1,4-Phenylene-di(N-l-alanylaminomethylphosphonate) a new diaminophosphonate peptide receptor for lysine and arginine

Article abstract of DOI:10.1016/j.molstruc.2007.03.022

New diaminophosphonate peptide receptor composed of two aminomethylphosphonic acid fragments attached to phenyl ring in para position was described. Intermolecular interactions between such host molecule and unprotected cationic forms of lysine and argini

Full text of DOI:10.1016/j.molstruc.2007.03.022

Available online at www.sciencedirect.com
Journal of Molecular Structure 873 (2008) 173–180
www.elsevier.com/locate/molstruc
1,4-Phenylene-di(N-L-alanylaminomethylphosphonate) a new
diaminophosphonate peptide receptor for lysine and arginine
a,
b
a
c
´
*
´
Piotr Młynarz , Adriana Olbert-Majkut , Sylwia Sliwinska , Grzegorz Schroeder ,
a
a
´
Bartosz Bankowski , Paweł Kafarski
a
Department of Chemistry, Wrocław University of Technology, Wybrzez_e Wyspian´skiego 27, 50-370 Wrocław, Poland
b
Faculty of Chemistry, University of Wrocław, F. Joliot-Curie 14, 50-383 Wrocław, Poland
Faculty of Chemistry, A. Mickiewicz University, Grunwaldzka 6, 60-780 Poznan´, Poland
c
Received 18 January 2007; received in revised form 8 March 2007; accepted 10 March 2007
Available online 19 March 2007
Abstract
New diaminophosphonate peptide receptor composed of two aminomethylphosphonic acid fragments attached to phenyl ring in para
position was described. Intermolecular interactions between such host molecule and unprotected cationic forms of lysine and arginine
(guest molecules) were studied by means of NMR spectroscopy and ESI-MS spectrometry in both methanol and water solutions. Unfor-
tunately, no complexation induced shifts in NMR spectra were observed for these host–guest systems in water solution. However, the
performed studies have shown that in methanol complexes in host to guest ratio of 1:1 and 1:2 are formed and they precipitate from the
1
solution, while in water 1:1 complexes are only seen and supramolecular interactions between host and guest discovered using H- and
31P NMR T1 relaxation method suggest that they are present as ionic pairs in aqueous solution. The supramolecular studies were sup-
ported by DFT theoretical studies performed for free receptor and its molecular complexes with Lys and Arg.
ꢀ 2007 Elsevier B.V. All rights reserved.
Keywords: Supramolecular chemistry; Host–guest interaction; Receptor for Arg; Receptor for Lys; NMR spectroscopy; ESI-MS spectrometry; Relax-
ation time
1. Introduction
by designing polymeric phosphonate receptors able to bind
amino acid [6–8].
At the end of last century interest in supramolecular
receptor containing phosphonic acids as binding units has
significantly increased, mostly because of works of Schrad-
er who demonstrated their ability to bind cationic organic
species [1]. The introduced receptors were designed to bind
biologically important molecules such as amino acids, pep-
tides, proteins and neurotransmitters [2–4]. One of the
described supramolecular systems composed of suitable
receptor for Arg was also the subject of extensive theoret-
ical studies [5]. Undoubtedly contributions to phosphonic
receptor development brought Yashima and co-workers
In contrary to phosphonic acids, aminophosphonate
compounds were much rarely of supramolecular chemistry
interest. There is a report on application of aminophosph-
onate functionalized calixarene as carrier for amino acids
via liquid organic membrane [9]. In turn, the adduct forma-
tion between aminophosphonic acids and their derivatives
with arginine and its N- and C-blocked derivatives were
also a subject of extensive ESI-MS investigation [10,11].
In this work the supramolecular properties of molecule
bearing two N-(L-alanyl)aminomethylphosphonic acid
fragments attached to phenyl ring (compound 1, Scheme
1) at para position was studied. We have found that molec-
ular complexes, formed between this host molecule and
lysine or arginine, spontaneously precipitate from the
methanol solution with host to guest ratio of 1:1 and 1:2.
*
Corresponding author. Tel.: +48 71 320 3616; fax: +48 71 328 4064.
E-mail address: piotr.mlynarz@pwr.wroc.pl (P. Młynarz).
0022-2860/$ - see front matter ꢀ 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.molstruc.2007.03.022

174
P. Młynarz et al. / Journal of Molecular Structure 873 (2008) 173–180
0.0025 mol (1.202 g) of tetraethyl 1,4-phenylene-di(amin-
H2O3P
PO3H2
omethylphosphonate) was added with stirring and solution
was left to reach room temperature, followed by slow
warming up to the boiling point. The resulting solution
was washed successively with water, 1 M hydrochloride
acid, water, saturated solution of sodium bicarbonate,
and water. Finally, it was extracted with brine and dried
over sodium sulphate. The solvent was distilled off and
the oily residue representing crude di-phosphono peptide
was dissolved in 20 ml of 40% hydrogen bromide in glacial
acetic acid and left overnight. Then acetic acid and excess
of hydrogen bromide were removed under reduced pres-
sure, the residue dissolved in absolute ethanol and crude
peptide precipitated by addition of propylene oxide. This
procedure resulted in crystalline product, which decom-
posed upon heating over 300 ꢁC. Yield 65%. ESI-MS m/z
O
O
N
N
H2N
H
H
NH2
Scheme 1. Structure of the host molecule.
Their structures despite of lack of chemical induced shift
were determined with the use of longitudal T1 relaxation
time in 1H- and 31P NMR spectroscopy as a intermolecular
interaction indicator. Experimental studies were supported
by DFT calculations of the structure and the interaction
energies of the studied systems.
2. Experimental
1
(ꢀ) 437 (for more data see in the paragraph below); H
2.1. Synthesis of tetraethyl 1,4-phenylene-di(aminomethylphos-
phonate)
NMR (D2O, TSP): d 1.38 and 1.50 (3H each, d,
J = 6.0 Hz, alanyl CH3); 4.10 (2H, m, J=6.0 Hz, alanyl
aCH); 4.99 and 4.96 (1H, d, J = 19.5 Hz, each, phospho-
nate aCH); 7.37 (4H, s, aromatic protons); 31P NMR
(D2O, H3PO4): d 14.51; 14.41. For NMR study in metha-
nol solution 1,4-phenylene-di(N-L-alanylaminomethyl-
phosphonic acids) was transformed into its sodium salt
by dissolving in equimolar aqueous solution of NaOH fol-
lowed by evaporation and drying in vacuo.
The reaction was carried out according to the procedure
reported earlier [12,13]. Terephthalic aldehyde 0.015 mol
(2.01 g) and benzylamine 0.03 mol (3.24 g) were dissolved
in 20 ml of methanol and stirred overnight at room temper-
ature. Resulting Schiff base was filtered and analyzed by
means 1H NMR (CDCl3, TMS): d 4.86 (2H, s, CH2);
7.24–7.42 (10H, m, Ar); 7.86 (4H, s, Ar); 8.42 (2H, s,
CH@N); yield 80%, mp 102 ꢁC. This product was of good
purity and therefore it was dissolved in 50 ml of toluene
and 0.012 mol (1.61 g) of diethylphosphite was added fol-
lowed by refluxing the reaction mixture for 8 h. Then the
volatile components of the solution were evaporated and
crude product was dissolved in dry diethyl ether and pre-
cipitated as oxalate upon addition of appropriate amount
of oxalic acid. The formed precipitate was filtered, dried
and suspended in dichloromethane (DCM). The suspen-
sion was washed with 1 M NaOH until all solid material
was dissolved. After evaporation of the solvent crude prod-
uct was dissolved in 1.7 M solution of hydrogen chloride
gas in methanol and hydrogenated over Pd/C catalyst
under atmospheric pressure during 8 h. The progress of
hydrogenation reaction was controlled by 31P NMR spec-
troscopy. When the hydrogenation was completed, reac-
tion mixture was filtered through 4 cm layer of Cellite
and solution was evaporated under reduced pressure. The
crude hydrochloride was purified by recrystallization from
methanol/ethyl acetate mixture.
2.3. Spectroscopic studies
Host and guest concentrations were adjusted to 0.4–
10 · 10ꢀ3 mol dmꢀ3
.
1H and 31P NMR spectra were
recorded at 300 MHz DRX Bruker and 600 MHz Bruker
Avance instruments at 298 K in D2O and CD3OD using
coaxial tube filled with TSP [trimethylsilylpropenesulfonic
acids sodium salt] or phosphorous acid dissolved in D2O
as external standards. Selective ROESY experiment was
recorded applying mixing time ranging from 250 to 400 ms.
2.4. Precipitation studies
Sodium salt of phosphono peptide host was dissolved in
methanol and appropriate portion of dihydrochloride salt
of guest (in MeOH) was added adjusting stoichiometry to
1:1, 1:3 and 1:20. Resulting precipitates were filtered,
washed with methanol and dried. Each precipitate was dis-
solved in D2O and examined by means of NMR spectros-
copy. pH was controlled using Mettler–Toledo pH-meter
supported with semi-micro combined electrode. The host
to guest ratios in precipitates were determined on the basis
2.2. Synthesis of 1,4-phenylene-di(N-L-alanylaminomethyl-
phosphonate)
1
of integration of their 1D H NMR spectra after dissolu-
tion of the resulting complexes in D2O.
The synthesis of this phosphono peptide was carried out
according to the general procedure described earlier [14]. A
solution of 0.005 mol (0.946 g) of the Boc-Ala and
0.0052 mol (0.72 ml) of triethylamine in 40 ml of dry chlo-
roform was cooled to about ꢀ5 ꢁC and treated with 0.53 ml
(0.0055 mol) of ethyl chloroformate. After 40 min
2.5. ESI spectrometry
The spectroscopic analyses were carried out using
Waters/Micromass (Manchester, UK) ZQmass spectrome-
ter equipped with Harvard Apparatus syringe pump. The

P. Młynarz et al. / Journal of Molecular Structure 873 (2008) 173–180
175
measurements were performed in aqueous and methanol
solutions when setting up the concentrations of host and
viously described procedures [14]. Since the starting amin-
ophosphonate was of meso configuration this procedure
resulted in product of (RS,SS) configuration and thus
exhibiting two peaks of 1:1 ratio in 31P NMR.
guest in the range of 5 · 10ꢀ4– 5 · 10ꢀ5 mol dmꢀ3
.
2.6. Computational details
3.2. Characterization of receptor
The full conformational search was undertaken applying
DFT calculations using Gaussian 03 package of computers
code [15] and employing the Becke’s three parameter
hybrid functional with the non-local correlation of Perdew
(B3PW91) [16,17]. The structure of receptor and both
amino acids as well as their complexes were fully optimized
using DGDZVP (Density-Gaussian double-zeta-split-
valence plus polarization functions) basis set [18]. The
interaction energy was computed as the difference in total
energy between the complex and its subunits (receptor
and amino acid) at the frozen structures as adopted in
the complex according to the following formula: DEinterac-
The structure of the synthesized host phosphono peptide
1 was proven by 1H and 31P NMR spectroscopy and ESI MS
spectrometry. According to the literature the performed syn-
thesis imposes formation of meso form of compound 1 [12].
Attachment of two L-alanine (S-alanine) residues to amine
groups of this compound resulted in differentiation of proton
signals of aminophosphonate a-carbon, which appeared as
two doublets corresponding to RS and SS fragment of the
molecule. Also two sets of signals deriving from two methyl
groups of alanine were visible in 1H NMR spectra. The addi-
tional evidence for the presence of the two different chiral
centers came from 31P NMR spectra where two equipotent
singlets of phosphorous signals were observed. The 1D
ROESY selective experiments have shown that one of the
methyl groups interacts stronger with phenyl protons than
the other one (spectrum not shown).
tion = Ecomplexꢀ(Ereceptor + Eaminoacid frozen
)
. All proton
transfer calculations were carried without geometry optimi-
zations at each point on the PES.
3. Results and discussion
Obtained phosphono peptide 1 exhibits very good solu-
bility in water in the wide range of pH from 1 to 13 reveal-
ing four pKa values: 8.93 and 8.22 (for two NH2 groups of
alanine), and 6.67 and 5.74 (responding to two
PO3Hꢀgroups) [19]. The quite significant divergence in
3.1. Synthesis of host molecule 1
Host phosphono peptide 1 was synthesized in a
sequence of reactions shown in Scheme 2, according to pre-
N
N
O
O
H
H
NH2
H
H
HP(O)(OEt)2
H2/Pd-C
H2N
NH2
NH
Et2O3P
HN
PO3Et2
Et2O3P
PO3Et2
1./ BocAlaOH/ClCOOEt/Et3N
2./ HBr/AcOH
H2O3P
PO3H2
O
O
N
N
H2N
H
H
NH2
1
Scheme 2. Synthetic scheme leading to host molecule 1.

176
P. Młynarz et al. / Journal of Molecular Structure 873 (2008) 173–180
the respective pKa values may be caused by different inter-
molecular interactions between phosphonic and ammo-
nium groups, assumption which was supported by
theoretical studies.
6.4 were performed. Obtained results were compared with
those obtained for individual compound 1 and solutions
of lysine and arginine performed in D2O at the same pD
(Table 2). Data presented in Table 2 strongly suggest that
decrease of T1 values observed for proton and phospho-
rous atoms reflect existence of host–guests interactions.
The biggest effect was observed in the case of a- and e-pro-
tons of lysine (decrease in T1 value from 3.05 to 2.0 s and
1.55 to 1.05 s respectively), a and d protons of arginine,
aryl protons of phenyl ring and a-protons of amin-
ophosphonate, and a-protons of alanyl residues of com-
pound 1. Interestingly, phosphorous signal is also
remarkably affected what confirms participation of phos-
phonic moieties in supramolecular recognition. As seen
from Table 2, the observed effect is stronger when precipi-
tate was freshly dissolved in water than this found for mix-
tures of phosphono peptide 1 and corresponding amino
acids (ratio 1:2). This finding can suggests that complexes
precipitated form exhibit in methanol stronger network
of binding forces.
ESI-MS studies of the receptor in aqueous solution
revealed main m/z signals at 437 (ꢀ) [R] and 218 (ꢀ) [R]
and 461 (+) [R+Na]. Its sodium salt in methanol exhibits
three main forms 503 (ꢀ) [R+3Na], 481 (ꢀ) [R+2Na]
and 459 (ꢀ) [R+Na] and 227 (+) [R+Na], whereas in water
only following m/z signals: 459 (ꢀ) [R+Na], 437 (ꢀ) [R],
218 (ꢀ) [R] and 191 (+) [R+Na] were observed. This shows
that in methanolic solutions the main form of compound 1
is a molecule which binds two sodium ions, while in the
aqueous solution only one Na+ ion is bound. This finding
indicates that affinity of compound 1 towards positively
charged species is higher in methanol than it is in water.
3.3. Host guest molecular complexes of compound 1 with
lysine and arginine
The clear decrease in relaxation time T1 observed for
protons as well as for phosphorous atom might be caused
by change of correlation time (increase) of both molecules
due to complexation processes between relatively small
host and guest [20–23]. Also, that example once more
shows that in the case when complexation-induced chemi-
cal shifts are too small to be significant the studies of relax-
ation times is very useful [23].
The standard titration of host molecule by basic amino
1
acids in water studied by H and 31P NMR spectroscopy
did not demonstrate complexation induced shifts neither
in the case of proton nor phosphorous spectra. The only
trace of interaction could be deduced from changes in the
shape of phosphorous signals during titration. Therefore
these systems were additionally studied in less polar solvent
– methanol. Addition of the methanolic solution of lysine
or arginine dihydrochlorides to phosphono peptide 1 dis-
solved in the same medium caused precipitation of supra-
molecular complexes. Systematic studies of precipitates
obtained using various host to guest molar ratios have
shown the formation of two molecular complexes of host
to guest ratio 1:1 and 1:2. This is despite the fact that 20-
fold excess of amino acids dihydrochloride caused signifi-
cant decrease in pD of precipitate (Table 1), which may
suggests additional binding of HCl, or more likely the pres-
ence of protonated COOH group.
The ESI-MS study of the precipitates dissolved in water
revealed formation of molecular complexes of host to guest
1:1 ratio, which is indicated by presence of signals at 583
(ꢀ) (peptide 1–lysine) and 611 (ꢀ) (peptide 1–arginine).
Measurements performed in methanolic solution obtained
after filtration of precipitate showed the presence of 1:2
complex of compound 1 and lysine, with the signal at m/
z 731 [R+2Lys].
The formed complexes seem to be weakly bound since
ROESY and NOESY (2D as well 1D selective) measure-
ments did not equivocally confirm those interactions. The
systematic studies of the remaining precipitates, namely
those appearing at lower pD values showed that T1 values
of protons as well as phosphorous are very close to these
observed in the case of free phosphono peptide. This indi-
cates that the forming supramolecule is most likely stable
only in restricted pH range. Anyway, the observed differ-
ences in relaxation times suggest that the formation of
supramolecules is governed by interactions between phos-
phonic acid groups and ammonium and/or guanidinium
ions of amino acids. Thus, the formed supramolecules
might be rather considered as ionic pairs. That observation
may confirm the theoretically calculated model for 1:1
complexes (vide supra) forming preferentially in methanol
as well as in aqueous solutions. The detected involvement
of Ala residue of the host molecule can not be easily
explained, however there are two options of its interaction
with lysine and arginine. They may result either from inter-
The T1 relaxation measurements of obtained precipi-
tates dissolved in water at pD (pH measured in D2O) of
Table 1
The dependence of host to guest ratio in the formed complexes from their starting ratio
Molar ratio of peptide 1
to Lysa or to Arg
Molar ratio of peptide 1
to Lys in precipitate
pD of
precipitate
Molar ratio of peptide 1
to Arg in precipitate
pD of
precipitate
1:1
1:3
1:20
a
1:1.5
1:2
1:2.5
7.80
6.39
2.54
1:1
1:1.5
1:2
7.99
6.17
2.50
concentration of phosphono peptide 1 sodium salt was set at 0.5 mM.

P. Młynarz et al. / Journal of Molecular Structure 873 (2008) 173–180
177
Table 2
T1 relaxation measurements of proton and phosphorous signals of compound 1, lysine and arginine in D2O at the 298 K, pH 6.1–6.4
T1[s] for 1H
T1[s] for 31P
Free species
Lys
Arg
Peptide 1
3.05 (a)
2.76 (a)
1.55 (Ar)
1.55 (e)
0.98 (d)
1.82 (a)
2.66 (a-Ala)
0.88 (CH3–Ala)
0.75 (CH3–Ala)
0.85
Mixed species
Lys fragment
Peptide 1 fragment
Arg fragment
2.64 (a)
1.20 (Ar)
2.17 (a)
1.41 (Ar)
1.34 (e)
1.77 (a)
0.86 (d)
1.67 (a)
1.96 (a-Ala)
2.26 (a-Ala)
0.76 (CH3–Ala)
0.77 (CH3–Ala)
0.74 (CH3–Ala)
0.72 (CH3–Ala)
0.74
0.77
Peptide 1 fragment
Dissolved precipitate
Lys fragment
Peptide 1 fragment
Arg fragment
2.00 (a)
1.21 (Ar)
1.63 (a)
1.27 (Ar)
1.05 (e)
1.52 (a)
0.78 (d)
1.58 (a)
2.01 (a-Ala)
2.15 (a-Ala)
0.77 (CH3–Ala)
0.76 (CH3–Ala)
0.74 (CH3–Ala)
0.72 (CH3–Ala)
0.57
0.70
Peptide 1 fragment
actions between carboxylic group of amino acid and
ammonium group of phosphono peptide yielding 1:1
molecular complexes, or from that amino acid are bridging
host phosphonate with ammonium group of the next host
by their ammonium and carboxylic groups forming
pseudo-polymeric species. This second assumption might
be ruled out because polymeric species were not observed
in ESI-MS spectrometry.
imental observations of the formation of 1:1 and 1:2 host–
guest complexes between this peptide and studied amino
acids. Hence, the syn structure of the studied compound
probably allows the formation of the 1:1 complex, while
the anti conformer favors formation of 1:2 complexes with-
out sterical constrains.
3.4.2. Phosphono peptide structure in acidic solution
For simulation of the structure of compound 1 in acidic
3.4. Theoretical calculations
environment optimization of the system bearing monoprot-
2ꢀ
3.4.1. Phosphono peptide structure in high pH
onated PO3
(PO3Hꢀ) and protonated NH2ðNH þÞ
3
In order to obtain the structure analogous to that
observed experimentally in methanolic solution, the
unprotonated PO32ꢀ and NH2 groups in the optimized pep-
tide 1 have been taken into consideration. Two stationary
points were found on the Potential Energy Surface of the
studied compound. The global minimum was predicted
for the conformer exhibiting anti relation of the two phos-
phonate groups with respect to the aryl ring plane (Fig. 1)
with the PCCP tortional angle equal to ꢀ162ꢁ. Anti con-
former is only by 3.89 kJ/mol more stable than the second
confꢀo2rmer that is characterized by syn arrangement of the
PO3 groups (Fig. 2). In this case the PCCP dihedral angle
is predicted to be equal +7ꢁ. As can be seen from Figs. 1
and 2 both arrangements of the phosphonate groups
allowed formation of the internal hydrogen bonds between
groups has been performed. In this case significant changes
in the internal arrangement and interactions of the individ-
ual parts on the R and S branches are observed in compar-
ison with non protonated system. In Fig. 3 the DFT
optimized structure of the syn peptide is presented.
As can be seen the structure shows the presence of the
two strong hydrogen bonds in the aminomethylphospho-
nate group oþf R configuration. The first one is formed
between NH3 and C@O groups in alanyl residue. In this
˚
case the N–H bond is elongated by ca. 0.77 A as compared
with the non-bonded N–H group. It is interesting to notice
that this interaction is characterized by almost linear
hydrogen bond with Oꢁ ꢁ ꢁH–N of the angle equal to 171ꢁ.
For the described R-moiety formation of additional hydro-
gen bond between oxygen atom of the phosphonate and H
atom of amide group is predicted. This interaction is char-
NH2 and C@O groups (NAHꢁ ꢁ ꢁO@C) as well as between
2ꢀ
˚
amide proton (NH) and PO3
moiety (NAHꢁ ꢁ ꢁOAP).
acterized by Hꢁ ꢁ ꢁO distance equal to 1.80 A.
For both conformers Hꢁ ꢁ ꢁO distance in N–Hꢁ ꢁ ꢁO@C
The different type of interaction is predicted for the S-
aminomethylphosphonate fragment of compound 1. In this
case very strong hydrogen bond between NH3þ and HPO3ꢀ
groups is observed. This interaction is accompanied by the
˚
hydrogen bridge is equal to 2.13 A for R-phosphono pep-
˚
tide fragment and 2.25 A for S-center. In turn, the Hꢁ ꢁ ꢁO
distance in N–Hꢁ ꢁ ꢁO–P interaction is computed to be
˚
˚
1.86 and 1.80 A for R and S aminomethylphosphonate
significant elongation of the N–H bond up to 1.72 A and
˚
fragments, respectively, and is the same in the case of both
confꢀo2rmers. This clearly indicates that the rotation of the
shortening of the Hꢁ ꢁ ꢁO distance up to 1.03 A. These val-
ues clearly indicate that the proton is bounded rather by
phosphonate than by amine group.
PO3 groups around the C –C bound does not influence
*
ar
þ
the spatial arrangement and internal interactions of both
phosphonic acid groups. This fact could explain the exper-
The DFT predicted inequivalence of NH3 and HPO3ꢀ
groups of each phosphono peptide unit, which is reflected

178
P. Młynarz et al. / Journal of Molecular Structure 873 (2008) 173–180
Fig. 3. The optimizedstructure of the lowestcalculated energyfor peptide 1.
Fig. 1. The optimized structure and calculated energy of compound 1 as
anti conformer.
Fig. 2. The optimized structure and calculated energy of compound 1 as
syn conformer.
Fig. 4. The optimized structure of the arginine-peptide 1 complex.
in significant difference of the experimentally determinated
pKa values being 0.71 for NH2 group and 0.93 for HPOꢀ.
3
3.4.3. Complexes of phosphono peptide 1 with arginine and
lysine
The Potential Energy Surface studies performed for the
1:1 receptor-Arg and receptor-Lys systems indicate the
existence of only one stable structure in which two hydro-
gen bonds stabilizing their structure are formed. As the
starting point for optimizations the non-protonated syn
phosphono peptide (Fig. 2) and protonated arginine and
lysine has been taken into consideration. In both cases
the presence of interactions between negatively charged
phosphonic groups and cationic units of amino acid was
assumed. As might be seen in Figs. 4 and 5 both arginine
and lysine molecules are bidentately bound to phosphono
peptide 1. The spatial arrangement of arginine-peptide 1
complex allows the formation of the hydrogen bonds
between the phosphonate moieties of aminomethylphosph-
onate groups and hydrogen atoms of the a-amine and
guanidinium groups of arginine. Analogical pattern of
interactions is predicted for complex formed between com-
pound 1 and lysine.
Fig. 5. The optimized structure of the lysine-peptide 1 complex.
plexes is ca. 4 kJ/mol and indicates that the strength of
interactions is comparable for complexes of both amino
acids.
For the PO32ꢀ ꢁ ꢁ ꢁ H3NþA Ca interaction in arginine–
peptide 1 complex the calculated hydrogen bridge length
˚
and angle areþ equal to 3.47 A and 157ꢁ, while for the
PO32ꢀ ꢁ ꢁ ꢁ H2N ACe system the corresponding values are
The DFT calculated interaction energy computed for the
studied systems after the proton transfer is equal to 104.06
and 99.86 kJ/mol for arginine and lysine complexes, respec-
tively. The energy difference between both molecular com-
˚
2.79 A and 167ꢁ.
The same tendency is observed for the lysine–compound
1 system. The calculated length of the hydrogen bridge is

P. Młynarz et al. / Journal of Molecular Structure 873 (2008) 173–180
179
-2656.05
-2656.06
-2656.07
-2656.08
-2656.09
-2656.10
-2656.11
-2656.12
Cε-N...H...O-P
Cα-N...H...O-P
A
B
ΔE=42.7 kJ/mol
ΔE=21.7 kJ/mol
ΔE=62.9 kJ/mol
ΔE=34.2 kJ/mol
0.8
1.0
1.2
1.4
P-O...H distance [Å]
Scheme 3. Energy barriers for proton transfer between arginine and receptor for (A) CeANꢁ ꢁ ꢁHꢁ ꢁ ꢁOAP and (B) CaANꢁ ꢁ ꢁHꢁ ꢁ ꢁOAP system.
1.6
1.8
2.0
2.2
Cε-N...H...O-P
Cα-N...H...O-P
-2546.54
-2546.56
-2546.58
-2546.60
-2546.62
-2546.64
A
B
ΔE=92.1 kJ/mol
ΔE=13.2 kJ/mol
ΔE=63.5 kJ/mol
ΔE=38.7 kJ/mol
0.8
1.0
1.2
1.4
1.6
1.8
2.0
2.2
2.4
2.6
2.8
P-O...H distance [Å]
Scheme 4. Energy barriers for proton transfer between arginine and receptor for (A) CeANꢁ ꢁ ꢁHꢁ ꢁ ꢁOAP and (B) CaANꢁ ꢁ ꢁHꢁ ꢁ ꢁOAP system.
þ
equal to 3.14 A for PO32ꢀ ꢁ ꢁ ꢁ H3N ACa and 2.80 A for
PO32ꢀ ꢁ ꢁ ꢁ H3NþACe. It is clearly indicated that in both
studied systems the double proton transfer reaction
appears. Schemes 3 and 4 show the results of the Potential
Energy Surface studies for the varying N–H bond length in
the CaANAHꢁ ꢁ ꢁOAP and CeANAHꢁ ꢁ ꢁOAP moieties per-
formed for both arginine and lysine complexes. As one can
see for donor–acceptor pair, the second minimum on the
right (for proton attached to the ammonium group) is
always located higher than the first minimum (for hydro-
equal 62.9 kJ/mol, while for Lys–peptide 1 system to be
63.5 kJ/mol. On the other hand, in the case of Ce–
Nꢁ ꢁ ꢁHꢁ ꢁ ꢁO–P entities the differences between two minima
on the PES are almost twice smaller as the previous ones
and are equal to 34.2 kJ/mol (for arginine complex) and
38.7 kJ/mol (for lysine complex).
˚
˚
The predicted energy barriers for proton transfer
between amino acids and peptide 1 are distinctly higher
2ꢀ
in the case ofþhydrogen bonding between PO3 group
and CaANH3
than these calculated for PO32ꢀ ꢁ ꢁ ꢁ
CeAH3Nþ and PO32ꢀ ꢁ ꢁ ꢁ CeAH2Nþ systems. The calculated
energy barrier for Ca–Nꢁ ꢁ ꢁHꢁ ꢁ ꢁO–P system is equal to ca.
43 and 92 kJ/mol for Arg and Lys complex, respectively,
while those predicted for Ce–Nꢁ ꢁ ꢁHꢁ ꢁ ꢁO–P hydrogen bond
is equal to 22 and 13 kJ/mol for Arg–peptide 1 and Lys–
peptide 1 system, respectively. These data clearly indicate
2ꢀ
gen atom attached to the PO3 moiety) as the proton is
more stable when placed on phosphonic group than on a
hydrogen bonded arginine or lysine. In Schemes 3 and 4
the significant energy differences between individual min-
ima are observed for the Ca–Nꢁ ꢁ ꢁHꢁ ꢁ ꢁO–P hydrogen
bridge. For Arg–peptide 1 complex DE is calculated to be

180
P. Młynarz et al. / Journal of Molecular Structure 873 (2008) 173–180
that proton transfer process is much easier in the case of
the hydrogen bonds formed between ammonium or guan-
idinium groups of the side chain of amino acid as com-
References
[1] T. Schrader, A.D. Hamilton (Eds.), Functional Synthetic Receptors,
Wiley-VCH verlag GmbH& Co. KGaA, 2005, pp. 48–164.
[2] M. Wehner, T. Schrader, P. Finocchiaro, S. Failla, G. Consiglio, Org.
Lett. (2000) 605.
[3] S. Rensing, T. Schrader, Org. Lett. (2002) 2161.
[4] S. Kolusheva, O. Molt, M. Herm, T. Schrader, R. Jelinek, J. Am.
Chem. Soc. 127 (2005) 10000.
[5] B. Kirchner, M. Reiher, J. Am. Chem. Soc. 127 (2005) 8748.
[6] H. Onouchi, K. Maeda, E. Yashima, J. Am. Chem. Soc. 123 (2004)
7441.
þ
pared with the Ca ꢀ NH3 moieties. This results seems to
be slightly surprising since the experimental pKa values
þ
determinated for Ca ꢀ NH3 entity is always lower than
those for guanidinium or ammonium Arg or Lys side chain
group.
4. Conclusion
[7] H. Onouchi, D. Kashiwagi, K. Hayashi, K. Maeda, E. Yashima,
Macromolecules 37 (2004) 5495.
[8] H. Onouchi, T. Hasegawa, D. Kashiwagi, H. Ishiguro, K. Maeda, E.
Yashima, Macromolecules 38 (2005) 8625.
[9] I.S. Antiopin, I.I. Stoikov, E.M. Pinkhassik, N.A. Fitseva, I. Stibor,
A.I. Konovalow, Tetrahedron Lett. 38 (1997) 5865.
[10] H. Shung, W. Linder, Int. J. Mass Spectrom. 235 (2004) 213.
[11] W. Linder, H. Shung, Int. J. Mass Spectrom. 241 (2005) 11.
[12] J. Lewkowski, M. Rzezniczak, R. Skowronski, Heteroatom Chem. 11
(2000) 144.
In this work we demonstrated that peptide derivative of
diaminophosphonate compound 1 (in the form of sodium
salt) is able to bind non-blocked 2HClxArg and 2HClxLys
in methanol solution causing formation of molecular com-
plexes of the host-to-guest ratio of 1:1 and 1:2. Moreover,
the supramolecular interaction in aqueous solutions
between peptide 1-Arg and peptide 1-Lys were revealed
by a considerable change in spin-lattice relaxation time
T1 calculated for protons and phosphorous atoms of inves-
tigated systems. These studies were confirmed by ESI-MS
spectrometry. Theoretical calculations performed for
unprotonated form of guest 1 have shown the existence of
two conformers (syn and anti), which may explain the forma-
tion of two kinds of complexes in methanolic solutions. On
the other hand, in the case of tetraprotonated receptor 1
the net of intramolecular hydrogen bonds was predicted,
which might be a cause of difference in pKa values found
for the same kind of dissociable group by potentiometry.
Interesting results were obtained by calculating proton
transfer energy barrier from a-ammonium group of Arg or
[13] J. Barycki, R. Gancarz, M. Milewska, R. Tyka, Phosphorus Sulfur
Silicon 105 (1995) 117.
[14] B. Lejczak, P. Kafarski, H. Sztajer, P. Mastalerz, J. Med. Chem. 29
(1987) 2212.
[15] Gaussian 03, Revision C.02, M.J. Frisch, G.W. Trucks, H.B. Schlegel,
G.E. Scuseria, M.A. Robb, J.R. Cheeseman, J.A. Montgomery Jr., T.
Vreven, K.N. Kudin, J.C. Burant, J.M. Millam, S.S. Iyengar, J.
Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega,
G.A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R.
Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao,
H. Nakai, M. Klene, X. Li, J.E. Knox, H.P. Hratchian, J.B. Cross, V.
Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R.E. Stratmann, O.
Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W. Ochterski, P.Y.
Ayala, K. Morokuma, G.A. Voth, P. Salvador, J.J. Dannenberg,
V.G. Zakrzewski, S. Dapprich, A.D. Daniels, M.C. Strain, O. Farkas,
D.K. Malick, A.D. Rabuck, K. Raghavachari, J.B. Foresman, J.V.
Ortiz, Q. Cui, A.G. Baboul, S. Clifford, J. Cioslowski, B.B. Stefanov,
G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R.L. Martin, D.J.
Fox, T. Keith, M.A. Al-Laham, C.Y. Peng, A. Nanayakkara, M.
Challacombe, P.M.W. Gill, B. Johnson, W. Chen, M.W. Wong, C.
Gonzalez, and J.A. Pople, Gaussian, Inc., Wallingford CT, 2004.
[16] A.D. Becke, J. Chem. Phys. 98 (1993) 5648.
2ꢀ
Lys to PO3 entity in comparison to proton transfer form
guanidinium or ammonium side chain group of these amino
2ꢀ
acids to PO3 unit. The calculated binding energy suggests
that energy is lower for Arg and Lys side chain group than
their a-ammonium groups thus supporting the presumable
structures of the formed supramolecular complexes.
[17] J.P. Perdew, Phys. Rev. B 33 (1986) 8822.
[18] N. Godbout, D.R. Salahub, J. Andzelm, E. Wimmer, Can. J. Chem.
70 (1992) 560.
Acknowledgements
´
´
[19] J. Gałe˛zowska, P. Mlynarz, S. Sliwinska, P. Kafarski, H. Kozłowski.
Polyhedron, accepted for publication.
This work was supported by Polish Ministry of Educa-
tion and Science (Grant 4 T09A 03325). A.O-M. gratefully
acknowledges a grant from Wrocław Center for Network-
ing and Supercomputing, and thanks the Foundation for
Polish Science for financial support.
[20] J.P. Behr, J.M. Lehn, J. Am. Chem. Soc. 98 (1976) 1743.
[21] W.L. Mock, N.-Y. Shih, J. Am. Chem. Soc. 11 (1989) 2697.
[22] E. Gaggelli, N. D’Amelio, N. Gaggelli, G. Valensin, Eur. J. Inorg.
Chem. (2000) 1699.
[23] L. Fielding, Tetrahedron 56 (2000) 6151.