Gamma Proteins

A little bit of science humour on International Talk Like A Pirate Day!

Antibody silencing - did we Fc it up? Since the 90s antibody researchers have primarily selected effector null Fc domains based simply on which variants have been used in the clinic. Frustratingly new variants are rarely compared with each other and all claim to be 'silent'. For the first time we now have a full comparison! A few years ago Geoff Hale and I analysed over 800 INNs and described their structural design and Fc mutations. To our surprise we found about 50 mutations that had been used for the purpose of Fc silencing (i.e. elimination of effector function) and we promised to follow up with a publication comparing them. It might have taken us 2 years but we did go a step further and looked at >70 silent variants in a wide range of assays (SPR, cell based, FcgR affinity chromatography and stability). This was in part because we were fed up of people asking "have you compared mAbsolve's STR with XYZ". We believe we have now tested every silencing variant that has ever been in the clinic and any that have been cited in the literature on multiple occasions. As you might expect, there are substantial differences between the variants. Some are so far from being silent it is worrying. Others are very good but fortunately for us STR still comes out on top. Our board had serious reservations on starting this project due to the risks of undermining STR but Geoff and I felt that it would be far better to have truly comparative data out in the open whatever the outcome. Now for the first time ever researchers across the world can get a clear view of how the mutations they are using compare to others, or for those new to Fc silencing it can help steer people towards a choice driven by scientific data rather than simply what has gone before (e.g. IgG4). The paper is open access and all the data can be downloaded in an Excel sheet in the supplementary data. DOI: https://lnkd.in/ejgedUNs Did we miss any mutations? Let me know in the comments! If you are interested in discussing licensing of the STR mutations drop me a message or contact us through mabsolve.com. And we didn't stop there. We've also been comparing most of the Fc enhancing mutations. That paper is in review and will hopefully be out soon. Thanks to our collaborators at Protein Stable (Alastair Davy) and RIC group (Jelle De Vos), and to Janice Reichart, mAbs and The Antibody Society for their continued support in our efforts to provide robust, reliable and publicly available data to enable data driven selection of optimal Fc variants. ----- I'm Ian, I post about antibody engineering, recombinant proteins and my journey to bootstrap Gamma Proteins into a leading supplier of Fc receptors. If you like my content please follow me to see more.

C-terminal lysine - should it stay, or should it go? In a recent posts a question was asked about the C-terminal lysine on IgG. It's readily cleaved in vivo and seems to serve no purpose but there must be some evolutionary pressure to maintain it? I didn't have an immediate answer. I suspect many people don't know so I’m sharing what I've learnt after a bit of research.... The C-term motif 'GK' is highly conserved across IgG subtypes and species. At the DNA level the GK encodes a splice site. This allows for switching between membrane and soluble forms of the antibody. This ensures the GK is conserved down to the specific codon usage. It's worth noting this as use of the GGT codon for Glycine has been shown to result in unexpected splicing with recombinant antibodies causing as much as 1-5% impurity. OK so let's remove the lysine and select the glycine codon carefully. This avoids splicing issues and any charge heterogeneity from antibodies with 2, 1 or 0 lysines. In fact about 20% of all Fc containing INNs have removed the lysine in the coding sequence, presumably for this reason. However, there are reports that removal of the lysine can result in a decrease in expression by about 50% in both antibodies and Fc fusion proteins in both transient and stable expression! OK lets keep the lysine then! I'd rather deal with a bit of heterogeneity than a 50% drop in titre. Well actually, it turns out the lysine might not be redundant at the protein level. A team at Genmab published work in 2015 that seems to have gone somewhat under the radar (at least I hadn't seen it before). By generating antibody variants with both lysines (K2), one lysine (K1) or no lysines (K0) they show that complement activation and CDC is reduced by about 3-fold when both C-terminal lysines are present. By mixing the variants the CDC is only reduced when K2 is above 50%. They propose that the positively charged lysines can inhibit hexamerization of IgG, which is critical for CDC. So rather than being a redundant amino acid the lysine may actually be part of a regulatory mechanism to protect antibody producing plasma cells which contain very high concentrations of antibody that could multimerize. IgGs are thus produced in a pro-form with the fully active IgG created rapidly on release into the blood stream by carboxypeptidase cleavage. So back to the original question. The lysine does have a role but for recombinant drugs should we retain it or remove it? It's your decision but hopefully now you can make it from a slightly more educated position. I've put some links to the relevant papers in the first comment. A big thanks to Geoff Hale for pointing me towards some of this literature and David Hayes the original question! ----- I'm Ian, I post about antibody engineering, recombinant proteins and my journey to bootstrap Gamma Proteins into a leading supplier of Fc receptors. If you like my content please reshare with your network and follow me to see more.

Did you know that viruses target FcRn for entry into cells? It wasn't something we had really thought about but in hindsight it seems like an obvious target. A very recent paper by Teressa S., Devra Huey, Cody Warren and colleagues at University of Wisconsin-Madison and The Ohio State University College of Medicine demonstrates that FcRn is a pan-arterivirus receptor. This follows on from a paper in 2019 by a different group at University of Pittsburgh that showed that FcRn is a pan-echovirus receptor. We wonder how many other viruses piggy back on FcRn for cell entry? Pan-arterivirus receptor paper DOI: https://lnkd.in/e2FVFA5d Pan-echovirus receptor paper DOI: https://lnkd.in/etzqRgF6

There seems no limit to the number of different Fc variants that antibody engineers are able to produce! A new paper from scientists at The University of Texas Health Science Center at Houston (UTHealth Houston) and University of California, Riverside describes a novel approach to screen Fc variants from CHO libraries to generate variants with improved affinity for FcgRIIa, IIb and IIIa. In particular they focus on a couple of variants that show greatly enhanced ADCC in a PBMC assay. If you are looking to enhance the effector function of your therapeutic antibody this is probably worth a read! DOI: https://lnkd.in/evbpFYkz

I'm making freely available probably the best Excel sheet for antibody sequence analysis and engineering. No license, no fee, no personal info required, no macros and no AI. Input the sequences of your mAb and it will:  - Identify CDRs (Kabat, Chothia, IMGT, you pick)  - Identify sequence liabilities  - Identify signal peptide, V domains and C domains  - Identify mutations in the C domains and their purpose  - Identify species, isotype, allotype  - Enable chimerization of your mAb into almost any other species and isotype at the click of a button (eg, switch mouse IgG1 to human IgG1)  - Introduce any mutation you want into your mAb. Type L234A and that mutation will appear in the correct place in your sequence.  - Provides analysis of 599 IgGs that have been in the clinic Download at antibodyengineering.com. If this proves popular then version 2.0 could potentially reformat any supplied sequences into scFv, Fab, minibody, BiTE, DART, other bispecific formats etc. Background on why I'm doing this: Like many people I spent years thinking Excel was just for accountants. That changed when I started building a small team at MedImmune. We had access to lots of expensive software but my team kept making very basic maths errors in the lab. Out of frustration I started building little Excel sheets. A morning transfecting cells is difficult enough without having to keep working out how much HC and LC to add. Simply input your DNA concentrations and desired transfection volume into my sheet and it will create a worksheet for the day for you. Suddenly our productivity and accuracy shot up. The major impact came when I realised that Excel is a character manipulation tool, this includes letters as well as numbers. With a bit of outside the box thinking and Googling almost anything becomes possible. Translating DNA to Protein is just substitution of one string of text for another. Easy for Excel. Molecular weight determination is just assigning a value to each letter and performing a sum. CDR identification, humanization, Fc analysis or mutation, it's all possible. Hell, I've even analysed Sanger sequencing data in seconds with a home made Excel tool. Of course there are many online tools that can do some of this. They tend to be specific to one task and don't always allow entry of many mAbs once - inputting one at a time is a killer if you have lots. Excel is easy for any user to get to grips with. It has limitations but it's taken me a long way. Many of my connections will have paid for tools that can do all of this and more. There are many people though that don't have access to these tools and might be able to make use of this. Please do comment or reach out if you find any errors or have ideas on useful tools for future versions. If you don’t already follow me, please do so to keep up to date with any new versions I launch. I was hoping to just attach the file to LinkedIn but that doesn’t work, hence the hastily put together website!

TRIM21 is an intracellular Fc receptor. What's the purpose of an intracellular Fc receptor when antibodies don't enter the cell?! Well it turns out they can do so when attached to viral capsids and thus TRIM21 has emerged as a pivotal player at the interface between adaptive and innate arms of the immune system in response to viruses. We still have a lot to learn about TRIM21 and this recent publication from Johannes R., Philippe Ringler, Christian Klein, Tilman Schlothauer and colleagues at Roche explores the interaction of TRIM21 with a range of different Fc-engineered antibodies. It also acts as a great introduction for those less familiar with this interesting intracellular Fc receptor. DOI: https://lnkd.in/ecuh9N9N Gamma Proteins doesn't supply TRIM21 yet but it is on our list of receptors we would like to add to our portfolio!

This new preprint from Paul Kremer, Elizabeth Lampros, Allison Blocker and Adam Barb at the The University of Georgia shows the importance of the glycoform on N162 of FcgRIIIa and how this impacts binding to the IgG Fc domain and resulting ADCC. In previous work they showed a high degree glycan heterogeneity at this site in healthy donors resulting in high and low-affinity variants of the Fc receptor. Now they explore the mechanism underlying this. DOI: https://lnkd.in/eQyy8q8Q Be careful when purchasing Fc receptors. There are many suppliers using many different expression platforms. Glyosylation of Fc gamma receptors is critical to their activity so at the very least please ensure any receptors you purchase have been produced in a mammalian cell line. See gammaproteins.com for our mammalian expressed human FcγRs.

This new paper from Johannes R., Jan Terje Andersen, Ulrich Rant and Tilman Schlothauer is a fantastic analysis of the role of avidity and affinity in the pH dependent interaction of IgG with FcRn. They demonstrate the use of DNA nanolevers with SPR to mimick the membrane bound state of the receptor and enable the determination of affinity and avidity for complex interactions in a single measurement. DOI: https://lnkd.in/eyVWNg4a

What the Fc is up with ADCs? ADCs are the hot topic in antibodies at the moment. Not a week goes by that we don't see news of another ADC entering the clinic or being acquired. Much of the appeal of ADCs is their phenomenal potency but this is a double edged sword and toxicity is a real risk. It's always struck me as odd that ADC developers don't seem to pay much attention to the Fc domain. Surely for most ADCs effector function (i.e. ADCC, ADCP or CDC) isn't part of the primary mechanism of action? If that's the case then the logical approach would be to utilise an Fc domain with reduced effector function to avoid unwanted toxicity from binding to and killing immune cells expressing Fc gamma receptors. ADCs are toxic enough as it is so why run this risk? I always had the feeling this was largely ignored in the ADC space but didn't have the stats to back this up. So yesterday I looked at 70 ADCs, which was most if not all the ADCs that had been in the clinic up to 2022. Over 90% had a fully active human IgG1 Fc domain. This is in stark contrast to antibodies as a whole (from an analysis of +800 INNs) where about 45% are 'silenced' (see https://lnkd.in/e2n3PpeJ). As a developer of one of the best silencing mutations (see https://lnkd.in/eq6P2sGT) I'm somewhat biased, but what's going on here? Are ADC developers missing a trick or is there some logic to the choice of wild type IgG1 in almost all cases? I'd love to hear from people developing ADCs! Note - my analysis doesn't account for the fact that some of these antibodies will utilise the glycans for site-specific conjugation of the payload and this will presumably reduce effector function to some extent. I don't have details on the conjugation process for each mAb in my dataset but I expect that glycan conjugated mAbs only account for a small number (although I accept this is becoming increasingly common in recent years). ----- I'm Ian, I post about antibody engineering, recombinant proteins and my journey to bootstrap Gamma Proteins into a leading supplier of Fc receptors. If you like my content please reshare with your network and follow me to see more.