Plasmids are getting larger. Gene-therapy and vaccine constructs above 10 kbp are now routine, and they have to be produced at scale and at high purity. Anion-exchange chromatography is the standard purification step — and it is exactly where large plasmids cause trouble: yields drop, pressure climbs, and columns clog.
A 2023 study in Electrophoresis — from a monolith manufacturer's own research group, which is worth knowing and in this case works in the argument's favour, because the effect was found on their own product line — looked at where that trouble comes from.
What the study measured
The authors ran plasmids of different sizes on weak anion-exchange monolith columns with channel sizes of 2, 3 and 6 µm. Rather than only reading yield from the elution peak, they followed the mass balance and the isoform composition of the plasmid at the column outlet. That is what makes the result useful: it shows not just that performance drops, but which form of the plasmid goes missing.
The open-circular form is the problem
A plasmid preparation is a mix of isoforms, mainly supercoiled and open-circular. The study identified the open-circular form as an important driver of reduced performance: in supports with smaller channels, open-circular pDNA is trapped as it is carried through by the flow. The same convective entrapment appeared on porous beads and on monoliths.
Porous beads showed a second effect on top. On an 11.6 kbp plasmid they also recovered less of the supercoiled form, because it became trapped by diffusion inside the bead pores. A convective support avoids that second mechanism; it does not avoid the first one unless the channels are wide enough.
The threshold: channels wider than 3.5 µm
One limit of the number itself: the columns tested were 2, 3 and 6 µm. The 3.5 µm figure is the authors' interpolation of where the transition lies, not a channel size that was run.
The conclusion is a number. Convective anion-exchange monoliths or membranes with a channel diameter above 3.5 µm increased yields and prevented irreversible pressure build-up and column clogging for plasmids at least up to 16 kbp. In narrower channels, the open-circular form can be trapped, and the pressure build-up that follows was not reversible.
Two limits are worth keeping in mind. The work used weak anion-exchange chemistry and a defined set of plasmids, so the exact threshold for another chemistry, a larger construct or a less clean feed can differ. And the threshold describes one cause of clogging. Staying above it does not make a column immune to fouling from other feed components.
Why channel size is a design decision
If wider channels are safer, why not make every channel wide? Because a channel is chosen against the molecule it has to carry. Permeability grows with the square of the channel diameter, so wider channels run at lower pressure and tolerate more of what the feed carries. For a small protein that headroom buys little, and the channel that suits a 16 kbp plasmid is not the one you would pick for a 150 kDa antibody.
Every convective medium sits somewhere on that curve. Channel classes of a few micrometres are common in commercial monoliths — the study itself compared 2, 3 and 6 µm. Choosing a class means matching the channel to the molecule that has to move through it and to how much the feed carries. For large plasmids, the study shows where the floor of that trade is.
Where MonoCore™ sits
That is above the published requirement, with headroom — and headroom is worth being precise about. The paper gives a second, tighter number: at least 4 µm of convective channel for high open-circular recovery of a roughly 15 kbp plasmid, and the 6 µm column in the study ran without the entrapment effect. So a channel class in the mid single digits already clears the published requirement. What additional width buys is hydraulic margin on a difficult feed and a longer run before the pressure climbs. If neither clogging nor pressure is your bottleneck, that is not a trade in your favour, and we would rather say so.
The same applies to us. A wider channel buys hydraulic headroom and tolerance to difficult feeds. For pDNA work, where clogging and yield loss are what actually limit a step, that is the right side of the trade — which is why MonoCore™ runs 15 µm rather than the 2 µm that is common commercially. Capacity figures for a given chemistry, with their measurement conditions, are on the datasheet.
What to check in your own process
Channel size is not the only lever on the open-circular form. Hydrophobic interaction separates supercoiled from open-circular directly and is the common orthogonal step next to anion exchange; size exclusion as a group separation removes RNA and endotoxin without binding the plasmid at all. If the open-circular fraction is a specification problem rather than a clogging problem, those are the steps to look at first. Channel size is the answer when the symptom is pressure and lost mass, not when it is the isoform ratio in the eluate. And where maximum surface per millilitre is what matters and the plasmid is small enough for a narrow class, that class is one we do not build — our range starts at 10 µm.
Four measurements tell you whether channel size is limiting a pDNA step. First, the open-circular content of the load, from an agarose gel or an HPLC isoform assay. Second, the isoform mass balance across flow-through, eluate and strip — a missing open-circular fraction points to entrapment rather than to poor binding. Third, the pressure trend over repeated cycles: a baseline that climbs and does not recover after cleaning is the signature of irreversible clogging. Fourth, plasmid size against the channel size of the media you are using.
If the open-circular fraction goes missing and pressure keeps rising, a wider channel is the first thing to try before tuning the gradient.
Source
Kralj Š, Kodermac ŠM, Bergoč I, Kostelec T, Podgornik A, Štrancar A, Černigoj U. Effect of plasmid DNA isoforms on preparative anion exchange chromatography. Electrophoresis 44(24):1953–1966 (2023). doi:10.1002/elps.202300035
Further reading
For why convective transport changes the flow-rate ceiling of packed beds, see our article on convective vs diffusive mass transport in chromatography. For how monoliths and membrane adsorbers compare on hydraulics, resolution and scale-up, see the monolith vs membrane comparison.
Still an open question?
Tell us the molecule, the feed volume and the system you run it on, and we will say whether a MonoCore™ capsule is the right starting point — or whether it is not.