Scientific investigations rarely adhere to even the most well thought out blueprints. Our study was no exception; however, it may be worthwhile to share some of the experiences we endured, as they offer insight for how one challenging student-mentor relationship grew into a productive, fun, and potentially innovative scientific experience. The upfront disclaimer is that I was the mentor in this story, and while I have attempted to present a neutral recollection, it may be prudent to read over any embellishments from my perspective of the story.
Several old reports, showing that the influenza NA (neuraminidase) protein was less stable than the HA (hemagglutinin) protein provided the foundation for our study. In fact, this property difference among the proteins aided in the discovery that the receptor binding and receptor destroying functions of the influenza virus were not linked to the same protein at the time of the reports1,2. As this difference in properties can potentially impact vaccine development, I decided to invest the time of a new graduate student to setup some assays to identify the stability determinants in the NA protein.
Although my new graduate student was less than enthusiastic about the concept of the project, due to the lack of a direct translational impact, he slogged through several assay iterations until we finally settled on one, a simple thermostability analysis. We chose this particular assay because it takes about 45 minutes to run and only requires three materials: virus, a PCR machine, and access to a plate reader, all of which are crucial when you have a budding young scientist who is learning and cannot afford expensive reagents or machines.
Early in our study, the measurements changed from one day to the next causing the results to be quite inconsistent. This issue at that time really made us rethink the whole project as the variation in results is not ideal when your goal is to create a robust assay for generating highly reproducible data.
At one point, we tested different columns in the 96-well PCR machine because I was convinced either the machine was broken, the wrong gradient program was selected, or it was experimental error. None of these preconceptions were well received by my hard-working student and did little to increase his interest in the project, so the challenge of producing consistent results grew into a large frustration for both mentor and student.
In retrospect, I am extremely grateful he did not give up, but instead, he continued following instructions, and most importantly, he stood behind his data. Soon, he realized that the two NAs he was examining had very large stability variations.
Once my student proved this observation, the follies of my initial evaluation (use of the wrong program or a broken machine) were understandable, but this experience also emphasized the shortcomings of my teaching and thought processes. To remedy this issue, I introduced positivity toward his findings and challenged the results rather than criticized them. As a result, the student-mentor relationship and the motivation for the project vastly improved.
Later, we discovered that the stability of each NA is affected by any subtle change in the solution explaining why the measurements were not as precise as we were hoping for in the beginning. At this stage, the study advanced more rapidly than it had in the first year, and we tracked the NA stability variation down to the presence of calcium in the surrounding environment.
This finding was when we hit the next hurdle in our collaboration (translation: graduate student loses faith in project and mentor and became disenchanted), as calcium contributions to NA stability had already been observed and had been referenced in some reviews1-8.
In our lab, we often describe the occurrence of repeated observations as rediscovering fire or reinventing the wheel, which is not a good thing in a world where two words ‘not novel’ can kill any study. Our solution to this problem was to identify every possible question surrounding the calcium observation and to determine if any had been answered. Amazingly, we found little or no data for most, if not all, of our questions, which provided new motivation to set out to answer them one by one.
In the following two years, I had the most fun thus far in my mentoring career as I was watching a graduate student transform into a scientist and the added bonus was that his results provided a link to several of our previous studies9-11.
A cliché in science is that we stand on the shoulders of the ones who came before us; however, I think it may be equally appropriate to say we (mentors) stand on the shoulders of those who came after us (students and post-docs). Yes, it is true, teaching students requires a significant investment of time and energy, which is seemingly unrewarding in the beginning; however, seeing the transformations take place, as occurred in this collaborative study with my student, make all the energy spent worthwhile and it is especially rewarding in terms of generating new ideas and hypothesis.
For more in-depth scientific details, we suggest reading the paper12. It includes a vast array of findings related to NA function and evolution, many of which were developed and tested by my student. You can find the published work here:
As a teaser, I will highlight that his study is a bit counter intuitive as it demonstrates that the unique structural restrictions for NA function actually provide the enzyme with more adaptive flexibility from both a homogeneous and heterogeneous population. In addition, his results suggest that the delicate balance between the functions of NA and HA, viral movement and viral attachment, respectively, may be more dynamic where NA function changes with the environment and time.
Regarding the translational implications of the study, we are currently attempting to improve the potency of NA antigens in vaccines and to develop better methods for assessing protective NA responses by applying several of the principle discoveries in his investigation. If there is one thing I regret in the end; it is that I have moved my lab and my student who is currently writing up his thesis at my old university is not here to see the applications of his work.
1. Francis, T. Dissociation of Hemagglutinating and Antibody-Measuring Capacities of Influenza Virus. J Exp Med 85, 1-7 (1947).
2. Briody, D.A. Characterization of the enzymic action of influenza viruses on human red cells. J Immunol 59, 115-27 (1948).
3. Carroll, S.M. & Paulson, J.C. Complete metal ion requirement of influenza virus N1 neuraminidases. Brief report. Arch Virol 71, 273-7 (1982).
4. Burmeister, W.P., Cusack, S. & Ruigrok, R.W. Calcium is needed for the thermostability of influenza B virus neuraminidase. J Gen Virol 75 ( Pt 2), 381-8 (1994).
5. Sultana, I. et al. Stability of neuraminidase in inactivated influenza vaccines. Vaccine 32, 2225-30 (2014).
6. Baker, N.J. & Gandhi, S.S. Effect of Ca++ on the stability of influenza virus neuraminidase. Arch Virol 52, 7-18 (1976).
7. Dimmock, N.J. Dependence of the activity of an influenza virus neuraminidase upon Ca2+. J Gen Virol 13, 481-3 (1971).
8. Chong, A.K., Pegg, M.S. & von Itzstein, M. Influenza virus sialidase: effect of calcium on steady-state kinetic parameters. Biochim Biophys Acta 1077, 65-71 (1991).
9. da Silva, D.V., Nordholm, J., Madjo, U., Pfeiffer, A. & Daniels, R. Assembly of subtype 1 influenza neuraminidase is driven by both the transmembrane and head domains. J Biol Chem 288, 644-53 (2013).
10. Nordholm, J., da Silva, D.V., Damjanovic, J., Dou, D. & Daniels, R. Polar residues and their positional context dictate the transmembrane domain interactions of influenza a neuraminidases. J Biol Chem 288, 10652-60 (2013).
11. da Silva, D.V. et al. The influenza virus neuraminidase protein transmembrane and head domains have coevolved. J Virol 89, 1094-104 (2015).
12. Wang, H., Dou, D., Östbye, H., Revol, R. & Daniels, R. Structural restrictions for influenza neuraminidase activity promote adaptation and diversification. Nature Microbiology (2019).