Phase transitions of multivalent proteins can promote clustering of membrane receptors
- Howard Hughes Medical Institute, University of Texas Southwestern Medical Center, United States
Figures
Figure 1 with 1 supplement
Reconstitution of p-Nephrin clusters on supported lipid bilayers.
(A) Cartoon illustrating the interaction of triply-phosphorylated His8-tagged Nephrin (p-Nephrin) with its partners Nck and N-WASP. Top panel illustrates p-Nephrin attached to bilayers. Bottom panel illustrates the model for clustered p-Nephrin, upon Nck and N-WASP binding. (B) Top: TIRF image of Alexa488-labeled p-Nephrin attached to a supported DOPC lipid bilayer doped with 1% nickel-chelating lipid (Ni2+-NTA DOGS), (corresponding to panel A, top). Bottom: TIRF image of analogous membrane-attached Alexa488-labeled p-Nephrin after addition of 1 µM Nck and 1 µM N-WASP (corresponding to panel A, bottom). (C) Line-scans of the images in panel B, at the positions depicted by the white dotted lines.
Figure 1—figure supplement 1
p-Nephrin, Nck and N-WASP colocalize to clusters formed on fluid supported lipid bilayers.
(A) Fluorescence recovery after photobleaching (FRAP) on a supported bilayer with p-Nephrin shows full recovery with exponential recovery time constant τ = 1.3 s. Line shows fit to a single-exponential. (B) Clusters do not form with only p-Nephrin on the membrane (left-panel, p-Nephrin Alexa488) or with p-Nephrin (Alexa488) + 1 μM Nck (Alexa568) (middle and right panels). The legend below each panel indicates the fluorophore imaged. (C) Three color imaging shows that p-Nephrin Alexa488, Nck Alexa568, and N-WASP Alexa647 co-localize at the clusters.
Figure 2 with 2 supplements
Nephrin clusters are created via a two-dimensional phase-transition.
(A) Fractional intensity in clusters (blue symbols, left ordinate) and signal variance (red symbols, right ordinate) of p-Nephrin fluorescence on a DOPC bilayer as a function of Nck concentration for 500 nM N-WASP and total p-Nephrin density of ∼2700 molecules/µm2. (B) Relative frequency with which a given number of clusters are found within 93 randomly selected 56 × 56 µm regions of a bilayer formed using ∼2500 molecules/µm2 Alexa488-labeled p-Nephrin, 1 µM Nck, and 1 µM N-WASP. (C) Size distribution of clusters formed using ∼2500 molecules/µm2 Alexa488-labeled p-Nephrin, 1 µM Nck, and 1 µM N-WASP. (D) Puncta formed using 1 µM Nck, 1 µM N-WASP, and low (left) or 4.7-fold higher (right) density of p-Nephrin. Images were autocontrasted for clarity.
Figure 2—figure supplement 1
Quantitative analysis of the measurement and control of His8-p-Nephrin density on supported lipid bilayers.
(A) Fluorescence intensity as a function of fluorescent lipid (OG-DHPE) concentration for a solution of small unilamellar vesicles. Fluorescence of liposomes (in solution) containing OG-DHPE were measured at the indicated concentrations of the OG-DHPE concentrations in the x-axis. (B) Fluorescence intensity as a function of p-Nephrin Alexa488 concentrations. Blue points represent data for protein alone, red points represent data for p-Nephrin in the presence of 9.5 μM of Ni2+-NTA DOGS. Concentrations of the protein for this standard plot were kept similar to those of the concentration of OG-DHPE in panel A. (C) Fluorescence intensity as a function of OG-DHPE density on supported lipid bilayers. Upper and lower x-axis labels list density as percent total lipid and molecules/µm2, respectively. Lines in (A–C) represent a linear fits. (D) Time course of His8-tagged p-Nephrin Alexa488 dissociation from supported lipid bilayers, monitored by TIRFM, following washes that left 2.8 nM protein in solution above the bilayer. (E) Fluorescence intensity of bilayers containing different percentages of p-Nephrin Alexa488 (with total p-Nephrin density ∼2000 molecules/µm2). The data suggest linearity up to ∼60% labeling.
Figure 2—figure supplement 2
Quantification of average p-Nephrin density on the bilayer for every titration point shown in Figure 2A.
Y-axis represents p-Nephrin density and x-axis represents the different Nck concentrations of the titration as in Figure 2A. Densities are averages of five different areas of each bilayer. Error bars representing standard deviations are smaller than the symbols.
Figure 3 with 3 supplements
Clusters are dynamic.
(A) Time-lapse TIRF imaging of bilayers containing ∼3100 molecules/µm2 Alexa488-labeled p-Nephrin after addition of 1 µM Nck and 1 µM N-WASP. Images represent time intervals of 2 min and show coalescence of clusters into larger structures. (B) Fluorescence recovery after photobleaching of Nck and N-WASP in clustered regions (left panel, red and blue, respectively) and p-Nephrin in clustered and unclustered regions (right panel, green and black, respectively). FRAP experiments were performed in separate experiments using Alexa-488 labeled p-Nephrin, Nck, or N-WASP. Lines show bi-exponential fits of the data, except for unclustered p-Nephrin, which was fit using a single-exponential. Bars represent standard deviation from three FRAP experiments on a single bilayer. Bottom table lists the parameters obtained from the fitting.
Figure 3—figure supplement 1
Cluster size-distribution analyses at different times suggest exponential behavior at lower densities.
(A) Log-linear plot of cluster number vs size at p-Nephrin density of ∼2500 μm2, 1 μM Nck, and 1 μM N-WASP. The distributions are plotted for times between 2 and 20 min. (B) Log–log plot of the same data. Lines in (A and B) represent the best linear fits of the data. The better fits in (A) than (B) indicate that the data are better described by exponential than power law functions. In other experiments (not shown), the sizes remain exponentially distributed to times as long as 60 min.
Figure 3—figure supplement 2
Cluster size-distribution analyses suggest power law behavior at higher densities.
(A) Log-linear plot of the size distribution of clusters at a Nephrin density of ∼4000 μm2, 1 μM Nck, and 1 μM N-WASP, recorded 60 min after clustering was initiated. (B) Log–log plot of the size distribution of clusters at a Nephrin density of ∼4000 μm2, 1 μM Nck, and 1 μM N-WASP. Lines in (A and B) represent the best linear fits of the data. The better fits in (B) than (A) indicate that the data are better described by power law than exponential functions (contrast with Figure 3—figure supplement 1).
Figure 3—figure supplement 3
Average cluster size is dependent on molecular density.
Samples contained either low (∼2000 molecules/µm2, red dots) or high (∼3500 molecules/µm2, blue dots) density of p-Nephrin. Clustering was initiated by addition of 1 μM Nck and 1 μM N-WASP. Images were taken every minute. (A) Average cluster size, (B) the fractional intensity in clusters.
Figure 4 with 1 supplement
Clustering is dependent upon the valency of the interacting motifs.
Plots show fractional intensity of fluorescent Nephrin proteins in clusters as a function of Nck protein concentrations for 500 nM N-WASP. (A) Top, middle, and bottom panels show data for p-Nephrin 3pY, 2pY, and 1pY, respectively. For these concentrations of N-WASP and Nck, only Nephrin 3pY shows clustering. At 2 µM N-WASP, Nephrin 2pY also clusters when Nck is added (Figure 4—figure supplement 1). (B) Top, middle, and bottom panels show data for p-Nephrin plus engineered Nck proteins containing 3, 2, or 1 repeat of the second SH3 domain of Nck. For these concentrations/densities of N-WASP/p-Nephrin, only the (SH3)3 protein can induce clustering. At 5 µM N-WASP, (SH3)3 also induces clustering (Figure 4—figure supplement 1). Note that the x-axis is Nck concentration in panel A but total SH3 domain concentration in panel B.
Figure 4—figure supplement 1
Di-valent molecules are stronger clustering agents than mono-valent molecules.
(A) In the presence of 5 µM Nck and 2 μM N-WASP, p-Nephrin 2pY forms clusters (right panel). Even in the presence of 10 µM Nck and 5 μM N-WASP, p-Nephrin 1pY does not form clusters (left panel). Bottom: line-scan of images in the top panels, with locations indicated by dotted lines. (B) At 5 μM N-WASP, 2.5 μM (SH3)2 (5 μM SH3 module concentration) produces p-Nephrin clusters (right panel), whereas (SH3)1 does not (left panel). Bottom: line-scan of images in the top panels, with locations indicated by dotted lines.
Figure 5 with 1 supplement
Molecular affinities affect macroscopic clustering.
(A) Fractional intensity of fluorescent pTyr proteins in clusters as a function of SH3 (module) concentrations for 500 nM N-WASP. Left and right panels show data for a p-TIR and p-Nephrin, whose pTyr motifs bind the SH2 domain of Nck with KD values of 40 nM and 370 nM, respectively. (B) Fluorescence recovery after photobleaching (FRAP) for Alexa488-labeled Nck in p-Nephrin clusters (blue) and p-TIR clusters (red). Nck recovers more slowly (larger τ), can be bleached more strongly (Y-intercept) and recovers to a lower value (plateau) with p-TIR than with p-Nephrin, all indicating slower dynamics in clusters with the higher affinity SH2 binding partner. The bars represent standard deviation from three FRAP experiments on a single bilayer.
Figure 5—figure supplement 1
Measurement of the affinity of Nck for p-TIR and p-Nephrin.
Isothermal titration calorimetry analysis of Nck binding to p-Nephrin and p-TIR. Nck (150 µM) in the syringe was titrated into 5 μM of either (A) p-Nephrin (3pY) or (B) p-TIR (3pY). Both datasets could be fit well to a three-site binding model with a single affinity for Nck. In the p-Nephrin and p-TIR titrations, ∼6% and ∼11% of the proteins, respectively, were found to be incompetent to bind Nck.
Figure 6
Mono-valent pTyr peptide can eliminate clusters.
(A) Time course following addition of 10 µM of a monovalent pTyr peptide derived from TIR (with KD of 40 nM for the Nck SH2 domain) to clusters formed from p-Nephrin /(SH3)3/N-WASP. (B) Time course of the fractional p-Nephrin intensity in clusters after addition of the TIR peptide. (C) Equilibrium fractional intensity of the p-Nephrin clusters as a function of p-TIR peptide concentration, performed in the presence of 1 µM (SH3)3 and 500 nM N-WASP.
Figure 7 with 2 supplements
Actin assembles specifically on p-Nephrin/Nck/N-WASP clusters.
(A) Alexa488-labeled p-Nephrin (2200 molecules/µm2) was clustered by addition of 2 μM N-WASP and 1 μM Nck. Images show time course of p-Nephrin (top row), actin (middle row) and merge (bottom row) after addition of 10 nM Arp2/3 complex and 1 µM actin (10% rhodamine labeled). (B) Half-times of actin assembly as a function of surface area (left-panel) and p-Nephrin intensity (right-panel) in individual clusters. Half-times were calculated using the data for the first 27 min of the time-lapse. (C) Fluorescence of rhodamine-actin on individual clusters as a function of time for 20 representative clusters. Individual curves represent average intensity across an individual cluster.
Figure 7—figure supplement 1
Actin localizes to and assembles on the clusters in an Arp2/3 dependent manner.
(A) Images of clusters formed by p-Nephrin (2173 molecules/μm2 on the supported lipid bilayer) plus soluble 1 μM Nck, 2 μM N-WASP, 10 nM Arp2/3 complex, and 1 μM actin (10% rhodamine labeled) at 0 and 3 min. Top panels show p-Nephrin (Alexa 488 labeled), bottom panels show fluorescence for rhodamine-actin. Note that the actin images are contrast-enhanced relative to those in Figure 7A to illustrate weak, but relatively uniform actin recruitment to the p-Nephrin clusters at 3 min. (B) Actin assembly reactions as in panel A, except that of the lower row lacks the Arp2/3 complex, imaged at 105 min after actin addition. Left, middle, and right panels show p-Nephrin Alexa488, rhodamine-actin, and phalloidin 647 staining, respectively. In the bottom row phalloidin stains only the morphologically elongated structures, suggesting that the actin filaments are formed only in the re-shaped structures on the membrane.
Figure 7—figure supplement 2
Actin assembly reorganizes p-Nephrin clusters.
Enlarged actin assembly images from Figure 7, including more time points between 36 and 45 min. Top, middle, and bottom rows show p-Nephrin Alexa488, rhodamine-actin and merge, respectively.
Videos
Video 1
Addition of Nck and N-WASP to p-Nephrin produces macroscopic clusters on supported bilayers.
Time-lapse images taken immediately after adding 1 μM Nck and 1 μM N-WASP to p-Nephrin Alexa488. Images were captured every minute.
Video 2
Clusters are dynamic.
Time-lapse of clusters made from 1 μM Nck and 1 μM N-WASP with p-Nephrin Alexa488 on the membrane. Images were captured every 30 s. In addition to fusion events, the clusters also occasionally undergo fission.
Video 3
Mono-valent peptide dissolves clusters.
Addition of 10 μM 1pY—TIR causes the clusters of 1 μM (SH3)3 and 500 nM N-WASP to dissipate. Images were captured every 30 s.
Tables
Table 1
Information on the protein constructs used in this study
| Proteins | Sequence information | Notes |
|---|---|---|
| Nck | GHMAEEVVVVAKFDYVAQQEQELDIKKNERLWLLDDSKSWWRVRNSMNKTGFVPSNYVERKNSARKASIVKNLKDTLGIGKVKRKPSVPDSASPADDSFVDPGERLYDLNMPAYVKFNYMAEREDELSLIKGTKVIVMEKCSDGWWRGSYNGQVGWFPSNYVTEEGDSPLGDHVGSLSEKLAAVVNNLNTGQVLHVVQALYPFSSSNDEELNFEKGDVMDVIEKPENDPEWWKCRKINGMVGLVPKNYVTVMQNNPLTSGLEPSPPQCDYIRPSLTGKFAGNPWYYGKVTRHQAEMALNERGHEGDFLIRDSESSPNDFSVSLKAQGKNKHFKVQLKETVYCIGQRKFSTMEELVEHYKKAPIFTSEQGEKLYLVKHLS | Human, WT, residues 1–377 |
| N-WASP BPVCA | GSEFKEKKKGKAKKKRAPPPPPPSRGGPPPPPPPPHSSGPPPPPARGRGAPPPPPSRAPTAAPPPPPPSRPGVVVPPPPPNRMYPHPPPALPSSAPSGPPPPPPLSMAGSTAPPPPPPPPPPPGPPPPPGLPSDGDHQVPASSGNKAALLDQIREGAQLKKVEQNSRPVSCSGRDALLDQIRQGIQLKSVSDGQESTPPTPAPTSGIVGALMEVMQKRSKAIHSSDEDEDDDDEEDFEDDDEWED | Rat, residues 183–193 fused to 273–501 |
| Nck (cysteine-modified) | GHMCMAEEVVVVAKFDYVAQQEQELDIKKNERLWLLDDSKSWWRVRNSMNKTGFVPSNYVERKNSARKASIVKNLKDTLGIGKVKRKPSVPDSASPADDSFVDPGERLYDLNMPAYVKFNYMAEREDELSLIKGTKVIVMEKSSDGWWRGSYNGQVGWFPSNYVTEEGDSPLGDHVGSLSEKLAAVVNNLNTGQVLHVVQALYPFSSSNDEELNFEKGDVMDVIEKPENDPEWWKARKINGMVGLVPKNYVTVMQNNPLTSGLEPSPPQSDYIRPSLTGKFAGNPWYYGKVTRHQAEMALNERGHEGDFLIRDSESSPNDFSVSLKAQGKNKHFKVQLKETVYSIGQRKFSTMEELVEHYKKAPIFTSEQGEKLYLVKHLS | Human, residues 1–377, with mutations: C139S, C232A, C266S, C340S |
| Nephrin3Y | GGSLEHHHHHHHHGGSCGGSGGSGGSGGSHLYDEVERTFPPSGAWGPLYDEVQMGPWDLHWPEDTFQDPRGIYDQVAGD | Human, residues 1174–1223, with mutations: Y1183F, Y1210F |
| Nephrin2Y | GGSLEHHHHHHHHGGSCGGSGGSGGSGGSHLFDEVERTFPPSGAWGPLYDEVQMGPWDLHWPEDTFQDPRGIYDQVAGD | Human, residues 1174–1223, with mutations: Y1176F, Y1183F, Y1210F |
| Nephrin1Y | GGSLEHHHHHHHHGGSCGGSGGSGGSGGSHLFDEVERTFPPSGAWGPLYDEVQMGPWDLHWPEDTFQDPRGIFDQVAGD | Human, residues 1174–1223, with mutations: Y1176F, Y1183F, Y1210F, Y1217F |
| TIR3Y | GGSLEHHHHHHHHGGSCGGSGGSGGSGGSHMHIYDEVAADPPPSGAWGHIYDEVAADPWDLHWPEDTFQDPRHIYDEVAADP | Human Nephrin, with pTyr sites replaced by those in EPEC Tir protein (underlined) |
| (SH3)3 | GHMPAYVKFNYMAEREDELSLIKGTKVIVMEKSSDGWWRGSYNGQVGWFPSNYVTEEGDSPLSARKASIVKNLKDTLGIGKVKRKPSVPDSASPADDSFVDPGERLYDLNMPAYVKFNYMAEREDELSLIKGTKVIVMEKSSDGWWRGSYNGQVGWFPSNYVTEEGDSPLSARKASIVKNLKDTLGIGKVKRKPSVPDSASPADDSFVDPGERLYDLNMPAYVKFNYMAEREDELSLIKGTKVIVMEKSSDGWWRGSYNGQVGWFPSNYVTEEGDSPLNNPLTSGLEPSPPQCDYIRPSLTGKFAGNPWYYGKVTRHQAEMALNERGHEGDFLIRDSESSPNDFSVSLKAQGKNKHFKVQLKETVYCIGQRKFSTMEELVEHYKKAPIFTSEQGEKLYLVKHLS | Human, three repeats of the second Nck SH3 domain, plus the Nck SH2 domain |
| (SH3)2 | GHMPAYVKFNYMAEREDELSLIKGTKVIVMEKSSDGWWRGSYNGQVGWFPSNYVTEEGDSPLSARKASIVKNLKDTLGIGKVKRKPSVPDSASPADDSFVDPGERLYDLNMPAYVKFNYMAEREDELSLIKGTKVIVMEKSSDGWWRGSYNGQVGWFPSNYVTEEGDSPLNNPLTSGLEPSPPQCDYIRPSLTGKFAGNPWYYGKVTRHQAEMALNERGHEGDFLIRDSESSPNDFSVSLKAQGKNKHFKVQLKETVYCIGQRKFSTMEELVEHYKKAPIFTSEQGEKLYLVKHLS | Human, two repeats of the second Nck SH3 domain, plus the Nck SH2 domain |
| (SH3)1 | GHMPAYVKFNYMAEREDELSLIKGTKVIVMEKSSDGWWRGSYNGQVGWFPSNYVTEEGDSPLNNPLTSGLEPSPPQCDYIRPSLTGKFAGNPWYYGKVTRHQAEMALNERGHEGDFLIRDSESSPNDFSVSLKAQGKNKHFKVQLKETVYCIGQRKFSTMEELVEHYKKAPIFTSEQGEKLYLVKHLS | Human, one repeat of the second Nck SH3 domain, plus the Nck SH2 domain |
| TIR-1pY | EEHIpYDEVAADPGGSWGGSC | N-terminal rhodamine labeled single pTyr motif from EPEC Tir protein |
| Lck | ANSLEPEPWFFKNLSRKDAERQLLAPGNTHGSFLIRESESTAGSFSLSVRDFDQNQGEVVKHYKIRNLDNGGFYISPRITFPGLHDLVRHYTNASDGLCTKLSRPCQTQKPQKPWWEDEWEVPRETLKLVERLGAGQFGEVWMGYYNGHTKVAVKSLKQGSMSPDAFLAEANLMKQLQHPRLVRLYAVVTQEPIYIITEYMENGSLVDFLKTPSGIKLNVNKLLDMAAQIAEGMAFIEEQNYIHRDLRAANILVSDTLSCKIADFGLARLIEDNEYTAREGAKFPIKWTAPEAINYGTFTIKSDVWSFGILLTEIVTHGRIPYPGMTNPEVIQNLERGYRMVRPDNCPEELYHLMMLCWKERPEDRPTFDYLRSVLDDFFTATEGQFQPQP | Human, 119–509, Y505F |
Table 2
Statistics of fitting for FRAP data
| p-Nephrin | Nck (with p-Nephrin) | Nck (with p-TIR) | N-WASP | |
|---|---|---|---|---|
| Null hypothesis | Single Exp. | Single Exp. | Single Exp. | Single Exp. |
| Alternate hypothesis | Double Exp. | Double Exp. | Double Exp. | Double Exp. |
| p value | <0.0001 | <0.0001 | <0.0001 | <0.0001 |
| Conclusion (alpha = 0.05) | Reject null hypo. | Reject null hypo. | Reject null hypo. | Reject null hypo. |
| Preferred model | Double Exp. | Double Exp. | Double Exp. | Double Exp. |
| F (DFn, DFd) | 64.16 (2282) | 47.33 (2635) | 46.72 (2379) | 48.64 (2379) |
Download links
A two-part list of links to download the article, or parts of the article, in various formats.
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
Phase transitions of multivalent proteins can promote clustering of membrane receptors
eLife 3:e04123.
https://doi.org/10.7554/eLife.04123
