Innovations Forum: Isoelectric titration curves
Isoelectric titration curves of viral particles as an
evaluation tool for ion exchange chromatography
S. Herzer*, P. Beckett*, T. Wegman†, and P. Moore*
*Amersham Biosciences Corp, Piscataway, NJ, USA; †Mayo Clinic, Rochester, MN, USA
Electrophoretic titration curves (ETC) were used to determine the charge characteristics of
viral particles as part of the development of a chromatographic purification strategy.
CyDye™ labelling of viral particles and in-gel detection using Typhoon™ scanner contributed
to a fast and sensitive tool for determination of chromatographic behavior of intact viruses
on ion-exchange columns. ETC results facilitated the choice of ion exchange media and
separation conditions. Results are reported for adenovirus and measles virus. The method
also worked for adeno-associated virus (AAV), murine leukemia virus, and bacteriophages
such as lambda and M13.
Introduction
Electrophoretic titration curves (ETC) are a powerful tool to
determine charge characteristics of biomolecules over a preset
pH range. ETC of protein mixtures is well documented (1–10) and
especially useful for the evaluation of ion exchange
chromatography (1, 7).
Gene therapy and the demand for high purity viral particles at
large scale have created an increased interest in chromatographic
purification (11–14). Complexity and fragility of the particle impair
approximations of charge behavior. Agarose gel electrophoresis in a
pH gradient offers the advantage of relatively mild conditions. We
describe here a fast, reproducible, sensitive method to determine a
useful working pH for chromatographic separation of intact
viral particles.
Materials
All materials including bacteriophages were obtained from Amersham
Biosciences unless otherwise mentioned. Generic chemicals for buffer
preparation were obtained from Sigma, Aldrich. SYPRO™ Ruby was
obtained from Molecular Probes. All other viruses were obtained from
ATCC, except measles virus (kind donation of Professor M. Federspiel,
Mayo Clinic, Rochester, MN).
Methods
CyDye™ fluor labelling was achieved at a slightly basic pH
compatible with virus stability. To avoid over-labelling and
crosslinking, only monoreactive dye was used. Labelling conditions
were kept at low temperature (on ice) or for short duration to
ensure that surface charge characteristics would not be affected by
over-labelling.
Manufacturers’ instructions were followed for all other procedures
unless otherwise mentioned. 5% glyercol was added to agarose ETC
gels to circumvent aggregation, which appears as excessive smearing
on the scanned gel.
16 Life Science News 13, 2003 Amersham Biosciences
Agarose ETC gels were run on a PhastSystem™ electrophoresis system
using the following program: a first step at 2000 V, 20 mA/gel, 15 °C,
7 W for 110 Vh to establish the pH gradient; electrophoresis was
stopped at 110 Vh to rotate the gel by 90°. Care was taken to mark
the direction of the cathode during the first and second step. The
sample containing intact, desalted viral particle alone or in cell lysate
was applied either with a titration curve sample applicator or into a
small incision across the width of the gel and pH gradient. The viral
sample was separated by charge/pH at 1000 V, 20 mA/gel, 15 °C,
7 W for 40–60 Vh. Progress of separation was scanned at intervals
on a Typhoon scanner at the appropriate setting.
Control agarose ETC gels were run on PhastSystem according to
manufacturer’s instructions and visualized with SYPRO Ruby on a
Typhoon scanner.
The HiTrap™ IEX Selection Kit, RESOURCE™ S, and RESOURCE Q
columns were used to separate out the viral particles by
chromatography. A pH value as determined by ETC was chosen for
separation and the desalted or dilute virus (at > 5 mS/cm,
appropriate pH) was applied to the pre-equilibrated column.
Generally, columns were equilibrated in a low strength buffer
(buffer A, e.g. 25–50 mM Tris at appropriate pH) and 1–3 column
volumes (CV) of virus was applied.
The column was washed for at least 1–2 CV and virus was eluted
with a 10 CV gradient to 100% B where B contains a salt suitable to
ensure virus stability, for example 1 M NaCl if compatible, in buffer A.
0.25–0.5 CV fractions were collected throughout.
Fractions were analyzed for virus by either infectivity or by applying
virus to an HR 5/10 column packed with Sepharose™ 6 Fast Flow
(17-0159-01) or a MicroSpin™ S-400 HR Column (27-5140-01).
The collected void volume was analyzed by gel electrophoresis for
characteristic viral banding pattern.
Innovations Forum: Isoelectric titration curves
Results and discussion
To ensure that CyDye labelling would not have a large effect on viral
charge behavior, two of the viral vectors tested were also separated
on agarose ETC without prior labelling. Gels were stained (15) and
only a slight shift in migration pattern was observed (data not shown).
Agarose gel electrophoresis of adenovirus (AV) labeled with Cy™5
indicated an overall negative surface charge of the AV from pH 5–10
and a positive charge at pH values below 5 (Fig 1). This coincides well
with the isoelectric points of hexon, penton and fiber proteins that
make up the adenoviral surface (16). To determine a useful pH range
for purification, the unlabeled starting material was also separated by
electrophoresis and stained with SYPRO Ruby (data not shown).
Fig 3. Separation of UV inactivated,
Cy3 labelled measles virus on a 2%
agarose gel pH gradient 3–10. pI,
pH values, and sample application
slot are indicated.
The virus could be separated from the bulk of the contaminants at
pH 8 on Q Sepharose XL (Fig 2). Lower pH values may be used;
however, binding capacity will be affected.
At pH 8, up to 300 mM NaCl can be used in buffer A, but this
reduces the binding capacity of the column by half compared with
separations performed without NaCl in buffer A (data not shown).
Fig 4. Analysis of viral sample by ETC in
an IEF gel of pH range 3–9. The gel was
stained with SYPRO Ruby and bands
were detected using Typhoon scanner.
Fig 1. Agarose gel electrophoresis of
Cy5 labelled adenovirus in a 2% agarose
IEF gel, pH gradient 3–10. The isoelectric
point and potential pH range for anion
and cation exchange (AIX, CIX,
respectively) are indicated.
Column:
Sample:
Sample volume:
Buffer A:
Buffer B:
Gradient:
Flowrate:
System:
Detection:
Q Sepharose XL in XK 16/10 column
recombinant adenovirus Ad5CMV-GFP
5 ml in buffer A
50 mM TrisCl (pH 8.0), 5% glycerol
buffer A + 1 M NaCl
0–100% buffer B in 20 CV
150 cm/h (5 ml/min)
sample applied at 30 cm/h (1 ml/min)
ÄKTAexplorer 10
260 and 280 nm
% buffer B
conductivity (mS/cm)
A260
A280
mAU
600
500
400
300
200
100
0
0
100
200
300
400
ml
Fig 2. Separation of adenovirus from cell lysate by anion exchange. Lysate was treated
with Benzonase™ endonuclease before loading. Adenoviral peak is indicated with an arrow.
Purity was comparable to double purification using CsCl centrifugation.
The use of capillary isoelectric focusing has been described to analyze
AV type 5 lot stability (17). However, this is the first time the use of
electrophoretic titration curves to determine charge behavior of
intact viral particles has been described.
Measles virus inactivated by UV irradiation at 320 nm for 15 minutes
was also analyzed by ETC. The agarose gel electrophoretic curve
indicated that the virus was positively charged at a pH below 7 and
negatively charged at a pH above 7 (Fig 3). However, severe
aggregation problems seemed to persist at a pH above 7. Based
on experiments with non-inactivated measles virus above pH 7,
aggregation problems seem to be common and not caused by the
UV irradiation process. However UV crosslinking is likely to have
increased aggregation.
Based on ETC results, cation exchange was chosen for separation
of measles virus. This method choice was also confirmed by
electrophoretic analysis of contaminants, which indicated that the
bulk of contaminants were negatively charged at a pH > 5.5 (Fig 4).
This was somewhat surprising because the two envelope proteins of
measles virus indicate that the virus would be negatively charged
above a pH of 5.5 according to their isoelectric points (18–20). These
observations demonstrate the usefulness of determining charge
characteristics empirically rather than based on protein sequence or
individual protein isoelectric points. The virus was successfully
purified using SP Sepharose XL at pH 6.5 (Fig 5).
Life Science News 13, 2003 Amersham Biosciences 17
Innovations Forum: Isoelectric titration curves
Ion exchange was also applicable for purification of
adeno-associated virus (AAV), murine leukemia virus, and
bacteriophages such as lambda and M13 (data not shown). Analysis
of AAV indicated an overall negative net charge above pH 7 and an
overall positive net charge below neutral pH. Contaminants appeared
to be negatively charged above a pH of 5–5.5. Purification of AAV
from cell lysate using SP Sepharose High Performance followed by
polishing on SOURCE™ 30 Q works well (21).
Moloney murine leukemia virus displayed an overall net negative
charge above pH 6 and a positive net charge below pH 6. This
corresponds well with isoelectric points described for surface proteins
(22–25). The charge transition was very sharp with a rapid increase in
migration velocity above and below that pH. Contaminants appeared
to be mostly negatively charged, as well above a pH of 5–5.5,
however, charge increase was moderate over a pH range of 5.5–8.
Anion exchange chromatography on RESOURCE Q was used to
separate the MLV from its major contaminants (data not shown).
Column:
Sample:
Sample volume:
Buffer A:
Buffer B:
Gradient:
Flowrate:
System:
Detection:
A260
A280
5. Brisabois, A. and Gullet, P. Isolation and characterization of carboxylesterase E3
from Salmonella enterica. J. Appl. Bacteriol. 75, 176–183 (1993).
6. Picard, B. et al. Genetic heterogeneity of Pseudomonas aeruginosa clinical
isolates revealed by esterase electrophoretic polymorphism and restriction
fragment length polymorphism of the ribosomal RNA gene region.
J. Med. Microbiol. 40, 313–322 (1994).
7. Watanabe, E. et al. Selection of chromatographic protein purification operations
based on physicochemical properties. Ann. N Y Acad. Sci., 721, 348–364 (1994).
8. Attanasio, F. et al. Analytical titration curves of glycosyl hydrolase Cel45 by
combined isoelectric focusing-electrophoresis. Electrophoresis, 20, 1403–1411 (1999).
9. Ameskamp, N. et al. Pilot scale recovery of monoclonal antibodies by expanded bed
ion exchange adsorption. Bioseparation 8, 169–188 (1999).
10. Sanchez, E. E. et al. Partial characterization of a basic protein from Crotalus
molossus molossus (northern blacktail rattlesnake) venom and production of a
monoclonal antibody. Toxicon 39, 523–529 (2001).
11. O'Riordan, C. A. et al. Scaleable chromatographic purification process for
recombinant adeno-associated virus (rAAV). J. Gene Med. 2, 444–454 (2000).
12. Gao, G. et al. Purification of recombinant adeno-associated virus vectors by
column chromatography and its performance in vivo. Hum. Gene Ther. 11,
2079–2091 (2000).
13. Tamayose, K., et al. A new strategy for large-scale preparation of high-titer
recombinant adeno-associated virus vectors by using packaging cell lines and
sulfonated cellulose column chromatography. Hum. Gene Ther. 7, 507–513 (1996).
SP Sepharose XL in HR 5/5 column
inactivated measles virus
2.4 ml in buffer A
25 mM sodium phosphate (pH 6.5)
buffer A + 1 M NaCl
0–100% buffer B in one step
75 cm/h (0.25 ml/min)
ÄKTAexplorer 10
260 and 280 nm
14. Huyghe, B. G. et al. Purification of a type 5 recombinant adenovirus encoding
human p53 by column chromatography. Hum. Gene Ther. 11, 1403–1416 (1995).
15. Westermeier, R. Electrophoresis in Practice, 3rd ed., WILEY-VCH Verlag GmbH,
Weinheim, (2001).
% buffer B
conductivity (mS/cm)
16. Adam, E. et al. Comparative studies on the soluble proteins of adenovirus type 1.
Acta Microbiol Acad Sci Hung. 24, 181–187 (1977).
mAU
17. Mann, B. et al. Capillary zone electrophoresis of a recombinant adenovirus.
J. Chromatogr. A 895, 329–337 (2000).
800
18. Kohama, T. et al. Maturation of measles virus hemagglutinin glycoprotein. Arch.
Virol. 85, 257–268 (1985).
600
400
19. Boriskin, YuS. et al. Measles virus persistent infection: modification of the virus
nucleocapsid protein. J. Gen. Virol. 67, 1979–1985 (1986).
200
0
0.0
2.0
4.0
6.0
8.0
10.0
12.0
ml
Fig 5. Separation of UV inactivated measles virus on SP Sepharose XL at a pH of 6.5.
Virus containing fractions (based on analytical gel filtration data) are indicated by the
black line.
Conclusion
The electrophoretic titration curve is a simple, rapid method for
prescreening conditions for ion exchange chromatography of viral
particles. The method described here relies on CyDye labelling and
detection using Typhoon scanner for rapid and sensitive detection
of viral particles. As demonstrated, ETC was used to define an
appropriate pH range for ion exchange chromatography of the intact
virus particle. In some cases, only empirical data generated by ETC
can be relied upon to determine the appropriate pH range because
published isoelectric points of capsid proteins may not be suitable
for determining purification conditions.
References
20. Andzhaparidze, O. G. et al. Analysis of the possible mechanisms for the
persistence of the vaccinal strain of the measles virus in a human cell culture. Vopr
Virosol. 33, 338–342 (1988).
21. Brument, N. et al. A versatile and scalable two-step ion-exchange
chromatography process for the purification of recombinant adeno-associated
virus serotypes-2 and -5. Mol. Ther. 6, 678–686 (2002).
22. Strand, M. and August, J. T. Polypeptide maps of cells infected with murine type
C leukemia or sarcoma oncovirus. Cell 13, 399–408 (1978).
23. Forchhammer, J. and Turnock, G. Glycoproteins from murine C-type virus are
more acidic in virus derived from transformed cells than from nontransformed cells.
Virology 88, 177–182 (1978).
24. Katoh, I. et al. Murine leukaemia virus p30 heterogeneity as revealed by
two-dimensional gel electrophoresis is not an artefact of the technique.
J. Gen. Virol. 65, 733–741 (1984).
25. Ikuta, K. and Luftig, R. B. Differences in the pI heterogeneity of virion and intracellular Moloney murine leukaemia virus p30s. J. Gen. Virol. 68, 487–498 (1987).
Ordering Information
HiTrap IEX Selection Kit
7 x 1 ml
17-6002-33
RESOURCE Q 1 ml
1
17-1177-01
RESOURCE S 1 ml
1
17-1178-01
SOURCE 30Q
10 ml
17-1275-10
SOURCE 30Q
50 ml
17-1275-01
1. Lindblom, H. et al. Separation of urine proteins on the anion-exchange resin
mono Q. J. Chromatogr. 273,107–116 (1983).
Q Sepharose XL
300 ml
17-5072-01
2. Nath, S. et al. Correlation of migration behavior in free-flow zone electrophoresis
and electrophoretic titration curve. Electrophoresis, 11, 612–616 (1990).
SP Sepharose XL
300 ml
17-5073-01
Cy3 bis-Reactive Dye Pack
1 kit
PA23000
Cy5 bis-Reactive Dye Pack
1 kit
PA25000
Typhoon 9400 & ImageQuant
Solutions for Windows 2000
1
63-0038-53
3. Avellana-Adalid, V. et al. Electrophoretic study of conformational changes of a
human soluble beta-D-galactoside-binding lectin upon storage. Electrophoresis, 13,
416–421 (1992).
4. Hull, H. H. and Wharton, D. C. Isoelectrophoretic characterization of
Pseudomonas cytochrome oxidase/nitrite reductase and its heme d1-containing
domain. Arch. Biochem. Biophys. 301, 85–93 (1993).
18 Life Science News 13, 2003 Amersham Biosciences
To request literature on ion exchange columns and media, please circle 3 on the reader
reply form provided.