Adsorbed surfactants for affinity chromatography: End-group modification of ethylene glycol polymers

Article abstract of DOI:10.1016/S0960-894X(03)00098-2

The hydroxyl end-groups of PluronicF108 {a tri-block copolymer surfactant of poly(ethylene glycol) and poly(propylene glycol) [PEG-PPG-PEG]} were converted into primary amine and quaternary ammonium equivalents for use in a new approach to affinity chromatography. The preparation of sulphonic acid end-groups was also attempted.

Full text of DOI:10.1016/S0960-894X(03)00098-2

Bioorganic & Medicinal Chemistry Letters 13 (2003) 1381–1384
Adsorbed Surfactants for Affinity Chromatography:
End-Group Modification of Ethylene Glycol Polymers
a
b,
a
c
Cemile Yanic¸ , Martin W. Bredenkamp, * Edmund P. Jacobs and Pieter Swart
aInstitute of Polymer Science, Stellenbosch University, Private Bag X1, Matieland 7602, South Africa
bDepartment of Chemistry, Stellenbosch University, Private Bag X1, Matieland 7602, South Africa
cDepartment of Biochemistry, Stellenbosch University, Private Bag X1, Matieland 7602, South Africa
Received 12 April 2002; accepted 13 December 2002
1
Abstract—The hydroxyl end-groups of Pluronic F108 {a tri-block copolymer surfactant of poly(ethylene glycol) and poly
(
propylene glycol) [PEG-PPG-PEG]} were converted into primary amine and quaternary ammonium equivalents for use in a new
approach to affinity chromatography. The preparation of sulphonic acid end-groups was also attempted.
2003 Elsevier Science Ltd. All rights reserved.
#
The non-covalent adsorption of surfactants onto the
surface of lipophilic membrane filters has been
reported as a means to reduce fouling by organic
allowed for the development of the reactions on a poly-
mer that is relatively easily characterisable.
1À6
compounds.
Modification of the end-groups of the
Yields reported for Pluronic derivatives are yields by
mass. No effort was made to remove the polymers with
unreacted or by-product end-groups. NMR investiga-
tion at each intermediate reaction step indicated pri-
marily the expected compounds, with no significant
surfactants, such as by sulphonation, may render
them even more suited as anti-foulants. Ligands can
also be attached to the end-groups of surfactants. If
the modified surfactant is adsorbed onto particle sur-
faces, the coated particles could be used for affinity
8
amount of unreacted or wrongly reacted termini. Due
7
separation.
to the difficulties involved in separating the correctly-
reacted polymers from those that did not react cor-
rectly, the presence of the undesired polymers will be
characterised and quantified in a later paper. The
In this paper, the conversion of the end-groups of Plur-
1
onic F108 (BASF; 14 600 Da) into amines and cations,
and the attempted conversion into anions is reported.
Furthermore, by attaching a ligand to the amine-termi-
1
potential use of the modified Pluronic F108 for affinity
chromatography was ascertained first and is reported
herein.
1
nated Pluronic F108 and immobilising the modified
surfactant by adsorption onto polysulphone capillary
membranes, serum albumin was extracted from serum
by affinity chromatography.
1
Amine-terminated Pluronic F108 was synthesised as
follows (Scheme 1): the hydroxyl group was activated
9
by conversion to a tosylate, that was in turn displaced
1
0
The synthetic work was initiated with model compound
studies. Di(ethylene glycol) methyl ether (DEG) simu-
lates the first two ethylene oxide nodes of PluronicF108
with the hydroxyl terminus. Hydroxyl-terminated
poly(ethylene glycol) [PEG] with a molecular mass of
with an azide, followed by reduction to the desired
1À13
1
primary amine.
The hydroxyl group(s) of DEG and PEG were con-
verted to the tosyl equivalents by sulphonylation with
50% excess p-toluenesulphonyl chloride in pyridine at
0 C for 12 h (84% yield for DEG). These reaction
mixtures and the products that were obtained were
600 Da was used as a transition model compound
1
between DEG and Pluronic F108. The use of PEG
ꢀ
ꢀ
never heated beyond 20 C since these tosylates dis-
integrate above that temperature. Satisfactory con-
version was obtained on treating a solution of 14.6
*Corresponding author. Fax: +27-21-808-3360; e-mail: mwb@sun.
ac.za
0
960-894X/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0960-894X(03)00098-2

1382
C. Yani c¸ et al. / Bioorg. Med. Chem. Lett. 13 (2003) 1381–1384
ꢀ
to minimise the loss of trimethylamine (bp: 3–4 C). No
reaction took place.
refluxed. The reflux condenser was maintained at À5 C
ꢀ
As an alternative to the direct displacement of tosylate
with trimethylamine, the following was considered more
appropriate: conversion of the tosylate to the alkyl
1
6,17
bromide
nucleophilically, with trimethylamine (Scheme 1).
and then displacement of the halide,
1
4,15
DEG tosylate was converted to DEG bromide with 2
mol equivalents of lithium bromide in dimethyl form-
amide for 3 h under an Ar atmosphere. The preparation
of PEG bromide from the tosylated equivalent was
similar to that of DEG bromide. Lithium bromide (2.64
g) was dissolved in a solution of pluronic tosylate (2.81
Scheme 1.
1
ꢀ
g Pluronic F108 in pyridine (50 mL) with 19.06 g of
tosyl chloride for 7 days. The reaction mixture was
poured onto ice and extracted with chloroform. The
chloroform extracts were washed with 6 M HCl, then
with water, and dried over K CO /Na SO , yielding
g) in dry DMF (20 mL) and maintained at 100 C under
Ar for 3 h. 3 M HCl (50 mL) was added to the mixture,
which was then extracted with chloroform, to yield (2.41
g) pluronic bromide (87%).
2
3
2
4
1
3.27 g pluronic tosylate (89% yield, 89% conver-
DEG bromide was converted to DEG-trimethyl-
ammonium bromide with 3.0 mol equivalents of tri-
methylamine. PEG- and pluronic-trimethyl-ammonium
salts were prepared likewise, using the same mass to
mass amount of trimethylamine. Pluronic bromide (1.5
g) was dissolved in 10 mL of 40% (m/m) aq trimethyl-
amine and refluxed for 4 h. The reaction mixture was
sion). Conversion was determined by the addition of
trichloroacetyl isocyanate to an NMR-sample and
integrating the terminal methylene H-signal of the
resultant carbamate thus derived from unconverted
1
1
Pluronic F108, relative to the equivalent tosylate
8
signal.
evaporated in vacuo, yielding 1.04
trimethylammonium bromide (69%).
g
pluronic-
DEG tosylate and PEG tosylate were converted to the
azides with a 20-fold excess of NaN (80% yield for
3
1
DEG tosylate). Sodium azide (7.95 g) was added to a
solution of pluronic tosylate (1.52 g) in dry DMF (25
The strategy for the transformation of Pluronic F108
to pluronic sulphonic acid was the conversion of
ꢀ
18,19
mL) under Ar atmosphere at 65 C and allowed to react
pluronic tosylate to pluronic-thiol,
1
followed by
oxidation with oxone to the corresponding sulphonic
for 18 h. The reaction mixture was then poured onto ice
water, extracted with chloroform, the extract washed
with 2.5 M HCl and distilled water, sequentially, and
2
0,21
acid.
dried over MgSO to yield 1.31 g pluronic azide (87%).
4
Focusing on the model compounds first, DEG tosylate
and PEG tosylate were converted to the corresponding
thiols with 2 mol equivalents of aqueous Na CS
DEG azide and PEG azide were reduced to the amine
using 8 mol equivalents of LiAlH per azide in ether, at
2
3
1
8
ꢀ
(33%). per tosylate at 65 C for 5 h. The thiolation
reaction is postulated to proceed via the sodium mono-
alkyl trithiocarbonate which, upon acidification,
decomposes spontaneously to liberate a thiol and car-
4
reflux temperature for 6 h. For pluronic azide, 0.50 g of
LiAlH was added to a solution of pluronic azide (0.40
4
g) in THF (15 mL). The reaction mixture was refluxed
for 6 h, then slowly quenched with water. The mixture
was made basic (pH 10) with 25 mL of 5 M sodium
hydroxide and then continuously extracted with
chloroform for 12 h, furnishing 0.30 g (76%) amino-
pluronic.
2
2
bon disulphide. Besides the formation of the desired
DEG-thiol, significant amounts of diDEG disulphide
and diDEG sulphide also formed. Oxidation of diDEG
disulphide should readily oxidise to the desired sul-
2
3,24
phonic acid,
however, the diDEG sulphide will most
likely yield the sulphone. A rationale for the formation
of the sulphide is nucleophilic attack of the tosylate by
the DEG sulphide ion instead of trithiocarbonate. The
sulphide is likely to originate from the dissociation of
the sodium alkyltrithiocarbonate into the alkyl sulphide
Two synthetic routes were considered to convert the
1
hydroxyl end-groups of Pluronic F108 into trimethyl-
ammonium end-groups: alkylation of amino-pluronic
by exhaustive methylation; or use of trimethylamine as
a nucleophile to substitute the tosylate of pluronic
tosylate. The one-step conversion of the tosylate to
quaternary ammonium equivalents using trimethyl-
and CS in a reversible process. With this rationale in
2
mind the reaction of DEG tosylate was repeated in the
presence of CS (75% yield for DEG sulphide). This
2
1
4,15
amine
seemed more expedient.
modified procedure circumvented the formation of the
sulphide. The preparation of pluronic thiol follows:
pluronic tosylate (2.00 g) was added to a 33% solution
The synthesis of DEG-trimethylammonium salt was
first attempted using DEG tosylate as precursor. A
mixture of DEG tosylate, trimethylammonium chloride
and sodium hydrogencarbonate in dry methanol was
of sodium trithiocarbonate and 41% CS in water (9
2
mL) under Ar. The reaction mixture was stirred and
ꢀ
gradually heated to 60 C. After 5 h the mixture was

C. Yani c¸ et al. / Bioorg. Med. Chem. Lett. 13 (2003) 1381–1384
1383
acidified (concd H SO ) to between pH 1 and 2, extrac-
2
ted with chloroform and dried (MgSO ), yielding 1.75 g
4
pluronic-thiol (89%).
d 60.7, indicating coupling of the PEG-block to Ciba-
cron Blue 3GA. The proton NMR spectrum did not
prove of any value in indicating the process of coupling
because of the signals of the bulk polymer at the critical
chemical shifts.
4
DEG- and PEG-thiols were successfully oxidised to the
corresponding sulphonic acids with a stoichiometric
1
amount
(
(oxone–mercaptan,
3:1)
of
oxone
The utility of Cibacron blue 3GA coupled amino-
pluronic (CBAP) adsorbed onto polysulphone fibers as
a non-covalent ligand for the affinity separation of
albumin was then investigated. A single polysulphone
capillary membrane (PSf matrix) was first used to test
the isolation of commercial bovine serum albumin.
Thereafter, a multi-fibre system was used to enable the
characterisation and quantification of the isolated albu-
2KHSO KHSO K SO ) for 1 h at ambient tempera-
5 4 2 4
ture. The same amount of oxone to substrate mass
was used for pluronic-thiol as was used for PEG-thiol
1
(
i.e., 22-fold excess). Characterisation of the product by
13
1
H and C NMR spectroscopy indicate very little, if
any, of the sulphonic acid derivative. The major deriva-
1
tive was Pluronic F108, which had not been present
1
29
before oxidation. Oxone had thus cleaved off the sul-
phur atom.
min. The PSf membranes in the multi-fibre module
were coated in the following manner. A 0.1% aqueous
solution of CBAP was circulated through the multi-fibre
ꢀ
The preparation of DEG-sulphonic acid was carried out
as follows: DEG-thiol (0.51 g) was dissolved in a 25%
PSf matrix for 25 h at room temperature, using a peri-
staltic pump. The fibres were then rinsed with distilled-
deionised water until the fractions emerging from the
fibre showed no absorbency at 610 nm.
(
m/v) aqueous solution of oxone1 (14 mL) and stirred
for 1 h. Concd HCl (14 mL) was added and the mixture
stirred for a further 15 min. The reaction mixture was
then extracted with ether and the ethereal extract dried
(MgSO ), and the organic solvents removed by eva-
4
poration, yielding 0.23 g DEG-sulphonic acid (33%).
Two buffer solutions were used to process the protein-
loaded, uncoated (control) and coated PSf matrices
(CBAP): Buffer A (washing buffer) 0.05 M Tris/HCl, pH
8.0, containing 0.05 M NaCl, and Buffer B (elution buf-
fer) 0.05 M Tris/HCl, pH 8.0, containing 0.2 M NaSCN.
Isolation of albumin from sheep serum was considered
1
to investigate the suitability of aminated Pluronic F108
as a means to immobilise ligands for affinity chromato-
graphy. Cibacron Blue 3GA is used as a ligand for the
Sheep serum (5 mL) containing 5 mg/mL of protein was
introduced into the fibres. buffer A was passed through
the system for 6 h under constant UV detection at 280
nm. Sheep serum proteins that were retained by the
system were eluted thereafter by passing buffer B
through the fibres for 4 h. Albumin retention was mon-
2
5À27
isolation of albumin.
2-Methoxyethylamine was
successfully used as a model amine for developing the
coupling reaction of Cibacron Blue 3GA to an amino
end-group (Scheme 2). Thereafter an aqueous solution
of 120 mg Cibacron Blue 3GA and 318 mg Na CO was
3
0,31
itored qualitatively by the Kaiser ninhydrin test
M
and
Frac-
2
3
32,33
added to an aqueous solution of 570 mg amino-pluro-
nic. The reaction mixture was maintained at 45 C for 3
quantitatively by the Pierce Protein Assay.
ꢀ
Cibacron Blue-amino-pluronic (CBAP) and
tions which contained sheep serum albumin were fur-
ther characterised by Sodium Dodecyl Sulphate
2
8
days.
3
4À36
unreacted dye were separated by eluting through a
Sephadex G-50-40 column. The fractions were analysed
by UV spectroscopy at 610 nm (absorption maximum
of Cibacron Blue). Product fractions were freeze dried,
Polyacrylamide Gel Electrophoresis
and Electro-
spray Mass Spectrometry. A 32 mL fraction containing
35 mg/mL of highly purified sheep serum albumin, as
determined by ESMS, was thus obtained.
1
3
1
and investigated by C and H NMR spectroscopy.
Even though the efficiency of conversion of amino-
pluronic to CBAP could not be ascertained by NMR
spectroscopy, because of the coincidence of the terminal
The potential use of non-covalent anchoring of ligands
for affinity separation has been proved. Improved pro-
cedures to prepare the ligand carrier, such as a
3
7À39
40
CH -signal of amino-pluronic with the signal of the
2
triethylamine-
pluronic-tosylate instead of pyridine, are currently being
or alkoxide-based preparation of
deuterated solvent, the chemical shift of the terminal
CH -group of the PEG-block of CBAP was observed at
4
1,42
investigated. The use of propane sultone
as a means
2
1
of sulphonylating the termini of Pluronic F108 is also
under investigation. The economic viability of this pro-
cedure for the isolation of albumin from abattoir efflu-
ent is now also being studied. Other applications of
affinity separation by non-covalent adsorption onto
membranes are also being explored.
Acknowledgements
The Water Research Commission and the National
Research Foundation of South Africa and the Uni-
versity of Stellenbosch are thanked for funding this
research.
Scheme 2.

1384
C. Yani c¸ et al. / Bioorg. Med. Chem. Lett. 13 (2003) 1381–1384
References and Notes
m, M.; Jarvinen, P. J. Membrane Sci. 1991, 60, 275.
. Fane, A. G.; Fell, C. J. D.; Kim, K. J. Desalination 1985,
3, 37.
. Kim, K. J.; Fane, A. G.; Fell, C. J. D. Desalination 1988,
0, 229.
. Kim, K. J.; Fane, A. G.; Fell, C. J. D. J. Membrane Sci.
989, 43, 187.
. Brink, L. E. S.; Romijn, D. J. Desalination 1990, 78, 209.
. Kim, K.-J.; Fane, A. G. J. Membrane Sci. 1995, 99, 149.
. Ho, C.-H.; Limberis, L.; Caldwell, K. D.; Stewart, R. J.
Langmuir 1998, 14, 3889.
. Yanic, C.; Bredenkamp, M. W.; Jacobs, E. P.; Spies,
H. S. C.; Swart, P. J. Appl. Pol. Sci. 2000, 78, 109.
. Fieser, L.; Fieser, M. In Reagents for Organic Synthesis,
John Wiley and Sons: New York, 1967; Vol. 1, p 1180.
23. Pandiarajan, K. In Synthetic Reagents; John Wiley and
Sons: Toronto, 1985; Vol. 6, p 111.
4. Pizey, S. S. In Synthetic Reagents; John Wiley and Sons:
Toronto, 1974; Vol. 1, p 27.
5. Travis, J.; Bowen, J.; Tewksbury, D.; Johnson, D.; Pan-
nell, R. Biochem. J. 1976, 157, 301.
2
1
2
5
3
7
4
1
5
6
7
. Nystro
¨
¨
2
2
3
2
6. Heyns, W.; De Moor, P. Biochim. Biophys. Acta 1974,
58, 1.
7. Bohme, H.-J.; Kopperschlager, G.; Schulz, J.; Hofmann,
¨ ¨
E. J. Chromatogr. 1972, 69, 209.
8. Naval, J.; Calvo, M.; Lampreave, F.; Pin
Ed. 1983, 11, 5.
9. Yanic C. Dissertation: Polymer End-Group Modifi-
2
˜
eiro, A. Biochem.
2
¸ ,
cation for Affinity Separation; Institute of Polymer
Science; University of Stellenbosch, South Africa,
8
¸
1
3
999.
0. Kaiser, E.; Colescott, R. L.; Bessinger, C. D.; Cook, P. I.
9
Anal. Biochem. 1970, 34, 595.
1
1
1
0. Boger, J.; Corcoran, R. J.; Lehn, J.-M. Helv. Chim. Acta
978, 61, 2190.
1. Martin, V. V.; Lex, L.; Keana, J. F. W. Org. Prep. Pro-
3
3
1. Troll, W.; Cannan, R. K. J. Biol. Chem. 1953, 200, 803.
2. Smith, P. K.; Krohn, R. I.; Hermanson, G. T.; Mallia,
A. K.; Gartner, F. H.; Provenzano, M. D.; Fujimoto, E. K.;
Goeke, N. M; Olson, B. J.; Klenk, D. C. Anal. Biochem. 1985,
ceed. Int. 1995, 27, 117.
1
5
1
1
3
1
1
1
2. Rosser, R. M.; Faulkner, D. J. J. Org. Chem. 1984, 49,
157.
3. Boyer, J. H. J. Am. Chem. Soc. 1951, 73, 5865.
4. Broxton, T. J.; Chung, R. P.-T. J. Org. Chem. 1986, 51,
112.
5. Hendrix, W. T.; Von Rosenberg, J. L. J. Am. Chem. Soc.
976, 98, 4850.
´
6. Stephenson, B.; Solladie, G.; Mosher, H. S. J. Am. Chem.
1
3
1
3
3
50, 76.
3. Redinbaugh, M. G.; Turley, R. B. Anal. Biochem. 1986,
53, 267.
4. Bury, A. F. J. Chromatogr. 1981, 213, 491.
5. Matheka, H. D.; Enzmann, P.-J.; Bachrach, H. L.; Migl,
B. Anal. Biochem. 1977, 81, 9.
36. Laemmli, U. K. Nature 1970, 227, 680.
37. Harris, J. M.; Struck, E. C.; Case, M. G.; Paley, M. S.;
Soc. 1972, 94, 4184.
Yalpani, M.; Van Alstine, J. M.; Brooks, D. E. J. Polym. Sci.,
Polym. Chem. Ed. 1984, 22, 341.
1
1
1
1
7. Stork, G.; Grieco, P. A.; Gregson, M. Tetrahedron Lett.
969, 1393.
8. Martin, D. J.; Greco, C. C. J. Org. Chem. 1968, 33, 1275.
9. Fieser, M.; Fieser, L. In Reagents for Organic Synthesis;
3
3
4
8. Simionescu, C. I.; Rabia, I. Polym. Bull. 1983, 10, 311.
9. Gao, B.; Kops, J. Polym. Bull. 1995, 34, 279.
0. Harris, J. M.; Hundley, N. H.; Shannon, T. G.; Struck,
John Wiley and Sons: Toronto, 1969; Vol. 2, p 5389.
E. C. J. Org. Chem. 1982, 47, 4789.
1. Linfield, W. M.; Abend, P. G.; Davis, G. A. J. Am. Oil.
Chem. Soc. 1963, 40, 114.
2. Parris, N.; Weil, J. K.; Linfield, W. M. J. Am. Oil. Chem.
Soc. 1972, 49, 649.
2
1
2
0. Kennedy, R. J.; Stock, A. M. J. Org. Chem. 1960, 25,
901.
1. Fieser, L. F.; Fieser, M. In Reagents for Organic Synthesis;
4
4
John Wiley and Sons: Sydney, 1967; Vol. 1, p 952.
2. Hantzsch, A.; Bucerius, W. Ber. 1926, 59B, 793.
2