Protein Kintsugi: repairing recombinant influenza neuraminidase

The majority of influenza vaccine research focuses on the viral surface protein hemagglutinin (HA). Although Neuraminidase (NA), the less abundant viral surface protein, is the target of licensed antivirals, it has been less explored as a viable vaccine target, due in part to its lesser stability. In addition, the current seasonal influenza vaccine does not account for NA content, providing ample room to improve influenza vaccines through the supplementation of properly-folded NA antigens.  

Our story outlined in Nature Communications began with a simple question: How does influenza NA behave when recombinantly expressed, and can we eventually use recombinant NA as a possible influenza vaccine candidate? This seemed to be a simple question, but the road to the answer proved to be both bumpy and winding. 

In the literature, NA has a history of being structurally presented as a “closed” tightly packed tetramer in which all four globular head domains are in close contact. This structural conformation is regularly seen in crystal structures and in viral tomography. When analyzing recombinant NA antigens by electron microscopy (EM), we expected to see similar structures. But that was not the case. We noticed that the NAs of certain subtypes instead formed “open” conformations in which heads were not properly interacting. With vaccines in mind and immunogenicity closely linked to the structural integrity of an antigen, these results put us in a tough position: Is this open state biologically relevant, or an artifact of recombinant expression?

Knowledge in the influenza field implied that this open state was a known issue that wasn’t yet formally addressed in literature. So we decided to “fix” this presumably non-biological open conformation to a closed, biologically relevant conformation.

"Closure"
Neuraminidase
Shifting clover form now clear
sNAp tetramers closed

After our initial results, we knew that we had to get protein designers involved and started our fruitful collaboration with Daniel Ellis and Neil King from the Institute for Protein Design at the University of Washington. Dan was excited to tackle the problem. We identified the misbehaving NAs, compared their structures with well-behaving (i.e., closed tetrameric) NAs, designed combinations of homology-directed and/or novel mutations and expressed, purified and characterized them. We quickly learned that there were patterns in the spatial locations we introduced our mutations. Some helped in closure, others helped in expression and some appeared irrelevant. 

Figure 1: Overview of our design strategy to close NA.

“Golden Joinery”
Protein kintsugi
Essence of NA restored
Beauty through EM

After many designs, we were able to stabilize the closed state of several different NAs from multiple subtypes and confirmed this in precise detail with a cryo-EM single particle reconstruction at 3.2 Å resolution obtained by Ollie Acton in David Veesler’s lab at UW. Further, our stabilized NA proteins (called “sNAps”) showed improved affinity to infection-induced monoclonal antibodies, which further suggested that this closed conformation is more biologically relevant than the open conformation.

"Untitled"
Hundreds of designs
In search for NA truth
We all need a break

Our story is admittedly only one part of a larger effort to understand this antigen, and there were so many more less successful constructs that were designed, expressed and characterized but were not included in the final manuscript. Keen readers might have also noticed that we did not yet answer the second part of our question: can we use the recombinant NA proteins as an influenza vaccine candidate? Our journey to close and stabilize NA was more challenging than we initially expected, but our efforts to translate sNAps into influenza vaccines continue. 

Neuraminidase, neuraminidase
Such a sweet taste
We desired to express you as a recombinant protein, 
But you decided to be open/closed/mixed in shape and not clean
Why oh why are you not beautiful, closed and decided to be mixed in shape? 
You weren’t easy to stabilize, like other proteins
You certainly got our passion and love to change your clover
To create a beautiful, closed tetramer, giving you a nano makeover
Only changing your inside, filling cavities and other spaces
One of many mysteries might be solved
However, there is still so much more to discover, we need others involved 

Thanks to Dan Ellis, Neil King, Masaru Kanekiyo, and Karla-Luise Herpoldt for creative input on NA-inspired poetry.