For aggregation-prone monoclonal antibodies, citrate buffer at pH 3.5 produced the best tested balance of elution efficiency, peak shape, and aggregate control in a Protein A capture study. Lower elution pH increased higher molecular weight species in both buffer systems tested, and pH 4.0 left the…
For an aggregation-prone monoclonal antibody, elute Protein A columns with citrate buffer at pH 3.5. That condition produced the best tested balance of elution efficiency, peak shape, and aggregate control in a study of an antibody designated Antibody X, run on the WorkBeads affimAb Edge Protein A resin. Avoid elution pH below 4.0 when aggregation is a known risk: the acidic conditions that release an antibody from Protein A also disturb the hydrogen bonds and salt bridges that hold the antibody together, and the loosened structure invites the intermolecular contacts that form aggregates.
The evidence comes from a single dataset: a Bio-Works application note carrying the publication timestamp May 12, 2026 1:04:01 PM, in which the company's team varied elution pH and buffer composition, measured higher molecular weight species by size exclusion chromatography, and tracked elution behavior through UV peak shape. The study is deliberately narrow, one antibody on one resin in one laboratory, and it should be read as such. Within those limits it tests the reader's question directly, and its conclusions fit what is known about how low pH destabilizes antibodies.
Protein A chromatography is the standard capture step in antibody purification, selected for its selectivity, dependability, and scalability. It is also typically the largest contributor to the cost of goods in an antibody process, which makes it a prime target for optimization. The difficulty is that the step demands the conditions most likely to damage the product. Antibodies bind Protein A through the Fc region, and elution depends on acid: when the surrounding pH falls, Protein A undergoes conformational changes that weaken its grip on the Fc. Typical elution pH values for antibodies on Protein A resins fall in the range of pH 2.7 to 3.5.
Those values sit below the stability threshold of most antibodies. Below pH 4.0, the hydrogen bonds and salt bridges that stabilize the native antibody structure are disrupted. The molecule becomes conformationally flexible, partially unfolds, and exposes hydrophobic regions normally buried in the folded state. Exposed hydrophobic surfaces promote non-native intermolecular interactions, and once a few molecules stick together they can nucleate higher molecular weight species HMWS . Concentration amplifies the effect: the more antibody molecules occupy a given volume during elution, the higher the probability of hydrophobic contact. This is why high-capacity resins carry a hidden risk. They concentrate antibody at the binding surface, raising the localized concentration during elution and thereby raising the chance of aggregation. Aggregates are a critical quality concern because they can reduce biological activity and provoke severe immunogenic responses in patients.
The Bio-Works team set the Antibody X runs against exactly this problem. In the initial pH-dependence series, elution was performed at pH 2.7, 3.5, and 4.0. The most acidic condition, pH 2.7, produced the highest HMWS levels, and HMWS decreased as elution pH increased in both buffer systems tested. The least acidic condition, pH 4.0, produced a broad and tailing elution peak: the antibody-Protein A interaction was only incompletely disrupted at that pH, so product continued to leak off the column. pH 3.5 cleared the column efficiently while staying above the destabilizing zone below pH 4.0.
One result in that series broke with a common expectation. With citrate buffer, the pH 3.5 elution produced a narrower peak than the pH 2.7 elution. A simple model predicts the harshest acid strips the column fastest and therefore tightest. The opposite result is what prompted the mechanistic buffer comparison that followed.
The follow-up compared two buffer systems, citrate and glycine-HCl, at elution pH values of 3.0, 3.2, and 3.5. The two buffers diverged on almost every outcome measured. The table below summarizes the results.
| Outcome | Citrate buffer | Glycine-HCl buffer |
|---|---|---|
| Peak shape trend as pH rose from 3.0 to 3.5 | Narrower peaks | Broader peaks |
| Peak width at matched pH | Narrower | Wider |
| Buffer front through the column | Uniform pH transition; pH tracked conductivity | pH overshoot at buffer breakthrough |
| Total HMWS at matched pH | Slightly higher | Slightly lower |
| pH 3.5 behavior | Narrow peak, efficient elution, controlled aggregation | Broader peak than citrate |
The central mechanistic difference was the buffer front. In citrate runs, the pH transition through the column closely followed the conductivity shift at the start of the elution peak; both shifts appeared at 0.9 column volumes. That coincidence indicates a relatively uniform buffer front. The citrate wave moved through the bed coherently, so the antibody experienced a clean, predictable pH drop that disrupted the Protein A interaction across the entire binding zone at once. The result was a narrow elution peak.
Glycine-HCl behaved differently. At elution buffer breakthrough, the pH transition overshot before settling, meaning portions of the column were transiently exposed to conditions more acidic than the target pH. The source text explaining the consequences of that overshoot is truncated, so the full mechanism is not available from this record. What the data do show is that glycine-HCl produced broader peaks than citrate at the same pH, and that the gap widened as pH increased. The two buffers also moved in opposite directions across the pH series: citrate peaks narrowed as pH rose, while glycine-HCl peaks broadened.
The aggregation data contain a tension that must be stated plainly. Citrate gave the better peak shape, but total HMWS in the pooled eluate was slightly higher with citrate than with glycine-HCl at matched pH. The Bio-Works team suggests the difference may reflect elution volume and antibody concentration in the pooled fractions. A narrower peak concentrates the product into a smaller pool, and higher concentration during elution is itself a driver of aggregation. In short, the buffer that elutes the antibody most cleanly also elutes it most densely, and density has a cost. Whether that trade-off matters depends on the process.
Fraction-wise SEC analysis added a detail that matters for pooling decisions. In the citrate elutions, peak broadening was associated with higher HMWS levels, and aggregates accumulated toward the trailing edge of the peak. The cause is physical as well as chemical: larger complexes migrate more slowly through the column because of steric hindrance and stronger interactions, so they lag behind the main product band. A process developer who pools the entire peak collects those trailing aggregates. Pooling conservatively and cutting the tail sacrifices some yield but buys aggregate control. Narrow peaks make that decision easier, because the aggregate tail is a smaller fraction of the total volume.
The application note fixed a set of operating conditions that provide context for judging the results. The feed was clarified CHO cell supernatant containing approximately 5 mg/mL of Antibody X. The column was loaded to 70% of dynamic binding capacity. The dynamic binding capacity of the WorkBeads affimAb Edge resin is specified at 60 mg/mL at 4 minutes residence time and 70 mg/mL at 6 minutes. Protein A capture alone delivered purity above 95%, the typical figure for this step in a platform process.
| Parameter | Value |
|---|---|
| Feed | Clarified CHO supernatant, approximately 5 mg/mL Antibody X |
| Loading | 70% of dynamic binding capacity |
| Capture purity | 95% by Protein A alone |
| Buffer breakthrough detected | 0.9 column volumes |
| Resin dynamic binding capacity | 60 mg/mL at 4 minutes; 70 mg/mL at 6 minutes |
Several practical rules follow, with the standing caveat that they were developed on one antibody.
Select elution pH with aggregation in mind rather than speed. In this study the entire range from pH 2.7 to 3.5 was sufficient for elution, and the mild end of that range produced the least HMWS. Nothing was gained, in peak shape or yield, by pushing below pH 3.5; the pH 2.7 runs simply added aggregate burden. For antibodies with known stability problems, screen elution pH in small steps starting near 3.5 rather than jumping to 2.7 or lower.
Consider citrate over glycine-HCl for aggregation-prone antibodies, and accept the trade-off. Citrate produced the narrower peaks and the uniform pH transition; glycine-HCl produced marginally lower total HMWS at matched pH. Narrow peaks reduce the volume sent to downstream polishing steps, saving time, resin, and buffer. The slightly higher HMWS seen with citrate, if it appears at manufacturing scale, can be managed at the pooling valve, whereas a broad and tailing peak is harder to fix after the fact.
Instrument the elution. The 0.9 column volume coincidence of the pH and conductivity shifts with citrate, and the overshoot with glycine-HCl, were visible only because the team fitted inline pH and conductivity probes. Those sensors cost little relative to a failed batch. A uniform transition, with pH tracking conductivity, is the signature of a well-behaved buffer front and is preferable to one that overshoots or lags.
Run fraction-wise SEC once at scale to learn where aggregates sit in the elution peak. The trailing-edge accumulation seen here is common, and knowing it in advance allows pooling rules to be written before manufacturing begins rather than after the first failed release test.
For challenging feeds, a guard column operated in flow-through mode upstream of the Protein A column can reduce the impurity load and extend resin lifetime. Bio-Works recommends the WorkBeads 40 TREN resin for this role. The practice is sensible independent of brand: anything that keeps host cell proteins, lipids, and aggregates off the capture column reduces the burden on the primary unit operation.
Finally, account for the concentration effect of high-capacity resins. A resin binding 60 mg/mL holds more antibody per unit volume, and when elution concentrates that bound mass into a small pool, the product may reach concentrations that drive aggregation on their own. When switching to a higher-capacity resin, re-examine elution conditions rather than carrying them over from the previous resin.
The indexed literature nearest to this topic does not test Protein A elution conditions at all. What it offers is the general chemistry of stabilizing labile peptide and protein molecules, and it is worth citing precisely so a reader knows where the boundaries of the evidence sit. A 2018 review of peptide chemistry toolboxes catalogs backbone, side-chain, and higher-order modifications for overcoming stability and bioavailability limits in natural peptides PMID 29395804 . A review of peptide and protein PEGylation details second-generation PEG derivatives, reversible conjugation strategies, and PEG architectures for improving pharmacological properties PMID 12052709 . A critical review of diketopiperazines covers their synthesis, conformation, and use as cyclic scaffolds in combinatorial chemistry PMID 12587880 . A 2024 review of peptide bioconjugation with radionuclides surveys non-classical click chemistries, including tyrosine-click, sulfo-click, SuFEx, thiol-ene click, azo coupling, hydrazone and oxime formation, and RIKEN click PMID 39458911 .
None of these addresses elution pH, buffer chemistry, or aggregation during affinity chromatography. They bear on the topic indirectly at best, as evidence for the general principle that peptide and protein stability is governed by structure and environment, and that chemical modification can rescue molecules that native conditions destabilize. For a process developer, that is a reminder that formulation and modification options exist downstream if chromatography cannot keep an antibody intact. It is not guidance on where to set an elution pH. The gap is real: most practical guidance on Protein A elution conditions lives in application notes and vendor technical reports rather than in peer-reviewed method comparisons, and this topic is no exception.
The limits of the evidence deserve the same emphasis as its findings. The subject was a single aggregation-prone monoclonal antibody, so the optimal pH and buffer may not transfer to other mAbs, bispecifics, or Fc-fusion proteins. The Bio-Works team reports that under the optimized citrate pH 3.5 conditions the resin delivered higher yield and lower aggregate levels than widely used commercial Protein A resins, but the available documentation provides no numerical comparison, no named comparators, and no experimental detail behind the claim. That is a vendor assertion, not a supported result. The dynamic binding capacity figures likewise come from the manufacturer. The glycine-HCl pH overshoot is described only in truncated form, so its full consequences for yield and purity remain unknown. The application note itself carries a header artifact reading 20:11, unlabeled in the source, possibly a reading time, a video length, or a formatting remnant; it is not data.
Several questions remain open. Why does citrate give a narrower elution peak at pH 3.5 than at pH 2.7? The note offers a hypothesis but no definitive answer. What happens after the glycine-HCl pH overshoot, and does it cost yield, purity, or aggregate control? Would citrate at pH 3.5 be the best choice for other antibody formats? How much lower were aggregate levels, and how much higher was yield, relative to the unnamed commercial resins? None of these is answered in the available record.
The state of the evidence can be summarized honestly. For this antibody, on this resin, in this laboratory, citrate at pH 3.5 was the best tested condition, and staying above pH 4.0 reduced aggregation without sacrificing elution performance. That is a useful starting point for a screening design, not a universal law of Protein A chromatography. The actionable principle is the design logic: test elution pH in the mild range, choose a buffer with a uniform pH front, measure aggregates by fraction, and pool accordingly.
Related reading: Peptide Therapeutics' Golden Age: Synthesis Advances, CMC Demands, Bio-Works Targets Peptide and Oligonucleotide Purification Growth, Fatty-Acyl CGKRK Peptide Improves Chitosan siRNA Delivery In Vitro, Stapled Antimicrobial Peptides: How They Work and What Is Proven.