Extreme Scale-Down Approaches for Rapid Chromatography Column Design and Scale-Up During Bioprocess Development - Measurement, Monitoring, Modelling, and Control of Bioprocesses - page 127

elution steps at scale-down provided an adequate mimic of the linear gradient
established over the larger column when operated on the ÄKTA. In the second
study, separation of human growth hormone (hGH) and a precursor was attempted.
This is challenging due to the structural similarity between the proteins as well as
the use of only very small feed quantities, meaning that any spectrophotometric
quantification may experience limit-of-detection problems. Elution was performed
by an increasing salt gradient, and different resins and elution buffers were tested.
The shape and position of the UV peaks using the HTS robot compared well with
the curves produced by the miniature column and the laboratory column on the
ÄKTA. Finally, an insulin precursor and a process-relevant contaminant were
separated. Retention volumes and resolution were studied using an elution gra-
dient, and this study showed general agreement in peak shape and position pro-
duced by the robotic separation compared with the ÄKTA.
In another case, Susanto et al. [ 29 ] used scale-down experiments to populate
models that predicted scale-up column outcomes in which lysozyme adsorbs to a
strong cation exchange matrix. This study looked at the suitability of HTS (batch
uptake and miniature columns) to determine parameters for a lumped rate column
model that accounted explicitly for axial dispersion effects while lumping other
phenomena into a rate coefficient. These authors studied binding behaviour at
different ionic strengths and determined isotherm parameters on a robot. Studies
were conducted in a filter plate prepared with resin from the ResiQuot device. The
model was validated by using it to predict the behaviour of a preparative column.
4.6 Microfluidic Chromatography
Especially when operated on robotic platforms, none of the microscale technologies
discussed above provide a true representation of the continuous liquid flow obtained
in a normal column. Microfluidic chip technology provides one possible route for
achieving this. Shapiro et al. [ 24 , 25 ] described the fabrication and flow character-
isation of a microfluidic chromatography device for evaluating adsorption and res-
olution-based separations during early development. These authors developed a
system for generating a packed IEX microfluidic column containing a 1-cm-high,
1.5-lL bed. This consisted of a glass chip with a 10-mm-long column into which
compressible 6 % agarose beads were packed in a two-bead-thick layer. This was
compared with a conventional-scale laboratory column with respect to the packing
quality and similarity of breakthrough and elution profiles for egg proteins. Break-
through curves were generated using a fluorescently labelled protein at linear
velocities that fell within the range used in normal columns (between 60 and 270 cm/
h). Binding capacities were found to be similar to those achieved in conventional
laboratory beds, and microfluidic gradient elutions equated well with laboratory
columns that were 1,000 times larger. Advantages of this method included a