Introduction

Monoclonal antibodies (mAb’s) have

gained much attention as therapeutics

due to their specificity for a single anti-

genic determinant. Various mAb’s are

approved by the Food & Drug

Administration (FDA) and used for clini-

cal therapy

. To use an mAb as a drug

substance, biological safety, quality,

purity and performance characteristics

of the mAb must be assessed during

manufacturing

. Therefore, the main

objectives towards manufacturing Ab-

based pharmaceuticals are production

and purification of high amounts of

antibody with good yields.

During biosynthesis, mAb’s can under-

go various modifications that lead to

charge heterogeneity. These provide a

great challenge for the complete char-

acterization of the mAb. Combinations

of charge and size analysis provide an

excellent assessment of antibody puri-

ty. Traditionally, slab-gel-based isoelec-

tric focusing (IEF) or capillary elec-

trophoresis (CE) methods are used to

monitor the charge profiles of antibod-

ies. However, although they provide

good resolution, gel-based or CE-based

IEF offer little opportunity to introduce

a second, quantitative dimension of

analysis.

OFFGEL electrophoresis is an alterna-

tive technique to slab-gel based IEF in

modern protein analysis

. It fraction-

ates peptides or proteins based on

immobilized pH gradients. However, in

contrast to gel based techniques, sam-

ples are recovered in solution for down-

stream applications such as size analy-

sis or liquid chromatography-mass

spectrometry (LC/MS). It is flexible in

resolution and in the amount of sample

that can be loaded (analytical to semi-

preparative sample amounts). IEF by

OFFGEL fractionation of proteins or

peptides can achieve a resolution of

0.1 pH. Isoelectric point (pI) values

obtained from OFFGEL runs serve as

additional support for MS analysis and

can be used to search for post transla-

tional modifications of proteins and

peptides.

For size analysis of pI fractions, the

Agilent 2100 Bioanalyzer is the tool of

choice. The High Sensitivity Protein 250

(HSP-250) assay provides size separa-

tion and highly sensitive detection of

proteins in the picogram range.

This study compares the two-dimen-

sional application of OFFGEL and

Bioanalyzer for monitoring the charge

heterogeneity of mAb’s under native

conditions and after addition of a deter-

gent (Tween-20). The study outlines the

methodology difference between dena-

tured and nondenatured conditions.

Materials and Methods

Tween-20 was obtained from Sigma

(MO, USA). Purified monoclonal anti-

body (1 mg/ml) was from Stratagene

(CA, USA).

For IEF, the Agilent 3100 OFFGEL

Fractionator with a 24-well frame setup

(OFFGEL High Resolution Kit, pH 3-10,

Agilent p/n 5188-6425) was used with

a slight deviation from the original pro-

tocol. For each OFFGEL fractionation,

100 µg of the mAb sample were dis-

solved in OFFGEL focusing buffer con-

taining 5 % glycerol, and 0.25 % IPG

buffer pH 3-10. Neither urea or thiourea

were added to maintain native condi-

tions because mAb’s can easily disinte-

grate under urea/thiourea denaturing

conditions (data not shown). A second

data set was generated with the addi-

tion of 1% Tween-20 to enhance the

solubility of the proteins. OFFGEL frac-

tionation was performed using the

default method for protein samples and

24 fractions (OG24PR01). The recov-

ered fractions were subjected to

Bioanalyzer protein analysis. Fractions

obtained after the OFFGEL runs are

directly compatible to the HSP-250

assay. OFFGEL recoveries were calcu-

lated by relating the mAb quantity in all

the OFFGEL fractions to the concentra-

tion measured in non-fractionated sam-

ple before OFFGEL fractionation

(referred to as sample load).

Protein analysis was done on the

Agilent 2100 Bioanalyzer with the

Agilent High Sensitivity Protein 250 kit

(Agilent Technologies GmbH,

Waldbronn, Germany). Protein labeling

and on-chip sample analysis was per-

formed as described in the High

Sensitivity Protein 250 Kit Guide

. All

samples were run under nonreducing

conditions.

Results and Discussion

In this study, to assess the influence of

protein structure on pI fractionation,

mAb samples were fractionated by pI

using OFFGEL electrophoresis under

native conditions with the addition of

Tween-20. The recovered OFFGEL frac-

tions were analyzed with the

Bioanalyzer HSP-250 assay. Tween-20

was added to the OFFGEL buffer system

to enhance the solubility of the protein

during focusing, because mAb’s are

known to disintegrate easily under urea

and thiourea buffer conditions (data not

shown). The main motivation for sepa-

rating mAb under native conditions is

the direct compatibility with down-

stream LC/MS analysis for identifica-

tion of the fractionated charge vari-

ants

Figures 1A and 2A show Bioanalyzer

gel-like images for OFFGEL fractions

and the initial sample load obtained

under native conditions and after addi-

tion of Tween-20, respectively. Various

mAb charge variants are well separated

by OFFGEL fractionation and subse-

quently detected with the HSP-250

assay. Charge variants from the main

mAb product are all found towards the

anode and are therefore exhibiting

acidic pI values. Figures 1B and 2B

show corresponding electropherogram

overlays for the OFFGEL fractions of the

main product and the acidic charge

variants. The initial sample load is

shown as well. The difference in focus-

ing pattern between the two conditions

applied is clearly visualized by the

Bioanalyzer protein assay. With the

addition of Tween-20, more main prod-

uct (fraction 9-11) and charge variants

(fraction 1-5) are observed as indicated

by the higher peak intensity values of

the corresponding electropherograms

(Figure 2). In terms of molecular weight,

charge variants differ from the main

products by a maximum of just 4 kDa

and represent only 10% of the total

mAb preparation. Detection of such

variants is heavily dependant upon the

initial sample fractionation by pI,

because their overall contribution and

difference in molecular weight do not

allow direct detection with the

Bioanalyzer (see sample load,

Figures 1 and 2).