solvents were removed under reduced pressure
(308C). The obtained residue was purified by flash column chroma-
tography as indicated.
Tetrasodium 1,3,4,6-(2,5-di-O-methyl-myo-inosityl) tetrakisphos-
phate (12a). Synthesized from octabenzyl derivative 11 a accord-
ing to Procedure I (1 atm, 48 h); 97% yield of white glassy solid;
3
1H NMR (D2O, 400 MHz): d=4.33 (q, JHH =3JHP =9.5 Hz, 2H, H-C(4)
and H-C(6)), 4.13 (td, 3JHH =3JHP =9.8 Hz, 3JHH =2.6 Hz, 2H, H-C(1)
and H-C(3)), 4.01 (brs, 1H, H-C(2)), 3.65 (s, 3H, OCH3), 3.59 (s, 3H,
OCH3), 3.32 (t, J=9.3 Hz, and1H, H-C(5)); 13C NMR (D2O, 100 MHz):
d=82.1 (brs, C5), 80.6 (brs, C2), 76.2 (brs, C4 and C6), 74.4 (brs,
C1and C3), 62.1 (OCH3), 60.3 (OCH3); 31P NMR (D2O, 162 MHz, 31P-1H
decoupled): d=0.61 (2P), ꢁ0.04 (2P); HRMS (ESI): m/z calcd for
C8H16Na5O18P4 [M+Na]+: 638.8770, found: 638.8779.
Octabenzyl 1,3,4,6-(2,5-di-O-methyl-myo-inosityl) tetrakisphos-
phate (11a): Synthesized from tetrol 10a according to Procedur-
e H. Eluent: heptanes ! 60% EtOAc in heptane; 91% yield of
Tetrasodium (ꢀ)-2,3,5,6-(1,4-di-O-methyl-myo-inosityl) tetrakis-
phosphate (17a). Synthesized from octabenzyl derivative 16a ac-
cording to Procedure I. (1 atm, 48 h); 99% yield of white solid;
1H NMR (D2O, 400 MHz): d=4.92 (brd, 3JHP =9.6 Hz, 1H, H-C(2)),
4.36 (q, 2JHH =3JHP =9.5 Hz, 1H, H-C(6)), 4.15–4.08 (m, 2H, H-C(3)
and H-C(5)), 3.64 (t, J=9.7 Hz, 1H, H-C(4)), 3.60 (s, 3H, CH3), 3.47 (s,
3H, CH3), 3.33 (brd, J=9.9 Hz, 1H, H-C(1)); 13C NMR (D2O,
1
thick, colorless oil. Rf =0.08 (65% EtOAc); H NMR (CDCl3, 400 MHz):
d=7.35–7.26 (m, 40H, ArH), 5.16–5.00 (m, 16H, 8ꢁCH2Ph), 4.94 (q,
3JHH =3JHP =9.4 Hz, 2H, H-C(4) and H-C(6)), 4.50 (brs, 1H, H-C(2)),
4.25 (ddd, 3JHH =9.6, 2.3 Hz, 3JHP =7.6 Hz, 2H, H-C(1) and H-C(3)),
3.57 (s, 3H, CH3), 3.50 (s, 3H, CH3), 3.25 (t, J=9.3 Hz, 1H, H-C(5));
13C NMR (CDCl3, 100 MHz): d=135.4 (d, 3JCP =6.9 Hz, 4ꢁC-iAr),;
135.1 (d, 3JCP =6.9 Hz, 2ꢁC-iAr), 135.0 (d, 3JCP =6.9 Hz, 2ꢁC-iAr),
128.00, 127.98, 127.9, 127.75, 127.71, 127.6, 127.5, 127.32, 127.25
(C-oAr, C-pAr, C-mAr), 80.2 (C5), 77.6 (C2), 76.2 (t, JCP =6.9 Hz, C4
100 MHz): d=80.4 (d, 3JCP =4.6 Hz, C4), 78.7 (C1), 77.3 (brt, JCP
=
2
3.8 Hz, C3 or C5), 76.0 (brd, JCP =6.4 Hz, C6), 73.8 (brt, JCP =3.1 Hz,
2
C3 or C5), 72.1 (d, JCP =6.1 Hz, C2), 60.0 (CH3), 58.0 (CH3); 31P NMR
2
and C6), 75.2 (brs, C1 and C3), 69.3 (d, JCP =5.3 Hz, 2ꢁCH2Ph), 69.0
(D2O, 162 MHz, 31P-1H decoupled): d=0.77, 0.12, 0.05, ꢁ0.10; HRMS
(ESI): m/z calcd for C8H16Na3O18P4 [MꢁNa]ꢁ: 592.8975, found:
592.8868.
2
2
(d, JCP =5.3 Hz, 2ꢁCH2Ph), 68.8 (d, JCP =6.1 Hz, 2ꢁCH2Ph), 68.7 (d,
2JCP =5.3 Hz, 2ꢁCH2Ph), 59.6 (CH3), 59.2 (CH3); 31P NMR (CDCl3,
162 MHz, 31P-1H decoupled): d=ꢁ1.22 (2P), ꢁ1.74 (2P); HRMS (ESI):
m/z calcd for C64H68NaO18P4 [M+Na]+: 1271.3248, found:
1271.3434.
Procedure J: General procedure for the preparation of triethyl-
ammonium salts from sodium salts.
A tetrasodium salt
Octabenzyl (ꢀ)-2,3,5,6-(1,4-di-O-methyl-myo-inosityl) tetrakis-
phosphate (16a). Synthesized from tetrol 15a according to Proce-
dure H. Eluent: 50% EtOAc in heptanes ! EtOAc; 95% yield of
(0.25 mmol) was dissolved in deionized H2O (10 mL). This solution
was passed through a column loaded with DOWEX 50WX-200 (H+)
(10 g). The resin was further washed with deionized H2O (20 mL).
Collected acidic fractions were transferred to a flask containing
Et3N (490 mL, 3.5 mmol). Volatiles were removed under reduced
pressure at 558C, and the residue was dried under high vacuum
for 12 h to give the corresponding tetrakis(triethylammonium) salt
as a highly hygroscopic solid, which was found to be sufficiently
pure by NMR.
1
thick, colorless oil. Rf =0.22 (EtOAc); H NMR (CDCl3, 400 MHz): d=
7.41 (brd, J=7.6 Hz, 2H, ArH), 7.33–7.17 (m, 38H, ArH), 5.29 (dt,
3JHP =9.1 Hz, 3JHH =1.8 Hz, 1H, H-C(2)), 5.25–4.95 (m, 16H, 8ꢁ
3
3
CH2Ph), 4.86 (q, JHP =3JHH =9.5 Hz, 1H, H-C(6)), 4.46 (q, JHP =3JHH
=
3
3
9.4 Hz, 1H, H-C(5)), 4.36 (ddd, JHH =9.6, 1.7 Hz, JHP =7.9 Hz, 1H, H-
C(3)), 3.66 (t, J=9.6 Hz, 1H, H-C(4)), 3.36 (s, 3H, CH3), 3.27 (s, 3H,
178
www.chemmedchem.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2011, 6, 169 – 180

Allosteric Effectors of Human Hemoglobin
Tetrakis(triethylammonium) 1,3,4,6-(2,5-di-O-methyl-myo-inosi-
tyl) tetrakisphosphate (23a). Synthesized from tetrasodium salt
12a according to Procedure J; 96% yield of a pale-yellow solid;
OCH3), 3.58 (t, J=9.2 Hz, 1H, H-C(5)); 13C NMR (D2O, 100 MHz): d=
2
80.0 (t, JCP =6.1 Hz, C5), 79.8 (t, JCP =6.5 Hz, C2), 77.2 (d, JCP
=
5.3 Hz, C4 and C6), 76.2 (d, 2JCP =5.3 Hz, C1 and C3), 61.9 (CH3),
59.7 (CH3); 31P NMR (D2O, 162 MHz, 31P-1H decoupled): d=ꢁ11.00
and ꢁ11.16 (ABq, 4P, 2JPP =17.8 Hz); HRMS (ESI): m/z calcd for
C8H12Na3O16P4 [MꢁNa]ꢁ: 556.8763, found: 556.8825.
3
1H NMR (D2O, 400 MHz): d=4.27 (q, JHH =3JHP =9.5 Hz, 2H, H-C(4)
and H-C(6)), 4.05 (td, 3JHH =3JHP =10.0 Hz, 3JHH =2.6 Hz, 2H, H-C(1)
and H-C(3)), 3.95 (t, J=2.6 Hz, 1H, H-C(2)), 3.57 (s, 3H, OCH3), 3.51
(s, 3,H, OCH3), 3.23 (t, J=9.4 Hz, 1H, H-C(5)), 3.11 (q, J=7.4 Hz,
24H, 12ꢁNCH2), 1.20 (t, J=7.3 Hz, 36H, 12ꢁCH2CH3); 13C NMR
Tetrasodium
(ꢀ)-1,4-di-O-acetyl-myo-inositol-2,3:5,6-bispyro-
phosphate (26d). Synthesized from triethylammonium salt 25d
according to Procedure L (3 h); 99% yield of white solid (245 mg,
99%); 1H NMR (D2O, 400 MHz): d=5.62 (t, J=9.9 Hz, 1H, H-C(4)),
(D2O, 100 MHz): d=82.1 (brs, C5), 80.5 (brs, C2), 76.1 (t, JCP
=
5.7 Hz, C4 and C6), 74.4 (brd, JCP =5.3 Hz, C1 and C3), 61.9 (OCH3),
60.0 (OCH3), 46.5 (12ꢁNCH2), 8.19 (12ꢁCH2CH3); 31P NMR (D2O,
162 MHz, 31P-1H decoupled): d=0.14 (2P), ꢁ0.51 (2P).
3
5.22 (brd, J=10.2, 1H, H-C(1)), 5.17 (brd, JHP =10.5 Hz, 1H, H-C(2)),
3
3
3
4.65 (td, JHP =3JHH =9.6 Hz, JHH =5.5 Hz, 1H, H-C(6)), 4.50 (td, JHP
=
3JHH =9.6 Hz, JHH =2.9 Hz, 1H, H-C(3)), 4.38 (td, 3JHP =3JHH =9.6 Hz,
3
Procedure K: General procedure for the preparation of triethyl-
ammonium salts by catalytic hydrogenation. An octabenzyl de-
rivative (0.2 mmol) was dissolved in a 1:1 mixture of EtOH and de-
ionized H2O (10 mL). Et3N (220 mL, 1.6 mmol) was added to the re-
sulting emulsion, followed by 10% Pd/C (200 mg). This mixture
was stirred vigorously under H2 atmosphere at room temperature
(H2 pressure and reaction time are given in each case). 1H NMR
spectra were taken and showed that the reaction was complete.
The catalyst was removed by filtration through an LCR/PTFE hydro-
philic membrane (0.5 mm); the membrane was washed with a 1:1
mixture of EtOH and deionized H2O (2ꢁ10 mL). The combined fil-
trates were evaporated under reduced pressure (558C), and the
obtained residue was dried under high vacuum for 24 h to give
tetrakis(triethylammonium) salt.
3JHH =6.1 Hz, 1H, H-C(5)), 2.23 (s, 3H, CH3), 2.22 (s, 3H, CH3);
13C NMR (D2O, 100 MHz): d=173.1 (CO), 173.0 (CO), 76.6 (d, JCP
=
2
2
2
6.1 Hz, C5), 75.0 (d, JCP =5.3 Hz, C6), 73.6 (d, JCP =7.6 Hz, C3), 71.7
2
3
3
(brd, JCP =2.3 Hz, C2), 71.1 (t, JCP =6.5 Hz, C4), 68.6 (t, JCP =6.1 Hz,
C1), 20.6 (CH3), 20.4 (CH3); 31P NMR (D2O, 162 MHz, 31P-1H decou-
2
pled): d=ꢁ9.80 and ꢁ14.67 (ABq, JPP =21.8 Hz, 2P, P-C(2) and P-
2
C(3)), ꢁ11.29 and ꢁ11.49 (ABq, JPP =19.8 Hz, 2P, P-C(5) and P-C(6));
HRMS (ESI): m/z calcd for C10H12Na3O18P4 [MꢁNa]ꢁ: 612.8662,
found: 612.8654.
Preparation of stripped hemoglobin. Human blood was with-
drawn from a healthy, non-smoking volunteer (CDD) in heparinized
microtubes and treated according to the procedure described by
Riggs.[30] Red blood cells were washed three times with 0.85%
saline and lysed by the addition of 1 volume of purified H2O per
volume of packed red blood cells. Hemolysate was kept cold and
passed through a column of Sephadex G-25 (1.0ꢁ20 cm) equili-
brated with 0.1m NaCl + 10ꢁ5 m EDTA pH 7.5 in order to remove
BPG. The concentration of oxy-Hb was assessed by UV/Vis spectro-
photometry (e=58400mꢁ1 cmꢁ1 at 577 nm per oxy-Hb tetramer).
Tetrakis(triethylammonium) 1,3,4,6-(2,5-di-O-acetyl-myo-inosityl)
tetrakisphosphate (23d). Synthesized from octabenzyl derivative
11 d according to Procedure K (1 atm, 20 h); 95% yield of a thick
1
pale-yellow oil; H NMR (D2O, 400 MHz): d=5.58 (brs, 1H, H-C(2)),
4.94 (t, J=9.6 Hz, 1H, H-C(5)), 4.39 (q, 3JHH =3JHP =9.6 Hz, 2H, H-
C(4) and H-C(6)), 4.20 (td, 3JHH =3JHP =9.9 Hz, 3JHH =2.3 Hz, 2H, H-
C(1) and H-C(3)), 3.06 (q, J=7.4 Hz, 20H, 10ꢁNCH2), 2.93 (q, J=
7.4 Hz, 4H, 2ꢁNCH2), 2.10 (s, 3H, COCH3), 2.06 (s, 3H, COCH3), 1.15
(t, J=7.4 Hz, 36H, 12ꢁNCH2CH3); 13C NMR (D2O, 100 MHz): d=
173.5 (CO), 172.7 (CO), 74.5 (brs, C4 and C6), 73.0 (C5), 72.3 (brs,
C1 and C3 and C2), 46.4 and 41.9 (NCH2), 20.9 (COCH3), 20.4
(COCH3), 10.5 and 8.2 (NCH2CH3); 31P NMR (D2O, 162 MHz, 31P-1H de-
coupled): d=ꢁ0.15 (2P), ꢁ0.39 (2P).
General procedure for oxygen equilibration curves measure-
ments. Solutions of the test compounds (100 mm) were prepared
in purified H2O, and the pH was adjusted to 7.0–7.4 prior to incu-
bations with stripped Hb in a 20:1 molar ratio. The mixtures were
diluted in 3 mL of TES-saline buffer (30 mm, 140 mm saline, pH 7.4),
and oxygen equilibrium curves were measured using a Hemox An-
alyzer apparatus (TCS Scientific Corp., USA). P50 values and Hill co-
efficients were calculated by linear regression analysis from data
points obtained between 40 and 60% oxygen saturation.
Procedure L: General procedure for the preparation of pyro-
phosphates. A solution of DCC (165 mg, 0.8 mmol) in CH3CN
The measured values reported here (as well as in our previous
paper on the same topic: Reference [20a]) are the means for ex-
periments performed in triplicate. The errors are indicated in
Table 1.
(6 mL) was added to
a solution of triethylammonium salt
(0.20 mmol) in deionized H2O (3 mL). The mixture was heated at
65–708C for the time given in each case. At that time, a 31P NMR
spectrum of sample showed complete consumption of starting
material. The mixture was left to cool to room temperature and re-
moved by filtration. The filtrate was evaporated at 508C, and the
resulting residue was diluted with deionized H2O (8 mL) and
passed through a column loaded with DOWEX 50WX-200 (H+)
(10 g). The resin was further washed with deionized H2O (20 mL).
The collected acidic fractions were transferred to a flask containing
DOWEX MARATHON (Na+) (22 g). The mixture was stirred vigorous-
ly at room temperature for 1 h. The resin was then filtered, and
H2O was removed by evaporation under reduced pressure (508C).
The residue was found to be pure desired product.
Acknowledgements
A.E.K. thanks the Aristotle University of Thessaloniki (Thessaloniki,
Greece) for financial support of his sabbatical leave. The authors
also thank Dr. Jean-Louis Schmit for 2D NMR assistance. C.D.D.
thanks the Coordenażo de AperfeiÅoamento de Pessoal de Nꢁvel
Superior (CAPES) (Brasilia, Brazil) and the Collꢂge de France
(Paris, France) for postdoctoral fellowships. The authors thank
NormOxys Inc. (Wellesley, MA, USA) for financial support.
Tetrasodium 2,5-di-O-methyl-myo-inositol-1,6:3,4-bispyrophos-
phate (24a). Synthesized from triethylammonium salt 23a accord-
ing to Procedure L (24 h). Yield 99%; white solid; 1H NMR (D2O,
Keywords: hemoglobin · myo-inositol · oxygen release ·
phosphates · structure–activity relationships
3
3
400 MHz): d=4.50 (td, JHH =9.5 Hz, JHP =5.3 Hz, 2H, H-C(4) and H-
C6)), 4.41 (ddd, 3JHH =9.3, 2.6 Hz, 3JHP =6.1 Hz, 2H, H-C(1) and H-
C(3)), 4.02 (t, J=2.6 Hz, 1H, H-C(2)), 3.65 (s, 3H, OCH3), 3.62 (s, 3H,
ChemMedChem 2011, 6, 169 – 180
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
179

J.-M. Lehn et al.
MED
[1] a) D. C. Billington, The Inositol Phosphates, Wiley-VCH, Weinheim, 1993;
b) B. V. L. Potter, D. Lampe, Angew. Chem. 1995, 107, 2085–2125;
Angew. Chem. Int. Ed. Engl. 1995, 34, 1933–1972; c) M. A. J. Ferguson,
S. W. Homans, R. A. Dwek, T. W. Rademacher, Science 1988, 239, 753–
759.
[16] a) M. P. Kunert, J. F. Liard, D. J. Abraham, Am. J. Physiol. 1996, 271 (Part
2), H602-H613; b) S. R. Khandelwal, R. S. Randad, P.-S. Lin, H. Meng, R. N.
Pittman, H. A. Kontos, S. C. Choi, D. J. Abraham, R. Schmidt-Ullrich, Am.
J. Physiol. 1993, 265 (Part 2), H1450-H1453; c) E. P. Wei, R. S. Randad,
J. E. Levasseur, D. J. Abraham, H. A. Kontos, Am. J. Physiol. 1993, 265
(Part 2), H1439-H1443; d) B. A. Teicher, G. Ara, Y. Emi, Y. Kakeji, M. Ikebe,
Y. Maehara, D. Buxton, Drug Dev. Res. 1996, 38, 1–11; e) S. R. Khandelw-
al, B. D. Kavanagh, P.-S. Lin, Q. T. Truong, J. Lu, D. J. Abraham, R. K.
Schmidt-Ullrich, Br. J. Cancer 1999, 79, 814–820.
[2] a) P.-J. Lu, D.-M. Gou, W.-R. Shieh, C.-S. Chen, Biochemistry 1994, 33,
11586–11597; b) S. J. Mills, S. T. Safrany, R. A. Wilcox, S. Nahorski, B. V. L.
Potter, Bioorg. Med. Chem. Lett. 1993, 3, 1505–1510; c) N. T. Burford,
S. R. Nahorski, S.-K. Chung, Y.-T. Chang, R. A. Wilcox, Cell Calcium 1997,
21, 301–310; d) R. A. Wilcox, R. A. Challiss, C. Liu, B. V. Potter, S. R. Na-
horski, Mol. Pharmacol. 1993, 44, 810–817; e) D.-M. Zhu, E. Tekle, EC. Y.
Huang, P. B. Chock, J. Biol. Chem. 2000, 275, 6063–6066; f) R. A. Wilcox,
R. A. Challiss, G. Baudin, A. Vasella, B. V. Potter, S. R. Nahorski, Biochem. J.
1993, 294 (Part 1), 191–194; g) A. L. Hsu, P.-J. Lu, C.-S. Chen, Biochem.
Biophys. Res. Commun. 1998, 243, 653–656; h) I. Ivorra, R. Gigg, R. F.
Irvine, I. Parker, Biochem. J. 1991, 273 (Part 2), 317–321; i) D. J. Gawler,
B. V. Potter, R. Gigg, S. R. Nahorski, Biochem. J. 1991, 276 (Part 1), 163–
167.
[17] K. C. Fylaktakidou, J.-M. Lehn, R. Greferath, C. Nicolau, Bioorg. Med.
Chem. Lett. 2005, 15, 1605–1608.
[18] a) See patent: C. Nicolau, R. Greferath, J.-M. Lehn, K. C. Fylaktakidou,
US 2006/0258626, 2006 [Chem. Abstr. 2006, 145, 483718]; b) C. Kieda, R.
Greferath, C. Crola da Silva, K. C. Fylaktakidou, J.-M. Lehn, C. Nicolau,
Proc. Natl. Acad. Sci. USA 2006, 103, 15576–15581; c) G. Sihn, T. Walter,
J.-C. Klein, I. Queguiner, H. Iwao, C. Nicolau, J.-M. Lehn, P. Corvol, J.-M.
Gasc, FEBS Lett. 2007, 581, 962–966; d) C. Nicolau, J.-M. Lehn, L’Arbre
Vasculaire 2009, 5, 24–25.
[3] a) A. P. Morris, D. V. Gallacher, R. F. Irvine, O. H. Petersen, Nature 1987,
330, 653–655; b) J. W. Putney, Jr., H. Takemura, A. R. Hughes, D. A.
Horstman, O. Thastrup, FASEB J. 1989, 3, 1899–1905; c) J. A. Ely, L. Hu-
nyady, A. J. Baukal, K. J. Catt, Biochem. J. 1990, 268, 333–338; d) A. H.
Guse, E. Roth, F. Emmrich, Biochem. J. 1992, 288 (Part 2), 489–495; e) A.
Lꢃckhoff, D. E. Clapham, Nature 1992, 355, 356–358.
[4] a) M. Hashii, M. Hirata, S. Ozaki, Y. Nozawa, H. Higashida, Biochem. Bio-
phys. Res. Commun. 1994, 200, 1300–1306; b) M. Hashii, M. Hirata, S.
Ozaki, Y. Nozawa, Y. Higashida, FEBS Lett. 1994, 340, 276–280; c) R. F.
Irvine, P. J. Cullen, Curr. Biol. 1993, 3, 540–543; d) P. M. Smith, Biochem.
J. 1992, 283 (Part 1), 27–30.
[5] P. H. Hirst, A. M. Riley, S. J. Mills, I. D. Spiers, D. R. Poyner, S. Freeman,
B. V. L. Potter, A. W. Smith, J. Appl. Microbiol. 1999, 86, 537–543.
[6] a) M. W. Ho, S. B. Shears, K. S. Bruzik, M. Duszyk, A. S. French, Am. J.
Physiol. 1997, 272 (Part 1), C1160–C1168; b) W. Xie, M. A. Kaetzel, K. S.
Bruzik, J. R. Dedman, S. B. Shears, D. J. Nelson, J. Biol. Chem. 1996, 271,
14092–14097.
[7] a) R. Benesch, R. E. Benesch, Biochem. Biophys. Res. Commun. 1967, 26,
162–167; b) R. Benesch, R. E. Benesch, C. I. Yu, Proc. Natl. Acad. Sci. USA
1968, 59, 526–532; c) R. Benesch, R. E. Benesch, Nature 1969, 221, 618–
622.
[19] A. Biolo, R. Greferath, D. A. Siwik, F. Qin, E. Valsky, K. C. Fylaktakidou, S.
Pothukanuri, C. D. Duarte, R. P. Schwarz, J.-M. Lehn, C. Nicolau, W. S. Co-
lucci, Proc. Natl. Acad. Sci. USA 2009, 106, 1926–1929.
[20] a) K. C. Fylaktakidou, C. D. Duarte, A. E. Koumbis, C. Nicolau, J.-M. Lehn,
ChemMedChem 2011, 6, DOI: 10.1002/cmdc.201000366; b) K. C. Fylakta-
kidou, C. D. Duarte, R. Jogireddy, A. E. Koumbis, C. Nicolau, J.-M. Lehn,
unpublished results.
[21] See patent: J.-M. Lehn, S. Pothukanuri, A. E. Koumbis, C. Duarte, C. Nico-
lau, (OxyPlus Inc., Medford, MA, USA), WO/2008/082658, 2008 [Chem.
Abstr. 2008, 149, 153314].
[22] Y. Watanabe, M. Mitani, T. Morita, S. Ozaki, J. Chem. Soc. Chem. Commun.
1989, 482–483.
[23] a) J. L. Meek, F. Davidson, F. W., Jr. Hobbs, J. Am. Chem. Soc. 1988, 110,
2317–2318; b) A. P. Kozikowski, A. H. Fauq, R. A. Wilcox, S. R. Nahorski,
Bioorg. Med. Chem. Lett. 1995, 5, 1295–1300; c) J. Chen, G. Dormꢄn,
G. D. Prestwich, J. Org. Chem. 1996, 61, 393–397; d) S. J. Mills, B. V. L.
Potter, J. Chem. Soc. Perkin Trans. 1 1997, 1279–1286; e) S.-K. Chung, B.-
G. Shin, Y.-T. Chang, B.-C. Suh, K.-T. Kim, Bioorg. Med. Chem. Lett. 1998, 8,
659–662.
[24] a) S. Adelt, O. Plettenburg, G. Dallmann, F. P. Ritter, S. B. Shears, H.-J. Al-
tenbach, G. Vogel, Bioorg. Med. Chem. Lett. 2001, 11, 2705–2708; b) S.-
K. Chung, Y.-U. Kwon, J.-H. Shin, Y.-T. Chang, C. Lee, B.-G. Shin, K.-C. Kim,
M.-J. Kim, J. Org. Chem. 2002, 67, 5626–5637; c) S. J. Mills, K. Backers, C.
Erneux, B. V. L. Potter, Org. Biomol. Chem. 2003, 1, 3546–3556; d) H. A. J.
Carless, K. Busia, Tetrahedron Lett. 1990, 31, 3449–3452; e) M. A. L. Po-
deschwa, O. Plettenburg, H.-J. Altenbach, Eur. J. Org. Chem. 2005,
3103–3115.
[8] a) A. Arnone, Nature 1972, 237, 146–149; b) R. E. Benesch, R. Benesch,
R. Renthal, W. B. Gratzer, Nat. New Biol. 1971, 234, 174–176.
[9] a) R. E. Benesch, R. Edalji, R. Benesch, Biochemistry 1977, 16, 2594–2597;
b) A. Arnone, M. F. Perutz, Nature 1974, 249, 34–36.
[10] B. P. Teisseire, C. Ropars, M. O. Vallez, R. A. Herigault, C. Nicolau, J. Appl.
Physiol. 1985, 58, 1810–1817.
[11] a) D. J. Abraham, M. F. Perutz, S. E. V. Phillips, Proc. Natl. Acad. Sci. USA
1983, 80, 324–328; b) D. J. Abraham, P. E. Kennedy, D. C. Mehanna, D. C.
Patwa, F. L. Williams, J. Med. Chem. 1984, 27, 967–978.
[25] J.-L. Montchamp, F. Tian, M. E. Hart, J. W. Frost, J. Org. Chem. 1996, 61,
3897–3899.
[26] J. P. Vacca, S. J. deSolma, J. R. Huff, D. C. Billington, R. Baker, J. J. Kula-
gowski, I. M. Mawer, Tetrahedron 1989, 45, 5679–5702.
[12] M. F. Perutz, C. Poyart, Lancet 1983, 322, 881–882.
[13] a) I. Lalezari, S. Rahbar, P. Lalezari, G. Fermi, M. F. Perutz, Proc. Natl. Acad.
Sci. USA 1988, 85, 6117–6121; b) I. Lalezari, P. Lalezari, J. Med. Chem.
1989, 32, 2352–2357; c) I. Lalezari, P. Lalezari, C. Poyart, M. Marden, J.
Kister, B. Bohn, G. Fermi, M. F. Perutz, Biochemistry 1990, 29, 1515–
1523.
[14] a) D. J. Abraham, F. C. Wireko, R. S. Randad, C. Poyart, J. Kister, B. Bohn,
J. Liard, M. P. Kunert, Biochemistry 1992, 31, 9141–9149; b) R. S. Randad,
M. A. Mahran, A. S. Mehanna, D. J. Abraham, J. Med. Chem. 1991, 34,
752–757; c) D. J. Abraham, J. Kister, G. S. Joshi, F. C. Wireko, M. K. Safo,
M. C. Marden, C. Poyart, Med. Chem. Res. 1998, 8, 478–486.
[15] a) M. Phelps Grella, R. Danso-Danquah, M. K. Safo, G. S. Joshi, J. Kister,
M. Marden, S. J. Hoffman, D. J. Abraham, J. Med. Chem. 2000, 43, 4726–
4737; b) A. M. Youssef, M. K. Safo, R. Danso-Danquah, G. S. Joshi, J.
Kister, M. C. Marden, D. J. Abraham, J. Med. Chem. 2002, 45, 1184–1195.
[27] L. F. Johnson, M. E. Tate, Can. J. Chem. 1969, 47, 63–73.
[28] a) S. Ozaki, Y. Kondo, N. Shiotani, T. Ogasawara, Y. Watanabe, J. Chem.
Soc. Perkin Trans. 1 1992, 729–737; b) N. J. Noble, D. Dubreuil, B. V. L.
Potter, Bioorg. Med. Chem. Lett. 1992, 2, 471–476; c) S. T. Safrany, S. J.
Mills, C. Liu, D. Lampe, N. N. Noble, S. R. Nahorski, B. V. L. Potter, Bio-
chemistry 1994, 33, 10763–10769.
[29] R. Benesch, R. Edalji, R. E. Benesch, Biochemistry 1976, 15, 3396–3398.
[30] A. Riggs, Methods Enzymol. 1981, 76, 5–29.
Received: October 4, 2010
Revised: October 20, 2010
Published online on December 10, 2010
180
www.chemmedchem.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2011, 6, 169 – 180