Mass photometry: an innovative technology to measure the masses of macromolecules using light
Part 2 of the Bio-protocol Ambassador blog series
Written by Arianna Picozzi
Master’s Degree Candidate, University of Groningen
Intern, LaCava Lab — Laboratory of Macromolecules and Interactomes
European Research Institute for the Biology of Ageing
University Medical Center Groningen
Groningen, The Netherlands
What is mass photometry?
Mass photometry (MP) is an innovative technology that has been recently developed by the Oxford-based company Refeyn© and Professor Kukura’s research group at the University of Oxford. MP relies on two main optical principles, interference reflection microscopy (IRM) [1] and interferometric scattering microscopy (iSCAT) [2], enabling the imaging, detection, and study of macromolecules during their collision and adhesion to a glass coverslip.
In our lab, in-solution, ostensibly native measurements are of immense value. We study endogenous protein complexes extracted from their cellular milieus and transferred into solutions in test tubes. The solution conditions of protein extraction (among other variables) may stabilize or destabilize different forms of target macromolecular complexes, and may induce the accumulation or rejection of non-specific interactions to different degrees — these effects will impact the apparent compositions of macromolecules obtained, for example, by immunoprecipitation — a workhorse technique in our lab. Prior to thorough optimization, immunoprecipitates are almost always composite mixtures of different macromolecules affiliated in varying proportions with the target protein [3–5]. We use several techniques to learn about the affinity enriched fractions we obtain — e.g. SDS-PAGE with protein staining, western blotting, and mass spectrometry (MS) among them. Although these readouts can inform us about the compositions and proportions of the protein constituents of our fractions, they do not readily reveal how many macromolecules those constituents compose in the mixture. How many discrete objects have we enriched?
MP effectively functions as a single-molecule counting machine — revealing the number of macromolecules at each mass present in the sample. Provided that native elution from the antibody can be achieved, MP allows researchers to know how many discrete macromolecules co-immunoprecipitated with the target protein, given the experimental conditions of capture. MP compares favorably to other methods that are commonly used to survey heterogeneity — such as size-exclusion chromatography (SEC) or density gradient sedimentation — MP requires very little sample (usually 1 or 2 µL at nM concentrations) and the readout is delivered in about two minutes.
How does it work?
MP is characterized by the presence of an incident beam of light coming from a polarizer that hits a glass surface (in this case a microscopy coverslip). At this point the light can be reflected by the glass surface or it can be scattered by the presence of an entity in a solution at the interface with the glass; in this example, single proteins or protein complexes colliding with the surface of the coverslip. The instrument detects the change in the refractive index that occurs when the aqueous solution is displaced by the presence of a molecule with higher density: specifically, it precisely measures the interference between the reflected and scattered light [6] (black spots, Figure 1a) when proteins bind the surface. The readout of the instrument is a series of contrast values corresponding to landing events of single entities. These contrast values usually fit a Gaussian curve, whose height is informative of the relative abundance of that particular species in the sample mixture (Figure 1b).
Reproduced from reference [6], link to referenced article
Thus, a fundamental property of MP is that the relative contrast values obtained correspond to the molecular masses of the macromolecules present: the magnitude of light scattering scales and it is proportional to molecular mass. The conversion of contrast values to molecular mass values is achieved with the aid of a standardized calibration curve, usually composed of a variety of proteins spread across a mass range from approximately fifty kilodaltons to a few megadaltons — the calibration against standards enables high accuracy (up to 3% of object mass) and high precision (up to 2% of object mass) measurements. Furthermore, the mass photometer exhibits detection of objects as small as ~30 kDa and a concentration sensitivity ranging from 10 pM up to 100 nM [7].
A major advantage of MP is that it does not require any type of labeling for macromolecule detection, thus avoiding perturbations of proteins’ native structures during measurement. Plus, it allows molecular mass measurements in-solution with a broad reagent compatibility range. Compared to native mass spectrometry, which has limited reagent compatibility, this offers a great advantage — avoiding the removal steps of interfering species and making the technique less time consuming.
On the other hand, mass photometry-based assays also present some limitations: there are some boundaries on reagent compatibility e.g. when high concentrations of a solute alter the refractive index of the solution and limit the resulting contrast detection. The absolute concentration of solute at which these effects become apparent will vary and should be determined empirically; at a recent video conference bringing together MP users, one participant shared observations indicating that they could get good data at 2 M but nothing at 5 M NaCl; the same participant stated that glycerol to 10% (v/v) and DMSO to 20% (v/v) did not affect the apparent mass or number of landings. Refeyn stated potential interest in launching a repository of tested / compatible conditions. Although most experimentalists are concerned with endogenous protein samples being too dilute to measure by their choice techniques, with MP it is possible to have a sample that is too concentrated, with the upper bound approaching ~100 nM; if the concentration of the sample is too high, particles overpopulate the glass coverslip, interfering with the readout signal given.
How are we using it?
In one example, we are exploiting MP to study variability within human RNA exosome complexes [8–9] (Figure 2 and Table 1). The exosome is a multiprotein complex [10] involved in RNA processing. Although numerous compositional variations have been described, often depending upon the cellular compartment of assembly [11–13]. The human exosome is mainly comprised of a nine-protein central barrel, two ribonucleolytic subunits, EXOSC10 and DIS3, and the RNA helicase MTREX; additional components include C1D, MPHOSPH6, and myriad adapter proteins. In a typical human exosome IP targeting 3xFLAG-tagged EXOSC10, we retrieve approximately stoichiometric complexes composed of EXOSC1 through EXOSC10 and MTREX — thus, in our hands, the human exosome complex retrieved has a nominal mass of around 493 kDa. Being the target of the IP, EXOSC10 exhibits somewhat greater abundance than other components (a common occurrence for the target protein). DIS3 is usually substoichiometric or absent unless the conditions of capture are adjusted. C1D and MPHOSPH6 are also often present, but in substoichiometric amounts in typical EXOSC10 IPs.
Through our exploration of different procedures, we have been able to obtain preparations of endogenously assembled, affinity enriched human exosomes possessing or lacking different components — for example +/- DIS3, +/- MTREX, +/- ZCCHC8 — notably, ZCCHC8 and MTREX have also been characterized as members of the exosome adapter ‘NEXT complex’ [4, 8, 14, 15]. We therefore sought a rapid and simple way to assess macromolecular mass heterogeneity in these samples — after all, endogenous is messy! By combining MP with MS based studies of the same fractions, we can explore if our conclusions from each technique adequately match — for example, do the mass-adjusted relative abundances of proteins (or e.g. copy numbers obtained by iBAQ [16, 17]) explain the macromolecule masses observed by MP? Or are there differences and ambiguities? Do we observe minor populations with substoichiometric interactors (a frequent outcome with immunoprecipitates)? Can our MS data combined with MP data permit us to accurately assign the compositions of major and minor macromolecular species without extensive additional fractionation (and associated signal losses)? Can the MP data inform us when global heterogeneity is sufficiently low to move an endogenous macromolecule preparation forward to e.g. electron microscopy-based studies — or when we need to strategize further polishing?
Ultimately we hope to be able to routinely run structural biology assays, chiefly crosslinking MS [18] and electron microscopy [19–21] but also native MS [22] on immunoprecipitated macromolecules derived from their endogenous sources, mitigating the needs for recombinant expression and exogenous reconstitution. We believe sample quality assurance by pre-screening with MP will play a vital role in making such objectives routinely achievable and generally accessible.
Known subunits and molecular masses of human endogenous exosome. *ZCCHC8 is not formally recognized as a bona fide exosome component and is instead considered part of the exosome adapter NEXT complex at present.
Acknowledgements
I would like to thank members of the Laboratory of Macromolecules and Interactomes (ERIBA, Groningen, NL) and Refeyn Ltd. (Oxford, UK) for editing and helpful comments on this text; the Laboratory of Macromolecules and Interactomes and Refeyn Ltd have an R&D collaboration, co-funded from Health Holland. Thanks to Dennis Nanninga for support in graphic design and illustration.
References
[1] Interference reflection microscopy in cell biology: methodology and applications; Verschueren H.; J Cell Sci.; 1985;75: 279–30.
[2] Interferometric Scattering Microscopy; Young, G., Kukura, P.; Annu Rev Phys Chem.; 2019;70:301–322.
[3] Affinity proteomics to study endogenous protein complexes: Pointers, pitfalls, preferences and perspectives; LaCava, J., Molloy, K., Taylor, M., et. al.; BioTechniques; 2015; 58(3), 103–119.
[4] Rapid, optimized interactomic screening; Hakhverdyan, Z., Domanski, M., Hough, L., et. al.; Nature methods; 2015; 12(6), 553–560.
[5] Affinity proteomic dissection of the human nuclear cap-binding-complex interactome; Dou, Y., Kalmykova, S., Pashkova, M., et. al.; 2020; bioRxiv
[6] Quantifying Protein-Protein Interactions by Molecular Counting with Mass Photometry; Soltermann, F., Foley, E., Pagnoni, V., et. al; Angewandte Chemie (International ed. in English); 2020; 59(27), 10774–10779.
[7] Refeyn OneMP Mass Photometer User Manual vers 1.0
[8] Purification and analysis of endogenous human RNA exosome complexes; Domanski, M., Upla, P., Rice, WJ., et al.; RNA; 2016;22(9):1467‐1475.
[9] Affinity Purification of the RNA Degradation Complex, the exosome, from HEK-293 Cells; Domanski, M., LaCava, J.; Bio-protocol; 2017; 7(8), e2238.
[10] Targeting RNA for processing or destruction by the eukaryotic RNA exosome and its cofactors; Zinder, JC., Lima, CD.; Genes Dev. 2017;31(2):88‐100.
[11] Comparison of the yeast and human nuclear exosome complexes; Sloan, KE., Schneider, C., Watkins, NJ.; Biochem Soc Trans; 2012;40(4):850–855.
[12] Dis3-like 1: a novel exoribonuclease associated with the human exosome; Staals, RH., Bronkhorst, AW., Schilders, G., et. al.; The EMBO journal, 2010; 29(14), 2358–2367.
[13] The human core exosome interacts with differentially localized processive RNases: hDIS3 and hDIS3L; Tomecki, R., Kristiansen, MS., Lykke-Andersen, S., et. al.; The EMBO journal; 2010; 29(14), 2342–2357.
[14] Affinity Proteomic Analysis of the Human Exosome and Its Cofactor Complexes; Winczura, K., Domanski, M., LaCava, J.; The Eukaryotic RNA Exosome; 2020; 2062(15), 291–325.
[15] Interaction profiling identifies the human nuclear exosome targeting complex; Lubas, M., Christensen, M., Kristiansen, M., et. al; Molecular cell; 2011; 43(4), 624–637.
[16] Global quantification of mammalian gene expression control; Schwanhaeusser, B., Busse, D., Li, N., et.al; Nature; 2011; 473(7347), 337–342.
[17] Stoichiometry of chromatin-associated protein complexes revealed by label-free quantitative mass spectrometry-based proteomics; Smits, A., Jansen, P., Poser, I., et.al; Nucleic Acids Res; 2013 ; 41(1), e28.
[18] Cross-linking mass spectrometry: methods and applications in structural, molecular and systems biology; O’Reilly, FJ., Rappsilber, J.; Nat Struct Mol Biol; 2018; 25, 1000–1008 (2018).
[19]Structural Analysis of Protein Complexes by Cryo Electron Microscopy; Costa, T., Ignatiou, A., Orlova, EV.; Methods in molecular biology (Clifton, N.J.); 2017; 1615, 377–413.
[20] Detection and Characterization of Extracellular Vesicles by Transmission and Cryo-Transmission Electron Microscopy; Cizmar, P., Yuana, Y.; Methods in molecular biology; 2017; (Clifton, N.J.), 1660, 221–232.
[21] Cryo-electron microscopy — a primer for the non-microscopist; Milne, JL., Borgnia, MJ., Bartesaghi, A., Tran, et.al; The FEBS journal; 2013; 280(1), 28–45.
[22] A robust workflow for native mass spectrometric analysis of affinity-isolated endogenous protein assemblies; Olinares, P. D., Dunn, A. D., Padovan, et. al; Analytical chemistry; 2016; 88(5), 2799–2807.
[23] The RNA exosome and RNA exosome-linked disease; Morton, D. J., Kuiper, E. G., Jones, S. K., Leung, et.al.; RNA (New York, N.Y.); 2018; 24(2), 127–142.
