Increasing permeability of phospholipid bilayer membranes to alanine with synthetic α-aminophosphonate carriers

Article abstract of DOI:10.1016/j.bmcl.2008.02.081

A series of aminophosphonates was synthesized, and their ability to carry alanine, a model hydrophilic molecule, across phospholipid bilayer membranes was evaluated. Aminophosphonates facilitate the membrane transport at moderate rates, which make them a suitable platform for the design of carriers for continuous drug release devices.

Full text of DOI:10.1016/j.bmcl.2008.02.081

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 2320–2323
Increasing permeability of phospholipid bilayer membranes
to alanine with synthetic a-aminophosphonate carriers
Delia C. Danila,a Xinyan Wang,a Heather Hubble,a
Igor S. Antipinb and Eugene Pinkhassika,*
aDepartment of Chemistry, The University of Memphis, Memphis, TN 38152, USA
bDepartment of Organic Chemistry, Kazan State University, Kremlevskaya 18, Kazan 420008, Russia
Received 15 January 2008; revised 27 February 2008; accepted 29 February 2008
Available online 6 March 2008
Abstract—A series of aminophosphonates was synthesized, and their ability to carry alanine, a model hydrophilic molecule, across
phospholipid bilayer membranes was evaluated. Aminophosphonates facilitate the membrane transport at moderate rates, which
make them a suitable platform for the design of carriers for continuous drug release devices.
ꢀ 2008 Elsevier Ltd. All rights reserved.
One of the most important challenges in designing lipo-
some based drug release systems is the precise control
of transport kinetics across bilayer boundaries.1,2 Encap-
sulated drugs should remain within liposomes during
storage, and release should only begin after the adminis-
tration. In an ideal system, a triggering event will initiate
continuous release with well-controlled rate. Synthetic
carriers of polar organic molecules through lipid bilayers
can potentially provide an elegant way to achieve this
goal. Insertion of carriers into bilayers of drug-contain-
ing liposomes immediately prior to administration would
trigger the release. Transport kinetics can be regulated by
varying the concentration of carriers in the bilayer.
concept’.15 Sunamoto et al. reported on the transport
of phenylalanine using a photoresponsive carrier.16
Traditionally, high selectivity and fast transport rates
were viewed as desirable albeit challenging. These char-
acteristics may not be necessary in the design of carriers
suitable for continuous drug release. This application re-
quires slow release of a single component over a long
period of time, measured between hours and weeks,
and neither selectivity nor fast transport is critical. In
fact, slow carriers are likely to have an advantage of
inherently low cytotoxicity, an important safety consid-
eration in the case a carrier molecule separates from a
liposome and inserts into a cellular membrane. Small
size of carriers is likely to be beneficial for rapid incorpo-
ration into the liposomal bilayer.
Previously, a relatively small number of studies focused
on synthetic carriers for the transport of organic mole-
cules across bilayer lipid membranes compared to well
explored areas of artificial ion channels3,4 and transport
in supported liquid membranes.5–8 In the recent years,
however, the field has been gaining considerable interest.
Several papers described synthetic carriers for carbohy-
drates,9,10 nucleosides,11 small peptides,12 and other
molecules.13,14 Among the most recent beautiful exam-
ples is the transport of oligonucleotides guided by um-
brella carriers in the innovative ‘needle and thread
Considering the above, we decided to synthesize and
evaluate a series of a-aminophosphonates as carriers
of hydrophilic organic molecules across phospholipid
membranes. We used alanine, an amino acid, as a model
hydrophilic bifunctional molecule for transport studies.
In our view, aminophosphonates are excellent carrier
platforms. They possess two binding sites, a hydrogen
bond donor and a hydrogen bond acceptor suitable
for two-point binding of hydrophilic molecules creating
in the case of amino acids a complex with hydrophobic
exterior.17,18 They are readily prepared in a one-pot syn-
thesis by the Kabachnik–Fields reaction from a primary
amine, a phosphite, and a carbonyl compound.19,20
Their hydrophilic–hydrophobic balance can be easily
Keywords: Carriers; Liposomes; Drug delivery; Amino acids; Bilayers;
Aminophosphonates.
*
Corresponding author. Tel.: +1 901 678 4430; fax: +1 901 678
3447; e-mail: epnkhssk@memphis.edu
0960-894X/$ - see front matter ꢀ 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.02.081

D. C. Danila et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2320–2323
2321
1
varied by using different substituents in the starting
materials. In previous studies, aminophosphonates did
not exhibit cytotoxicity or mutagenic properties.21,22
Previously reported non-toxic concentrations as high
as 15 mM support the feasibility of successful in vivo
applications.21,22 Base aminophosphonate platform can
be further modified with substituents that provide steric
interactions or additional binding of a substrate mole-
cule. This can be utilized to adjust the selectivity and
strength of binding of a wide range of substrates.
Three-point interactions were shown effective in chiral
recognition.23
in H NMR spectra that indicated the formation of the
condensation product was the long-range coupling be-
tween phosphorus and hydrogens contributed by the
carbonyl compound.
Transport properties of aminophosphonate carriers were
evaluated by liposome efflux experiments coupled with
enzymatic assays.25 In these experiments, unilamellar
liposomes loaded with alanine (300 mM) were prepared
from dipalmitoyl phosphatidyl choline (DPPC) and cho-
lesterol (3:1 molar ratio) by a standard literature proto-
col,26 and unentrapped alanine was separated by size
exclusion chromatography. The batch of alanine-loaded
liposomes was separated into several samples, and iden-
tical amounts of carriers were introduced to bilayers with
the help of a solvent vector, DMSO, so that each sample
contained a different carrier with the same concentration.
The rate of alanine efflux was determined from the con-
version of alanine to pyruvate, catalyzed by Glutamic-
Pyruvic Transaminase, and further to lactate, catalyzed
by lactate dehydrogenase, which is accompanied by the
oxidation of nicotinamide cofactor NADH to NAD+.
Disappearance of NADH was monitored by UV
spectroscopy.27
The synthesis of the a-aminophosphonates was done by
the Kabachnik–Fields reaction, a condensation between
an amine, a carbonyl compound, and a phosphite
(Scheme 1).24
By using aniline and benzylamine, acetone and cyclo-
hexanone as carbonyl compounds and dimethylphos-
phite, dibutylphosphite, or bis-2-ethyhexylphosphite as
phosphites, we synthesized a series of seven aminophos-
phonates. We selected substituents for achieving varia-
tions in size and hydrophobic–hydrophilic balance of
products. Thus compound 3 is almost twice the size of
compound 1. Compounds 2 and 4 are similar in size
but are different in structure. Products were isolated
by column chromatography on silica gel using hexane/
ethyl acetate (5:1) as eluent. The most prominent feature
The carrier mechanism of transport was supported by
the lack of evidence of aminophosphonate induced
membrane lysis. In this experiment, we prepared lipo-
somes containing self-quenching fluorescent dye calcein.
Calcein loses 98% of its fluorescence at concentrations
above 100 mM.28 In the lysis experiment, liposomes with
calcein were mixed with aminophosphonate solution in
DMSO and the fluorescence was monitored for 1 h,
the amount of time comparable with the amino acid ef-
flux experiments. No detectable change in emission was
observed. In the end of this period, liposomes were trea-
ted with Triton X-100 solution that caused lysis26 and an
immediate huge increase of emission, indicative of dye
dilution below self-quenching concentration due to re-
lease from liposomes. We conclude that the aminophos-
phonates did not cause bilayer membranes to become
more permeable, as this would have been accompanied
by a continuous increase of emission.
O
O
P
O
O
R4
R4
R1 NH2
+
+
R3
R2
H
H
O
O
O
R4
R4
N
R2
P
R3
R1
1
2
3
4
5
6
7
R1 = C6H5;
R1 = C6H5;
R1 = C6H5;
R1 = C6H5;
R1 = C6H5;
R2 = R3 = CH3;
R2 = R3 = CH3;
R2 = R3 = CH3;
R4 = CH3
R4 = C4H9
Data in Table 1 summarize average flow rates for ala-
nine using different aminophosphonates. All carriers in-
crease permeability of phospholipid bilayers to alanine.
In control experiments, we observed no measurable
enhancement of alanine transport when adding amines
or phosphites to the bilayers. Flux was calculated in
nmol/h · m2 using volume and surface area of 100 nm
liposomes, which was the average size confirmed by
transmission electron microscopy. We chose these units
of measurement to enable comparison with other stud-
ies. Thus, we calculated the photoresponsive carrier-as-
R4 = CH2CH(CH2CH3)C4H9
R2 + R3 = (CH2)5; R4 = CH3
R2 + R3 = (CH2)5; R4 = C4H9
R1 = CH2C6H5; R2 = R3 = CH3;
R4 = CH3
R1 = CH2C6H5; R2 + R3 = (CH2)5; R4 = C4H9
Scheme 1. Synthesis of a-aminophosphonates 1–7 from an amine
(aniline or benzylamine), a carbonyl compound (acetone or cyclohex-
anone), and a phosphite (dimethyl phosphite, dibutyl phosphite, or
bis-(2-ethylhexyl)-phosphite).
Table 1. Transport of alanine across lipid bilayers, average flux from three experiments
Carrier
Nonea
1a
2a
3a
4a
5a
6a
7a
Flux, nmol/h · m2
0.7 ( 0.4)b
4.4 ( 1.6)
3.1 ( 1.0)
4.3 ( 2.0)
7.2 ( 1.8)
4.4 ( 0.5)
7.8 ( 2.3)
4.4 ( 1.8)
a Values are means of three experiments, standard deviation is given in parentheses.
b Measured by a liposome efflux experiment coupled with alanine dehydrogenase assay.

2322
D. C. Danila et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2320–2323
sisted transport of phenylalanine to be 14 nmol/h · m2
from the data reported by Sunamoto et al., which is
comparable to the values we observed.16 Since all carri-
ers were always run together using the same batch of lip-
osomes, relative transport rates offer a meaningful
comparison. Compounds 4 and 6 exhibited fastest trans-
port rates in the series, and compound 3 was the slowest
carrier.
6. Scrimin, P.; Tonelatto, U.; Zanta, N. Tetrahedron Lett.
1988, 29, 4967.
7. Tsukube, H.; Shinoda, S.; Uenishi, J.; Schiode, M.;
Yonemitsu, O. Chem. Lett. 1996, 11, 969.
8. Mohler, L. K.; Czarnik, A. W. J. Am. Chem. Soc. 1993,
115, 7037.
9. Smith, B. D. Supramol. Chem. 1996, 7, 55.
10. Lyutikova, I. V.; Pletnev, I. V.; Matveeva, I. G.;
Torocheshnikova, I. I. Russ. Chem. Bull. 1998, 47, 177.
11. Furuta, H.; Furuta, K.; Sessler, J. L. J. Am. Chem. Soc.
1991, 113, 4706.
12. Janout, V.; DiGiorgio, C.; Regen, S. L. J. Am. Chem. Soc.
2000, 122, 2671.
13. Smith, B. D.; Gardiner, S. J. Adv. Supramol. Chem. 1999,
5, 157.
14. Janout, V.; Jing, B.; Staina, I. V.; Regen, S. L. J. Am.
Chem. Soc. 2003, 125, 4436.
15. Janout, V.; Regen, S. L. J. Am. Chem. Soc. 2005, 127, 22.
16. Sunamoto, J.; Iwamoto, K.; Mohri, Y.; Kominato, T. J.
Am. Chem. Soc. 1982, 104, 5502.
17. Antipin, I. S.; Stoikov, I. I.; Garifzyanov, A. R.; Kono-
valov, A. I. Russ. J. Gen. Chem. 1996, 66, 391.
18. Antipin, I. S.; Stoikov, I.; Pinkhassik, E.; Fitseva, N.;
Stibor, I.; Konovalov, A. Tetrahedron Lett. 1997, 38, 5865.
19. Kabachnik, M. I.; Medved, T. Ya. Dokl. Akad. Sci. USSR
1952, 83, 689.
20. Fields, E. K. J. Am. Chem. Soc. 1952, 74, 1528.
21. Pietri, S.; Miollan, M.; Martel, S.; Le Moigne, F.; Blaive,
B.; Culcasi, M. J. Biol. Chem. 2000, 275, 19505.
22. Martel, S.; Clement, J.-L.; Muller, A.; Culcasi, M.; Pietri,
S. Bioorg. Med. Chem. 2002, 10, 1451.
We observed no direct relationship between hydropho-
bicity and efficiency of aminophosphonate carrier. It is
likely that optimum transport efficiency is the function
of availability of the carrier at the bilayer boundary,
hydrophobicity of the complex, and mobility of both
complex and free carrier. We observed significant differ-
ences between carrier 2 and carriers 4 and 6 as well as
carrier 5 and carriers 4 and 6. This can open an oppor-
tunity for finding relationships between carrier structure
and transport efficiency.
Based on transport rates measured in this study, 50% re-
lease of alanine, encapsulated in a 100-nm liposome with
the concentration of 300 mM, over a period of 3 weeks
would require 4–5 mol % of aminophosphonate carriers
relative to membrane phospholipids. Such concentra-
tions are attainable, and can be scaled down for even
longer release. Further tuning of transport efficiency
through additional binding by side-arm substituents
may offer control of mass transfer kinetics in a wider
range. Rapid progress in long-circulating liposomes
makes artificial transporters an attractive component
of the drug delivery systems.29
23. Davankov, V. A. Chirality 1997, 9, 99.
24. Reagents and conditions: 1, 2, 3, 6: amine (0.01 mol),
phosphite (0.01 mol), acetone (5 ml), 56 ꢁC, overnight; 4,
5, 7: amine (0.01 mol), cyclohexanone (0.01 mol), phos-
phite (0.01 mol), toluene (3 ml), 80 ꢁC, overnight. Solvent
was evaporated, and the mixture was separated by column
chromatography on silica gel with hexane/ethyl acetate
(5:1) eluent. Yields and spectroscopic data for the a-
aminophosphonates are as follows (1H NMR, CDCl3,
270 MHz):
In summary, we synthesized a series of a-aminophos-
phonates and evaluated their transport properties. Ami-
nophosphonates are capable of transporting alanine, a
model hydrophilic compound, across bilayer lipid mem-
branes at moderate rates. Due to ease of structural vari-
ations, a-aminophosphonates can be attractive
platforms for building carriers for facilitating transmem-
brane transport and regulating permeability of long-cir-
culating continuous release drug delivery devices.
Compound 1: Yield 75%, 1.48 ppm (d, 6H, J = 15.6 Hz),
3.72 ppm (d, 6H, J = 10.4 Hz), 6.82 ppm (t, 1H,
J = 7.4 Hz), 7.02 ppm (d, 2H, J = 1.2 Hz), 7.16 ppm (t,
2H, J = 20.0 Hz).
Compound 2: Yield 68%, 0.9 ppm (t, 6H, J = 32.4 Hz),
1.36 ppm (m, 4H, J = 51.3 Hz), 1.46 ppm (d, 6H,
J = 15.6 Hz), 1.56 ppm (m, 4H, J = 21.6 Hz), 4.09 ppm
(m, 4H, J = 21.0 Hz), 6.82 ppm (t, 1H, J = 7.4 Hz),
7.02 ppm (d, 2H, J = 1.2 Hz), 7.16 ppm (t, 2H,
J = 20.0 Hz).
Acknowledgments
We thank Chelsey Hunter for help with synthesis of
compounds 2 and 5. This work was supported in part
by the NSF CAREER award (CHE-0349315) and Uni-
versity of Memphis New Faculty Research Initiation
Award to E.P.
Compound 3: Yield 50%, 0.86 ppm (t, 12H, J = 54.0 Hz),
1.28 ppm (m, 10H, J = 27.0 Hz), 1.33 ppm (m, 4H,
J = 13.5 Hz), 1.48 ppm (d, 6H, J = 15.6 Hz), 1.55 ppm
(m, 2H, J = 35.1 Hz), 3.93 ppm (m, 4H, J = 21.0 Hz),
6.82 ppm (t, 1H, J = 7.4 Hz), 7.02 ppm (d, 2H,
J = 1.2 Hz), 7.16 ppm (t, 2H, J = 20.0 Hz).
Compound 4: Yield 61%, 1.26 ppm (m, 2H, J = 27 Hz),
1.51 ppm (m, 4H, J = 14 Hz), 1.80 ppm (m, 2H,
J = 40.5 Hz), 2.2 ppm (m, 2H, J = 19 Hz), 3.68 ppm (d,
6H, J = 10.4 Hz), 6.82 ppm (t, 1H, J = 7.4 Hz), 7.02 ppm
(d, 2H, J = 1.2 Hz), 7.16 ppm (t, 2H, J = 20.0 Hz).
Compound 5: Yield 39%, 0.86 ppm (t, 6H, J = 13.5 Hz),
1.30 ppm (m, 6H, J = 35.1 Hz), 1.51 ppm (m, 4H,
J = 27 Hz), 1.8 ppm (m, 2H, J = 35.1 Hz), 2.2 ppm (m,
2H, J = 21.6 Hz), 4.09 ppm (m, 4H, J = 21.0 Hz),
6.79 ppm (t, 1H, J = 7.4 Hz), 7.02 ppm (d, 2H,
J = 1.2 Hz), 7.14 ppm (t, 2H,J = 20.0 Hz).
References and notes
1. Chonn, A.; Cullis, P. R. Curr. Opin. Biotech. 1995, 6, 698.
2. Lian, T.; Ho, R. J. Y. J. Pharm. Sci. 2001, 90, 667.
3. Stoikov, I. I.; Antipin, I. S.; Konovalov, A. I. Russ. Chem.
Rev. 2003, 72, 1055.
4. Kobuke, Y. Adv. Supramol. Chem. 1997, 4, 163.
5. Brice, L. J.; Pirkle, W. H. Chiral Separations: Applications
and Technology; American Chemical Society: Washington,
1996.
Compound 6: Yield 59%, 1.44 ppm (d, 6H, J = 15.6 Hz),

D. C. Danila et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2320–2323
2323
3.83 ppm (d, 6H, J = 10.4 Hz), 3.96 ppm (s, 2H), 6.79 ppm
(t, 1H, J = 7.4 Hz), 7.02 ppm (d, 2H, J = 1.2 Hz),
7.14 ppm (t, 2H, J = 20.0 Hz).
Compound 7: Yield 51%, 1.24 ppm (m, 2H, J = 14.3 Hz),
1.51 ppm (m, 4H, J = 14.2 Hz), 1.73 ppm (m, 2H,
J = 27.2 Hz), 1.87 ppm (m, 2H, J = 16.2 Hz), 3.78 ppm
(d, 6H, J = 10.4 Hz), 3.9 ppm (s, 2H) 6.79 ppm (t, 1H,
J = 7.4 Hz), 7.02 ppm (d, 2H, J = 1.2 Hz), 7.14 ppm (t,
2H, J = 20.0 Hz).
from 0.03 mmol of dipalmitoyl phosphatidyl choline and
0.01 mmol of cholesterol in 1 ml of alanine solution
(300 mM) in tris buffer (50 mM, pH 7.4) and separated
from unentrapped alanine on Sephadex-25 column,
100 lL of solution containing enzymes GPT and LDH
and NADH, 795 lL of buffer, and 5 lL of a-amin-
ophosphonate solution in DMSO (100 mM). NADH
concentration change was calculated from absorbance at
340 nm. Background absorbance change in the control
sample containing DMSO but no carrier was subtracted
from all measurements to single out the effect of amin-
ophosphonate carriers on alanine transport. The enzy-
matic assay was calibrated with aliquots of alanine of
known concentrations. In all cases, complete conversion
of alanine was observed.
25. Bergmeyer, H. U. Methods of Enzymatic Analysis, 3rd ed.;
John Wiley and Sons, 1986.
26. Torchilin, V.; Weissig, V. Liposomes, 2nd ed.; Oxford
Press, 2003.
27. Kinetic data were collected on Agilent 8453 UV–vis
spectrophotometer equipped with 8-cell cuvette holder
maintained at 37 ꢁC. In a typical experiment, samples were
prepared by mixing 100 lL of liposome solution prepared
28. Allen, T. M. Liposome Technol. 1984, 3, 177.
29. Torchilin, V. P. Nat. Rev. Drug Disc. 2005, 4, 145.