How do you make an antibody let go of its antigen at pH 6?
When you need this
Three situations account for most pH-engineering programmes we run.
- Antigen sink is destroying your PK. A high-abundance or rapidly turning-over antigen consumes antibody by target-mediated clearance. An antibody that releases antigen in the endosome is recycled by FcRn and returns to circulation, so one molecule neutralizes many antigen molecules instead of one.
- Dose or dosing frequency is the commercial problem. Recycling behaviour is one of the few engineering levers that changes exposure without touching affinity at physiological pH.
- You want activity confined to acidic tissue. The tumour microenvironment sits below plasma pH. Inverting the logic — binding preferentially where pH is low — is a way to widen a therapeutic window without a masking domain or a protease trigger.
How we do it
We select clones that bind at pH 7.4 but release rapidly at pH ~5.8, using STEM™ library engineering to expand diversity around the parent and then applying the pH condition as the selection pressure rather than as a post-hoc screen. That distinction matters: histidine substitutions can be proposed computationally, but whether a given substitution buys real pH dependence without costing affinity, expression or stability is an empirical question, and a selection answers it across thousands of variants at once instead of one at a time.
We have run this from a pH-insensitive parent. Starting from an anti-TNFα antibody with no intrinsic pH sensitivity, we engineered pH-sensitive clones and confirmed binding at pH 7.4 with dissociation at pH 5.8 by biolayer interferometry (Octet).
We report the measured kinetics from your own campaign. We do not quote a headline fold-change from an unrelated programme as if it predicted yours — pH-dependence is highly epitope-dependent, and a number from someone else’s antibody is not a forecast for yours.
- pH-sensitive variants derived from your parent antibody, with sequences
- pH-dependent binding and dissociation data (Octet) at pH 7.4 and pH ~5.8
- Affinity at physiological pH for each variant, so you can see what the pH-dependence cost
- Expression titre for the panel, so a variant that cannot be made does not reach your decision table
What we will tell you before you commit
pH engineering does not succeed on every epitope. If the paratope has no residues positioned where a histidine could act as a protonation switch, the achievable pH window is narrow and the honest answer is that the effort is better spent elsewhere. Send the sequence and the antigen and we will say which of those two situations you are in before a purchase order exists.
Have a lead that gets consumed by its own target?
Send the sequence and the antigen. A PhD scientist — not a sales rep — reads it and replies within two business days with a view on feasibility. No NDA needed to start.
Talk to a scientist →Frequently asked questions
Do I need to give you my antibody sequence to start?
Not for a first conversation. We can assess feasibility from the antigen, the format and a description of the problem. Sequences are needed once a campaign is scoped, and they stay confidential.
Will pH engineering cost me affinity at pH 7.4?
Often some, which is why we measure it and report it per variant rather than only reporting the pH ratio. Whether the trade is acceptable depends on your target; that is a decision for you to make with the numbers in front of you.
Can you combine this with affinity maturation?
Yes, and it is common. STEM™ affinity maturation and pH engineering use the same library and selection machinery, so they can be run as one continuous campaign rather than two sequential projects.
Who owns the resulting variants?
You do. Standard engagements return the sequences to the sponsor and carry no downstream royalties in standard scope.
Can we order through Scientist.com?
Yes. Abwiz Bio is a registered supplier, so if your organisation already buys through Scientist.com there is no new vendor set-up. Direct contracting is equally welcome.