Yusuf Johari

by Yusuf Johari, Ph.D.
Senior Principal Scientist, R&D


Smarter expression approaches for multi-chain therapeutic protein formats.

Our customers play a critical role in driving innovation, often advancing multi-chain and new therapeutic formats to transform patient outcomes. To keep pace, cutting-edge therapeutic protein expression technology is essential - boosting titers, product quality, and stability while accelerating timelines. These tools are critical to improving bioproduction outcomes and increasing the chances of success on the path to first-in-human trials and beyond.

The rise of multi-chain therapeutic proteins

In recent years, there has been a sharp increase in the development of multi-chain therapeutic proteins, including bispecifics, multispecifics, fusion proteins, and other advanced formats. The appeal is clear: these molecules can offer enhanced targeting capabilities and the potential to deliver transformative outcomes for patients. However, innovation can bring challenges and the need to develop novel solutions, especially when it comes to biomanufacturing.

Through my 16 years’ experience in bioproduction, spanning both academic research and my role as a Senior Principal Scientist at Lonza, I’ve seen firsthand the hurdles associated with expressing these multi-chain molecules efficiently. Low expression titers, complex vector construction workflows, and product quality issues (e.g. incorrect chain pairing) are among the most persistent and limiting challenges.

Building better tools to advance multichain therapeutic protein expression

In bioproduction, the design of the expression vector is foundational. It’s the first step in the biomanufacturing workflow and plays a critical role in determining how successfully a protein can be produced. While expression vectors include multiple components (you can find a detailed review I co-authored here), in this blog post I’ll focus on one of the most influential elements: the gene promoter.

Promoters are the regulatory sequences that drive expression of the transgene (the gene that encodes the therapeutic protein). Traditionally, strong viral promoters like CMV have been widely used and have contributed to the success of CHO cells as the gold standard expression platform. However, for more complex, multi-chain proteins, traditional promoters don’t always provide the flexibility or control needed for optimal expression.

That’s why my team set out to develop novel synthetic promoters - engineered regulatory elements that offer a range of transcriptional strengths, that could be used alone, or in a combinatorial fashion, to enhance multi-chain protein expression. Drawing from our unique understanding of both viral and endogenous regulatory elements, my team has been able to construct promoters with distinct characteristics. This has enabled us to design novel strong CHO promoters as well as promoters with lower transcriptional activities to co-express multiple recombinant genes at different levels (Figure 1, panels A and B).


Figure 1: Varying expression strengths with novel synthetic promoters. As assessed by fed-batch harvest titre (panel A) and fold-change heavy chain mRNA level (panel B). LC; light chain. HC; heavy chain.

One of the key advantages of our synthetic promoter approach is the ability to introduce features that better align with cellular biology, such as the ability to regulate gene expression dynamically across different growth phases. For example, we’ve engineered promoters that upregulate expression during the stationary phase of culture, effectively decoupling growth and production. This helps direct cellular resources more efficiently, first toward proliferation and then toward protein production, potentially improving both product quality and expression stability (see Figure 2).



Figure 2: Upregulation of expression in stationary phase. Across a fed-batch process for stable pools expressing a mAb, average cell-specific productivity (qP) is relatively consistent when using the standard CMV promoter to control HC and LC expression. However, using a new synthetic promoter upstream of the HC and LC genes achieves a substantial increase in qP during the latter stages of fed-batch when growth of the culture slows, and cells typically enter the major production phase.

The output from this work informed the design of Lonza’s new GSquad® Pro vector system and its constituent high-performing LHP-1 promoter. When considering expression strength, a high performing in-house engineered gene promoter candidate, LHP-1, contributes to increased titers for a range of molecules relative to the CMV promoter, while supporting excellent product quality and expression stability. This vector system enables the construction of an expression vector capable of co-expressing of up to 4 product genes, offering a streamlined process that can save time, and also reduce inherent variability associated with co-transfection of single gene vectors.

Helping biotechs on the path to success

At Lonza, we have extensive experience in expressing multi-chain products such as bispecifics. Around 40% of molecules expressed at Lonza for early discovery and development in 2023 were bispecific formats. My colleagues in cell line development recently tackled a customer project for a difficult-to-express bispecific format, where the use of the GSquad® Pro vector system containing the LHP-1 promoter, boosted titers by over 40%, achieving ~ 5 g/L (Figure 3).



Figure 3: Customer case study

Engineering the future of bioproduction through synthetic biology, automation, and predictive tools

At Lonza, we apply the design-build-test-learn framework of synthetic biology to continually refine and optimize our promoter designs, ensuring they deliver predictable performance across scalable bioproduction systems. But we’re not stopping there.

To future-proof our vector platform, we’re building high-throughput, automated pipelines for vector assembly and stable pool generation within R&D. This infrastructure will enable us to rapidly evaluate hundreds of promoter and gene-order combinations, as well as assess the effects of additional regulatory elements on expression performance.

Looking ahead, large datasets generated by this approach could power AI models that suggest optimized vector designs tailored to each molecule, greatly streamlining empirical screening and guiding more targeted experiments.

About the author:

Yusuf Johari, Ph.D. is a Senior Principal Scientist in R&D at Lonza, Cambridge. He leads a dedicated research team focused on the development of advanced expression vector technology to improve bioproduction outcomes in CHO cells. Yusuf has extensive experience in the field of bioprocessing research, both in academia and industry, and holds an MSc and PhD from the University of Sheffield.

 

Need support with your own biotherapeutic protein expression projects?

The expression of multi-chain therapeutic protein formats can be challenging, but by using optimized technologies from an early stage, even expression of these more complex molecules can be made easier. If you need support and guidance, don’t hesitate to reach out to a member of our team so that you may learn how we can support you all the way from discovery to commercial production.

 

* The presented information was correct and current at the time of publication.
Latest briefing from the Knowledge Center