A Deep Dive into China Protein A Resins: The Road to Replacing Imported Affinity Chromatography Resins

In the commercialization of antibody drugs and recombinant protein biologics, affinity chromatography resins are a core strategic consumable in downstream purification processes, directly determining drug purity, yield, batch consistency, and production costs. For a long time, China’s high-end biopharmaceutical purification market has been heavily dependent on overseas products. The supply chain has long lead times, high procurement costs, and slow customization response, which has become a key bottleneck restricting the large-scale and independent development of domestic biopharmaceuticals.

According to QYResearch, the global chromatography purification resin market is expected to reach $335 million in sales by 2025, growing to $607 million by 2032. Data from the Forward Industry Research Institute shows that China’s biochromatography media market is around 6–9 billion yuan, but the market share of domestically made resins in the high-end sector is still relatively limited.

Under the major theme of replacing imported chromatography resins, domestic affinity chromatography resins are the core track, and domestic Protein A resins are the top technical stronghold and most valuable target on this track—the relationship of the three is from large to small, from broad to specific. With the fast iteration of domestic protein engineering and synthetic biology technologies, domestic Protein A resins now fully match international high-end standards in key metrics like binding capacity, alkali resistance, cycle life, and impurity removal, officially moving from a 'cheap alternative' to a 'superior alternative' stage.

1. Affinity Chromatography Resins: Why Are They a "Bottleneck" Consumable?

To understand the strategic significance of domestic affinity chromatography resins, we first need to grasp how important they really are in biologic drug production.

First, they account for a high proportion of costs. Taking monoclonal antibody production as an example, downstream separation and purification typically account for 30%-50% of total production costs (higher for small-scale processes, lower for mature large-scale processes), and Protein A affinity resins are among the single largest consumable expenses in downstream purification. The industry has a rough estimate: for a 2000L antibody fermentation scale, the annual procurement cost of just imported Protein A resins can reach tens of millions of yuan. Each 10% increase in resin binding capacity or extending its lifespan by 50 cycles can mean significant cost savings.

Second, the technical barriers are high. A seemingly simple microsphere resin actually has three layers of technical challenges behind it:

Matrix Microsphere Preparation: The goal is to achieve highly uniform particle size, even pore distribution, sufficient mechanical strength, and low nonspecific adsorption. The process for preparing highly monodisperse, precisely controllable high-crosslinked agarose microspheres has long been a core technological barrier for overseas companies;

Surface Modification and Ligand Coupling: Making the microspheres is just the beginning. You need to evenly modify functional groups on the microsphere surface and then stably, directionally, and densely couple the ligands — even with the same Protein A ligand, different coupling methods can lead to several-fold differences in loading, lifespan, and shedding;

Regulatory Compliance and Validation: Since the filler comes into direct contact with drugs, regulatory requirements are extremely strict. Once a customer locks in a supplier, the cost of switching is very high and the stickiness is strong.

Third, high import reliance. The global market is highly concentrated, with a few overseas giants holding the main share. High-end Protein A industrial fillers have long been monopolized. Imported products are not only expensive, but some customized products have supply cycles of 6-8 months, posing a risk of supply chain disruption.

2. The Core Bottleneck: Why are high-end ligands so hard to make?

 Three Technical Barriers to Filler Production

In the three major technological barriers, significant progress has been made in the domestic alternatives for the matrix microspheres and coupling processes. However, ligand design—especially high-end alkali-resistant ligands—remains the core technical bottleneck. This is also the key challenge that domestic Protein A resins must overcome to truly replace high-end imports.

The upper limit of resin performance essentially depends on the upper limit of the ligand performance. The microsphere matrix determines the resin’s 'hardware foundation' (flow rate, pressure resistance, pore size, etc.), while the ligand determines the 'functional core' (specificity, binding capacity, alkali resistance, elution conditions, etc.). Using the same matrix but coupling different ligands can result in drastically different resin performances.

High-end alkali-resistant Protein A ligands are difficult because of three 'balances':

- Balance between alkali resistance and affinity. Natural Protein A rapidly inactivates in high-concentration NaOH. The main chemical mechanism is non-enzymatic deamidation of surface Asn residues (especially at sites in flexible sequences like Asn-Gly, Asn-Ser) via succinimide intermediates, converting them into Asp/isoAsp; at high pH, peptide bond hydrolysis and other degradations can also occur. To improve alkali resistance, these sensitive sites must be mutated—but mutations often affect overall protein folding, which in turn affects binding affinity.

- Balance between capacity and elution. Higher ligand density theoretically increases binding capacity, but overly high density can cause steric hindrance, reducing effective binding and potentially making elution harder while increasing residual impurities.

- Balance between stability and specificity. The firmer the ligand coupling, the lower the shedding rate; however, overdoing the coupling chemistry may damage the ligand’s spatial structure, reducing specificity and increasing non-specific adsorption.

Traditional ligand engineering relies on expert experience and high-throughput screening—modify one site, test a batch of activity, then tweak the next site... Optimizing an industrial-grade alkali-resistant ligand from the initial template to maturity typically takes 2–5 years, screening thousands of variants, with extremely high trial-and-error costs.

This is precisely where AI ligand design comes in as a breakthrough.

3. AI Closed-loop Wet-Dry: A "Shortcut" Opportunity for Domestic Protein A Resins

AI Closed-Loop Iteration of Dry and Wet Processes

The reason ligand design has become a major bottleneck is fundamentally that traditional methods rely on long cycles of trial and error. AI protein design technology, however, offers a paradigm shift from 'trial and error' to 'design'—and this is precisely where domestic reagents have the chance to take a leap forward.

Shanghai Matwings Technology’s MatwingsVenus™ (Xiaowu™) AI protein large model platform is exploring a path to independently controllable high-end domestic affinity ligands. The core idea is to use AI to improve ligand design efficiency, and use a dry-wet closed loop to ensure implementation success, starting at the ligand—the most critical functional layer—to provide high-performance, independently controllable ligand technology for domestic affinity chromatography resins.

Specifically, this system supports performance upgrades of domestic Protein A resins in three ways:

3.1. 'Precision guidance' for alkaline-resistant ligand modification

Traditional alkaline-resistant modifications rely on structural biology experience combined with high-throughput screening: identifying candidate Asn sites, making point or combination mutations, and then testing activity and alkaline resistance. The incremental value of AI is not 'creating from scratch,' but in modeling multiple conflicting objectives at once—alkaline resistance, affinity, folding stability, coupling orientation—and proposing combination mutation strategies across a much larger sequence space in one go.

Take the AI-designed alkaline-resistant Protein A developed on the MatwingsVenus™ (Xiaowu™) platform as an example: the team used the AI large model to precisely design key sequence features and structural determinants of Protein A ligands, successfully producing a new generation of recombinant Protein A variants with both stronger alkaline resistance and high capacity. These can handle 1M NaOH CIP (clean-in-place) while maintaining high dynamic binding capacity. The R&D cycle was shortened from the traditional 2-5 years to 2-6 months, with experimental sample numbers reduced by over 90%.

3.2. Joint optimization of ligand and coupling process

Even if the ligand itself is excellent, poor coupling processes can ruin it—the wrong coupling site can cause disordered ligand orientation and greatly reduce effective binding, while unstable coupling chemistry leads to high ligand loss. AI can design the ligand sequence itself and assist in designing optimal coupling sites and strategies—for example, choosing specific amino acids on the protein surface for site-specific coupling, ensuring correct ligand orientation (boosting effective binding) and enhancing chemical stability of the coupling bond (reducing loss).

3.3. Rapid Iteration of Dry and Wet Closed Loops

The unique value of MatwingsVenus™ (Xiaowu ™) lies in its fact that it is not a pure computing platform, but rather a closed-loop system for automated wet experiment verification. Candidate combinations designed by AI can be directly integrated into automated experimental platforms for expression, purification, activity detection, and packing coupling testing. Experimental data is fed back to the next round of AI optimization, forming a complete closed loop of "design-validation-iteration."

The value of this model lies in its elevation of basis research from "purely computational theoretical talk" to the "real world of experimental verification"—no matter how accurate the computational predictions are, they still rely on experimental data for calibration; The data accumulated from each round of experiments helps the model more accurately understand the relationship between the ligand sequence and performance, guiding the next round of more efficient optimization. As validation data accumulates, the iterative efficiency of the basis performance will continue to improve.

4. Deep Zone of Domestic Substitution: Three Major Trends Have Become Clear

Looking at the domestic affinity chromatography packing industry from the perspective of 2026, three trends are already very clear.

First, domestic substitution has entered deep waters. According to industry data, by 2025, the overall market penetration rate of chromatography fillers in China will be about 41%, with a selection rate of 68.5% for clinical R&D, but only 28.9% for GMP commercial production. The replacement of clinical services is basically complete, and the next main battlefield is commercial production—not just price and supply chain, but also a comprehensive contest of performance, stability, compliance documentation, and technical service capabilities. It is worth noting that domestic substitution has shifted from customers' "passive choice" to "active selection," with driving factors evolving from initial supply chain security and price advantages to multi-party collaboration of cost competition, technological breakthroughs, and local services.

Second, ligator technology has become a new focus of competition. As domestic microsphere matrix technology gradually catches up with imports, the focus of competition is shifting from "hardware" to "software"—that is, base mixing. The next-generation competition for domestic Protein A fillers is essentially a contest of ligand performance. Whoever can create high-end alkali-resistant bases with performance comparable to or even surpassing imports will secure their ticket to the high-end market.

Third, AI-driven base design offers an opportunity to "overtake competitors by changing lanes." The traditional approach to base engineering is pursued by domestic companies—overseas companies have decades of technological accumulation and patent barriers. But in the new track of AI design and base matching, the gap between domestic and international markets is much smaller. China has deep technical expertise in AI large models and protein design, giving domestic companies the opportunity to leap from follower to alongside or even leading through AI ligand design technology routes.

This is precisely the significance of the MatwingsVenus™ (Xiaowu ™) platform—it serves as the "technical foundation" for providing high-performance substrates for the domestic packing industry. When AI-designed substrates meet domestic microsphere matrices and mature coupling processes, truly independently controllable high-end domestic affinity polymerization fillers can be born.

5. Conclusion

The domestic replacement of imported affinity chromatography resins is a tough battle with no shortcuts.

From catching up in microsphere matrices, to refining coupling processes, and tackling ligand technology, the domestic resin industry is gradually climbing up to the higher end of the technology chain. Meanwhile, the maturation of AI protein design technology has given this battle a new weapon—it allows the development of high-end ligands to rely less on expert experience and long trial-and-error processes, and more on data-driven precise design to speed up breakthroughs.

When domestic Protein A resins are no longer just "cheap alternatives to imports" but have their own core technologies and differentiated advantages, the story of replacing imported chromatography resins will truly reach its most exciting chapter.