Macromolecular crowding in cells influences processes such as folding, association and diffusion of proteins and polynucleic acids. Direct spatiotemporal readout of crowding would be a powerful approach for unraveling the structure of the cytoplasm and determining the impact of excluded volume on protein function in living cells. Here, we introduce a genetically encodable fluorescence resonance energy transfer (FRET) sensor for quantifying macromolecular crowding and discuss our application of the sensor in bacterial and mammalian cells.
Subscribe to Journal
Get full journal access for 1 year
only $20.17 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Rent or Buy article
Get time limited or full article access on ReadCube.
All prices are NET prices.
Zhou, H.X., Rivas, G. & Minton, A.P. Annu. Rev. Biophys. 37, 375–397 (2008).
Fulton, A.B. Cell 30, 345–347 (1982).
Cayley, S., Lewis, B.A., Guttman, H.J. & Record, M.T. Jr. J. Mol. Biol. 222, 281–300 (1991).
Weiss, M., Elsner, M., Kartberg, F. & Nilsson, T. Biophys. J. 87, 3518–3524 (2004).
Konopka, M.C., Shkel, I.A., Cayley, S., Record, M.T. & Weisshaar, J.C. J. Bacteriol. 188, 6115–6123 (2006).
Parry, B.R. et al. Cell 156, 183–194 (2014).
Dix, J.A. & Verkman, A.S. Annu. Rev. Biophys. 37, 247–263 (2008).
van den Bogaart, G., Hermans, N., Krasnikov, V. & Poolman, B. Mol. Microbiol. 64, 858–871 (2007).
Bermejo, C., Haerizadeh, F., Takanaga, H., Chermak, D. & Frommer, W.B. Nat. Protoc. 6, 1806–1817 (2011).
Guo, J., Sachs, F. & Meng, F.J. Antioxid. Redox Signal. 20, 986–999 (2014).
Le Coeur, C., Teixeira, J., Busch, P. & Longeville, S. Phys. Rev. E 81, 061914 (2010).
Mikaelsson, T., Aden, J., Johansson, L.B. & Wittung-Stafshede, P. Biophys. J. 104, 694–704 (2013).
Golynskiy, M.V., Rurup, W.F. & Merkx, M. Chembiochem 11, 2264–2267 (2010).
Zacharias, D.A., Violin, J.D., Newton, A.C. & Tsien, R.Y. Science 296, 913–916 (2002).
Venturoli, D. & Rippe, B. Am. J. Physiol. Renal Physiol. 288, F605–F613 (2005).
Lavrenko, P.N., Mikriukova, O.I. & Okatova, O.V. Anal. Biochem. 166, 287–297 (1987).
Konopka, M.C., Weisshaar, J.C. & Record, M.T. Jr. Methods Enzymol. 428, 487–504 (2007).
Minton, A.P. Biopolymers 99, 239–244 (2013).
Benton, L.A., Smith, A.E., Young, G.B. & Pielak, G.J. Biochemistry 51, 9773–9775 (2012).
Wang, Y., Sarkar, M., Smith, A.E., Krois, A.S. & Pielak, G.J. J. Am. Chem. Soc. 134, 16614–16618 (2012).
Neidhardt, F.C., Bloch, P.L. & Smith, D.F. J. Bacteriol. 119, 736–747 (1974).
This work was supported by the Netherlands Organization for Scientific Research (NWO Veni and ChemThem) and the Dutch Technology Foundation STW, which is part of the Netherlands Organization for Scientific Research (NWO) and is partly funded by the Ministry of Economic Affairs. Part of the work was performed at the University Medical Center Groningen Microscopy and Imaging Center (UMIC), which is sponsored by NWO grant 175-010-2009-023.
The authors declare no competing financial interests.
Integrated supplementary information
Conditions: 10 mM NaPi, 100 mM NaCl, 2 mg/ml BSA, pH 7.4. Starting concentration is on the order of ~10 nM (1000-times dilution from a stock solution, containing 10 µM sensor determined with the Nanodrop ND-1000 spectrophotometer, using ɛ280 = 54,000 M–1 cm–1).
Supplementary Figure 2 Example of the emission spectrum upon excitation of mCitrine at 515 nm at different concentrations of Ficoll PM70 (% w/w).
No difference between the spectra is observed, other than the experimental variation. Conditions: 10 mM NaPi, 100 mM NaCl, 2 mg/ml BSA, pH 7.4.
Proteinase K cleaves flexible linker regions, and leaves β-barrels intact. The linker was cleaved by addition of 2 µL enzyme (5 mg/ml stock in H2O) to 18 µl sensor solution (in 10 mM NaPi, 100 mM NaCl, pH 7.4). After 1 minute 2 µl PMSF (100 mM in isopropanol) was added. (a) Fluorescence emission spectrum after mCerulean excitation (420 nm), before and after Proteinase K treatment. The mCitrine emission is absent after linker cleavage. (b) Fluorescence emission spectrum upon mCitrine excitation (515 nm). The mCitrine concentration does not decrease, verifying that the fluorophores are intact. (c) Effect of added Ficoll on the mCitrine/mCerulean ratio. Without proteinase K treatment the ratio increases, whereas the cleaved sensor does not show a FRET effect, excluding aggregation phenomena. (d) SDS-PAGE (10% polyacrylamide) analysis of the cleavage reaction, showing single bands before and after cleavage. The band after cleavage is less sharp, possibly due to the different cleavage sites of proteinase K in the linker. In-gel fluorescence was recorded using excitation at 460 nm, filter Y515.
(a) Scanning confocal microscopy images of the crowding sensor in 40% w/w PEG showing coacervate formation.1,2 Conditions: 10 mM NaPi, 100 mM NaCl, pH 7.4. Image dimensions: 32 x 32 µm. The fluorescent spots are not due to aggregation, because YFP excitation did not indicate self-quenching (Supplementary Figure 2). The crowding sensor partitions in the coacervates, which are highly crowded with PEG and induce FRET ratios (YFP/CFP) of >20. (b) Emission spectrum upon excitation of mCitrine at 515 nm at different % w/w PEG 6000. No difference between the spectra is observed, other than the experimental variation. Conditions: 10 mM NaPi, pH 7.4. NaCl was omitted to prevent coacervate formation.
Supplementary Figure 5 The effect of crowding induced by bovine serum albumin (BSA) on the sensor readout.
(a) Emission spectrum upon excitation of mCerulean at 420 nm at different % w/w BSA. The relative contribution of mCitrine increases according to an increase in crowding, but the overall fluorescence decreases. The decrease is most likely due to an inner filter effect, because the BSA solution is strongly absorbing at 420 nm. Conditions: 10 mM NaPi, 100 mM NaCl, pH 7.4. (b) Emission spectrum upon excitation of mCitrine at 515 nm at different % w/w BSA. An approximately linear decrease in intensity is observed upon increasing the concentration of BSA. The decrease is likely due to the inner filter effect, because the BSA solution absorbs at 515 nm (most likely due to traces of myoglobin). Conditions: 10 mM NaPi, 100 mM NaCl, pH 7.4. (c) The FRET ratio, calculated from Supplementary Figure 5a, plotted versus the concentration of BSA. The FRET ratio increases with increasing concentration of BSA. The error bars are standard deviations from three independent experiments.
Conditions: 10 mM NaPi, pH 7.4. NaCl was omitted to prevent coacervate formation. The hydrodynamic radii1,2 were used to calculate the total PEG volume for 0.1 g PEG (Fig. 1d). The error bars are standard deviations from three independent experiments.
(a) Effect of NaCl on the sensor readout. NaCl was titrated from a 5M stock solution into a 1.0 ml solution of the sensor with the corresponding amount of Ficoll, and 10 mM NaPi, 2 mg/mL BSA, pH 7.4. The FRET ratio does not change due to NaCl, irrespective of the Ficoll concentration, demonstrating that electrostatic effects do not influence the read-out. A very small decrease can be observed at higher salt concentrations as more stock solution of salt is added, diluting the Ficoll, which in turn results in less crowding. (b) Effect of potassium glutamate on the sensor readout. Titration up to 500 mM glutamate from a 5M stock solution does not alter the readout of the sensor, demonstrating that electrostatic effects do not influence the readout. Conditions: 10 mM NaPi, 2 mg/ml BSA, pH 7.4. Addition of glutamate did not alter the pH significantly. (c) Effect of glycine betaine on the sensor read-out. Titration of up to 1M glycine betaine does not alter the readout of the sensor. Conditions: 10 mM NaPi, 2 mg/ml BSA, pH 7.4. (d) Effect of DNA on the sensor readout. DNA does not have an effect on the sensor. Conditions: 10 mg/ml salmon testes DNA, 10 mM NaPi, 100 mM NaCl, 0.1 mg/ml BSA, pH 7.4.
(a) Effect of glycerol on the sensor readout. A solution of 10 mM NaPi, pH 7.4 was mixed with a 85% (v/v) solution of glycerol to obtain the desired concentration of glycerol. Glycerol did not affect the readout, showing that the sensor does not report viscosity changes. (b) Effect of temperature on the sensor readout with and without Ficoll. Conditions: 10 mM NaPi, 100 mM NaCl, 2 mg/ml BSA, pH 7.4. A small temperature dependence is observed both in the presence and absence of Ficoll. (c) Effect of pH on the sensor readout. Conditions: 10 mM NaPi. Below pH 7 the FRET ratio drops, indicating measurements should be performed at pH >7. The drop in FRET ratio at pH <7 is likely due to protonation of the glutamates in the α-helices.
Supplementary Figure 9 Examples of histograms of the FRET ratio distributions of E. coli after different osmotic upshifts.
Data correspond to Fig. 2a,b. The distributions fit to a Gaussian.
Supplementary Figure 10 Comparison of the FRET ratios of E. coli after osmotic upshift by sucrose or NaCl.
Osmotic upshift induced by introducing 0.5 M NaCl or 1.0 M sucrose to MOPS minimal medium in the absence of K2HPO4 and glucose (n = 40 - 50 cells). The error bars reflect the error in the maximum of the Gaussian fit of the histogram generated from the measured FRET ratios.
Supplementary Figure 11 Box plot depicting the reversibility of the sensor readout in E. coli (n = 50–100 cells).
Resuspension of the cells in medium of higher osmolarity increases the mCitrine/mCerulean ratio. Subsequent centrifugation of the cells and resuspension in medium of lower osmolarity leads to the same ratios as before the osmotic upshift, showing reversibility of the sensor within the accuracy of the measurement.
This calibration applies to the microscope settings employed for the E. coli experiments. Each data point is an average over 9 measurements; the error bars represent the standard deviations. The data fit mCitrine/mCerulean = 7.08 × 10-3 × [% w/w Ficoll] + 0.910, with R2 = 0.99. Conditions: 10 mM NaPi, 100 mM NaCl, 2 mg/ml BSA, pH 7.4.
Supplementary Figure 13 SDS-PAGE (10% polyacrylamide) analysis of the crowding sensor expressed in HEK293 cells.
The crowding sensor was expressed using the protocol described above. After expression the medium was replaced with cold PBS buffer. The cells were scraped and collected by centrifugation. After resolubilization in sample buffer and 3 min of incubation at room temperature, the mixture was loaded on a 10% SDS-PAA gel. The gel was analyzed by in-gel fluorescence (right) and Coomassie staining (left). Only full-length sensor was obtained.
Supplementary Figure 14 Illustrative example of the mCitrine intensity plotted versus the mCerulean intensity in HEK293 cells expressing the sensor.
Determination of the slope yields the mCitrine/mCerulean ratio. Crucially, the linearity shows that the ratio is independent of sensor concentration. The R2 values are generally higher than 0.99.
Supplementary Figure 15 Comparison of the increase in the percent Ficoll equivalent in E. coli calculated from the volume decrease with the value determined using the crowding sensor.
The volume decrease was estimated from the length and width of the bacteria before and after the osmotic upshift and assuming a cylindrical shape of the cells. The % Ficoll equivalents after addition of 125 mM NaCl was subsequently calculated by multiplying the relative volume change with the % Ficoll equivalents before the addition of salt. The % Ficoll equivalents before addition of salt was determined using the crowding sensor and the calibration in Supplementary Figure 12. Median values of 5 independent experiments ± standard deviations are depicted.
About this article
Cite this article
Boersma, A., Zuhorn, I. & Poolman, B. A sensor for quantification of macromolecular crowding in living cells. Nat Methods 12, 227–229 (2015) doi:10.1038/nmeth.3257
Impact of Molecular Crowding on Translational Mobility and Conformational Properties of Biological Macromolecules
The Journal of Physical Chemistry B (2019)
Molecular Biology of the Cell (2019)
Chemistry – A European Journal (2019)
Crowding Effects on Energy-Transfer Efficiencies of Hetero-FRET Probes As Measured Using Time-Resolved Fluorescence Anisotropy
The Journal of Physical Chemistry B (2019)