Improved biotransformations on charged PEGA supports

Article abstract of DOI:10.1039/b301680c

PEGA supports functionalised with permanent charges show superior swelling properties in aqueous media when compared to neutral PEGA; a novel positively charged PEGA resin significantly improves penicillin G amidase (PGA) catalysed biotransformation on solid support, by favouring accessibility of the negatively charged enzyme.

Full text of DOI:10.1039/b301680c

Improved biotransformations on charged PEGA supports
a
a
a
b
b
Alessandra Basso, Luigi De Martin, Lucia Gardossi,* Graham Margetts, Ian Brazendale, Annie
c
c
c
Y. Bosma, Rein V. Ulijn* and Sabine L. Flitsch
Dipartimento di Scienze Farmaceutiche, Università degli Studi, Piazzale Europa 1, 34127, Trieste, (Italy).
E-mail: gardossi@units.it; Fax: +39/04052572; Tel: +39/0405583110
Polymer Laboratories Ltd., Essex Road, Church Stretton, Shropshire, UK SY6 6AX
School of Chemistry, The Edinburgh Centre for Protein Technology, The University of Edinburgh, King’s
Buildings, West Mains Road, Edinburgh EH9 3JJ, UK. E-mail: rein.ulijn@ed.ac.uk;
Tel: +44/01316504792
a
b
c
Received (in Cambridge, UK) 13th February 2003, Accepted 7th April 2003
First published as an Advance Article on the web 7th May 2003
PEGA supports functionalised with permanent charges
show superior swelling properties in aqueous media when
compared to neutral PEGA; a novel positively charged
PEGA resin significantly improves penicillin G amidase
A significant improvement in swelling was observed for
PEGA (from 18 to 35 mL g dry) and PEGA (from 13 to 32
mL g21 dry) when the buffer concentration was lowered from 0.1
to 0.001 M (Fig. 1). This observation reflects the increased ionic
interactions between chains at low ionic strength. Both charged
polymers have the same amount of acrylamide monomers
substituted with charged residues, thus explaining the very
similar swelling behaviour. As expected, no ionic strength
dependence on polymer swelling was observed for neutral
+
21
2
(
PGA) catalysed biotransformation on solid support, by
favouring accessibility of the negatively charged enzyme.
Applications of enzyme catalysis on substrates that are
immobilised onto solid supports are on the increase. Examples
1
are chemo-enzymatic synthesis of compound libraries, ‘on-
PEGA1900.
bead’ screening for enzyme substrates or inhibitors in combina-
The next step was then to assess whether improved swelling
had any effect on the efficiency of the biotransformation. PGA
was chosen as a model enzyme because of its applicability on
enzyme cleavable linkers and in enzymatic deprotection
strategies.3 As the solid phase substrate for PGA, a N-
phenylacetylated -Phe (1) was linked to the resin via a Wang-
type linker. This linker allows for the separate analysis of both
a,b
torial libraries,2 enzyme cleavable linkers and protecting
3
groups for release of polymer bound libraries, applications in
4
5
micro-array technologies and enzymatic optical resolution.
The ability of the enzyme to catalyse reactions within the solid
phase resin beads strongly depends on the dimension of the
protein compared to the pore size of the resin. This in turn is
determined by the swelling properties of the resin.2
L
,6
PEGA resins are hydrophilic solid supports that have been
7
designed for batch and continuous flow peptide synthesis. It
has been reported that PEGA1900 (the subscript refers to the
PEG molecular weight), is freely permeable to enzymes with
molecular weight up to 35 KDa and compatible with their
activity. Thus, PEGA1900 was found to be ideally suitable for
the screening of peptide libraries, affinity purification and on-
resin enzyme assays.2 For the use of large enzymes such as
penicillin G amidase (PGA, MW = 88 KDa) in solid phase
biocatalysis, the access of the protein to the inside of solid
polymers is the limiting step. This translates into the low yields
,5
3
reported in the literature on PEGA1500. Relatively better results
have been obtained either by using PEGA resins with longer
PEG chains (PEGA1500) that allow access of large proteases2 or
c
by using soluble resins for PGA catalysed hydrolysis of enzyme
Scheme 1 a) PEGA1900 is a co-polymer of ethylene glycol and acrylamide.
scissible linkers.3b However, both the loading and the ease of
The molecular weight of the building blocks determines the chain extension
+
and the pore size (chain length between cross-links) respectively. b) PEGA
handling of these resins were compromised.
2
and PEGA are prepared by the substitution of 10% of the acrylamide with
In this communication we suggest a different approach to
achieve better accessibility of large proteins into PEGA
supports. Upon introducing permanent positive or negative
charges into the acrylamide backbone, electrostatic repulsion
between adjacent chains is expected to increase the swelling of
the polymer, thereby increasing the pore size. In addition,
electrostatic attraction between oppositely charged polymer and
protein molecules is expected to aid enzymatic diffusion into
the beads.
monomers containing (3-trimethylammonium chloride) propyl and 1,1-di-
methyl-2-(sulphonate) ethyl residues, respectively. Loadings: 0.15 mmol
2
1
21
+
21
g
(
dry (PEGA1900) 0.23 mmol g dry (PEGA ) and 0.10 mmol g
dry
PEGA2).
To study both these effects, neutral PEGA1900 and novel
+
2
positively charged PEGA and negatively charged PEGA
were prepared by Polymer Laboratories (Scheme 1). The
charged polymers have the backbone of PEGA1900 and a partial
substitution with charge side chain monomers (10% substitution
of the acryl amide monomer).
+
2
The swelling behaviour of PEGA1900, PEGA and PEGA
was studied at different concentrations of phosphate buffer at
pH 8.0 (Fig. 1).† This pH value was chosen because it
represents the optimum value for hydrolysis reactions catalysed
by PGA.
+
2
Fig. 1 Swelling of PEGA
, PEGA and PEGA1900 in Kpi buffer at pH 8.0
as a function of buffer concentration.
1
296
CHEM. COMMUN., 2003, 1296–1297
This journal is © The Royal Society of Chemistry 2003

the cleaved phenylacetic acid (2) and the remaining amino acid
in the conversion is observed when comparing neutral (14.6
2
1
(
3) (Scheme 2).
µmol g dry resin) or negatively charged resin (14.4) to the
+
Enzymatic hydrolysis of (1) was observed with all three
PEGA (109.5). The maximum concentration of cleaved
+
resins (Fig. 2) indicating that all three supports are compatible
with PGA activity.
PhAcOH (2) observed was 0.033 mM (for PEGA at pH 8.0 and
0.001 M) which is below the inactivation constant of PhAcOH
+
10
For PEGA a correlation between the increase in the swelling
for PGA (0.098 mM), suggesting that product inhibition does
of the resins in the presence of diluted buffers and the yield of
the hydrolytic reaction was found. A conversion of 48% in
not play a role here.
We previously described the importance of using a low resin
to volume ratio when studying amide hydrolysis to overcome
the preference for amide synthesis on solid support. Upon
increasing the reaction volume from 6 to 18 mL, no further
increase in hydrolysis yield was observed (43%) thus making
the conversion yields obtained with PGA not fully limited by
equilibrium but still diffusion dependent. Leaving the enzy-
matic reaction for longer periods did not result in increased
hydrolysis either.
0
.001 M phosphate buffer was observed, which represents a
5
b
dramatic improvement if compared to the low yields previously
reported on solid phase supported chemistry by using PGA
3
2
(1–15%). For PEGA the reaction yield did not increase by
lowering the buffer concentration despite the increase of
swelling.
The difference between the two resins can be explained in
terms of electrostatic interactions between the polymer and the
protein molecules. PGA has an overall negative charge at pH
In summary, we show that by taking advantage of the
improved swelling properties of positively charged resins
8
+
8
.0 (pI = 5.2–5.4), and it is attracted by PEGA , while it is
2
+
repelled in PEGA . As expected, the yield obtained with
neutral PEGA1900 was independent of the buffer concentra-
tion.
(PEGA ) when using enzymes as bulky as PGA, a significant
enhancement in hydrolysis yields can be obtained. This effect is
a consequence of increased protein accessibility due to a better
swelling and electrostatic attraction between the negatively
charged enzyme and the positive resin. Overall we provided a
further step in the understanding of enzyme catalysis on solid
supports.
The authors gratefully acknowledge financial support from
the E. C. (COMBIOCAT project), the Wellcome Trust, the
BBSRC and MIUR (Rome).
The amount of protein that entered the different resin beads
9
was determined through the Pierce test. It was found that, after
2
4 h of stirring, almost 50% of the protein was inside the beads
+
in the case of PEGA , while this percentage was lower than 30%
in the case of neutral PEGA1900 and of PEGA . These results
2
are in line with the hydrolysis conversions observed and provide
evidence that electrostatic attraction occurs between the
+
positively charged PEGA and PGA.
When conversion data are expressed in terms of micromoles
of product (2) released per gram of dry resin an 8 fold increase
Notes and references
†
Polymers were washed in the appropriate buffer and then allowed to swell.
The swelling was quantified by measuring weight differences of dried and
swollen resins.
‡
Enzymatic hydrolysis was performed by washing about 100 mg of the wet
functionalised resin with Kpi buffer. The resin was then suspended in 6 mL
of the same buffer and in the presence of 5 mg of lyophilised enzyme. The
reactions were allowed to mix in a blood rotator for 24 h at rt. The samples
were analysed with a RP-HPLC system. Products amounts were calculated
using calibration curves. At the end of the reaction the mixtures were
filtered and the resins washed with 36 mL (12 3 3 mL) of ACN–H
2
O (1 :
1
). The liquid phase was recovered in a flask, dried under vacuum, re-
2
dissolved in 1 ml of ACN–H O (1 : 1), eventually centrifuged, and filtered
through 0.45 µm membrane filters. The peptide structures were then cleaved
through the Wang linker with a solution of TFA (95%) to confirm
conversion results.
Scheme 2 Synthesis of PhAc-
washing and HMPA coupling in DMF; (ii) PhAc-
capping with acetic anhydride; (iii) PGA hydrolysis in 6 ml Kpi buffer pH
L
-Phe-Wang linker on PEGA polymers. (i)
1
2
N. Bezay, G. Dudziak, A. Liese and H. Kunz, Angew. Chem., Int. Ed.,
001, 40, 2292.
(a) M. Meldal, Biopolymers, 2002, 66, 93; (b) P. M. St. Hilaire, M.
Willert, M. A. Juliano, L. Juliano and M. Meldal, J. Comb. Chem., 1999,
L
-Phe coupling and
2
8
:
2
.0, rt, 40 RPM, 24 h; (iv) cleavage of the Wang linker with TFA–H O (95
5).
1
, 509; (c) M. Meldal, I. Svendsen, L. Juliano, M. A. Juliano, E. D. Nery
and J. Scharfstein, J. Pept. Sci., 1998, 4, 83.
3
4
(a) D. W. Kadereit and H. Waldmann, Chem. Rev., 2001, 101, 3367; (b)
U. Grether and H. Waldmann, Chem. Eur. J., 2001, 7, 959.
(a) F. G. Kuruvilla, A. F. Shamji, S. M. Sternson, P. J. Hergenrother and
S. L. Schreiber, Nature, 2002, 416, 653; (b) H. Zhu and M. Snyder,
Curr. Opin. Chem. Biol., 2003, 7, 55.
5
6
(a) R. V. Ulijn, N. Bisek and S. L. Flitsch, Org. Biomolec. Chem., 2003,
1
, 621; (b) R. V. Ulijn, B. Baragaña, P. J. Halling and S. L. Flitsch, J.
Am. Chem. Soc., 2002, 124, 10988.
J. Kress, R. Zanaletti, A. Amour, M. Ladlow, J. G. Frey and M. Bradley,
Chem. Eur. J., 2002, 8, 3769.
7
8
M. Meldal, Tetrahedron Lett., 1992, 33, 3077.
V. Kasche, U. Haufler, D. Markowsky, A. Zeich and B. Galunsky, Ann.
N. Y. Acad. Sci., 1987, 501, 97.
9
T. Stich, Anal. Biochem., 1990, 91, 343.
Fig. 2 Yields of PGA catalysed hydrolysis of (1) on solid supports at pH 8.0
in Kpi buffers with different buffer concentration.‡
10 S. H. Done, J. A. Brannigan, P. E. C. Moody and R. E. Hubbard, J. Mol.
Biol., 1998, 284, 463.
CHEM. COMMUN., 2003, 1296–1297
1297