SUPPLEMENTARY INFORMATION. Single-molecule site-specific detection of protein phosphorylation with a nanopore

Transcription

1 SUPPLEMENTRY INFORMTION Single-molecule site-specific detection of protein phosphorylation with a nanopore Christian B. Rosen 1,2*, David Rodriguez-Larrea 1* and Hagan Bayley 1 Department of Chemistry, University of Oxford, Oxford, OX1 3T, United Kingdom 1 Department of Chemistry, University of Oxford, Oxford, OX1 3T, United Kingdom. 2 Center for DN Nanotechnology, Department of Chemistry and inno, arhus University, 8 arhus C, Denmark. *Both author contributed equally Correspondence should be addressed to D. R.-L. or H.B. david.rodriguez-larrea@chem.ox.ac.uk hagan.bayley@chem.ox.ac.uk

2 Construct I RES verage () I RES S.D. () I n (p) I n S.D. (p) TrxS112 'P '(dc) TrxS112 +P '(dc) TrxS17 'P '(dc) TrxS17 +P '(dc) TrxS95 'P '(dc) TrxS95 +P '(dc) TrxS17 'P /S112 'P '(dc) TrxS17 +P /S112 +P '(dc) Trx17/S112 'P '(dc) Trx17/S112 +P '(dc) TrxS17 'P /112'(dC) TrxS17 +P /112'(dC) Supplementary Table 1: Pore to pore variation. Residual current (I RES ) and noise (I n ) for all the constructs used in this work at +14 mv. Each construct was studied by using 3 different pores in 3 independent experiments (analyzing at least 5 events per experiment). Note that constructs TrxS17/S112-(dC) 3, Trx17/S112-(dC) 3, and TrxS17/112-(dC) 3 contain two sub-levels in level 3 unless S17 is phosphorylated.

3 a b c d e Supplementary Figure 1: Voltage dependence of the co-translocational unfolding of TrxS112 -P -(dc) 3 within the αhl pore. (a) Representative current trace with 4 levels at +14 mv. (b) Molecular model of the thioredoxin mutant TrxS112 -P showing the PK recognition sequence (blue sticks) and the phosphorylation site (red ball). (c) Voltage dependence of the rate of step 2 3, (Ο) V5-C19-(dC) 3 1 ; (Ο) TrxS112 -P -(dc) 3. (d) Voltage dependence of the rate of step 3 4, (Ο) V5-C19-(dC) 3 1 ; (Ο) TrxS112 - P -(dc) 3. (e) Voltage dependence of the rate of step 4 1, (Ο) V5-C19-(dC) 3 1 ; (Ο) TrxS112 -P -(dc) 3. Error bars represent the standard deviation between independent experiments, each using a different pore (n = 3).

4 (2.483) TOF MS ES e CRL (4.857) TOF MS ES+ 1.3e (2.417) TOF MS ES e5 1 CRL (4.221) TOF MS ES+ 1.3e (2.482) TOF MS ES e CRL (4.23) TOF MS ES+ 1.25e6 a b c 122 m/z m/z m/z d e f Phos EXT 2days m/z HELIX phos 2days m/z Phos LOOP 3days m/z m/z m/z Supplementary Figure 2: ESI LC-MS in positive ion mode before and after phosphorylation of constructs TrxS112, TrxS17 and TrxS95 (deconvoluted spectra). (a) TrxS112 -P (expected 1222). (b) TrxS17 -P (expected 11832). (c)trxs95 -P (expected 11916). (d) TrxS112 +P. (e) TrxS17 +P. (f)trxs95 +P. The expected gain after phosphorylation is 8 Da. Note that the spectra in (e) and (f) contain peaks for the non-phosphorylated protein.

5 a b c Supplementary Figure 3: Voltage dependence of the co-translocational unfolding of Trx112 +P -(dc) 3 within the αhl pore. (a) Voltage dependence of the rate of step 2 3, (Ο) Trx112 -P -(dc) 3 ; (Ο) Trx112 +P -(dc) 3. (b) Voltage dependence of the rate of step 3 4, (Ο) Trx112 -P -(dc) 3 ; (Ο) Trx112 +P -(dc) 3. (c)voltage dependence of the rate of step 4 1, (Ο) Trx112 -P -(dc) 3 ; (Ο) Trx112 +P -(dc) 3. Error bars represent the standard deviations between independent experiments, each with a different pore (n = 3).

6 a b c.5 s.5 s d e f.5 s.5 s g h i.5 s.5 s j k l.5 s.5 s m n o.5 s.5 s p q r.5 s.5 s Supplementary Figure 4: Spectral noise analysis of level 3 in various unphosphorylated and phosphorylated Trx constructs. In general, phosphorylation gave a decrease in the low frequency spectral density and a small increase in the corner frequency (f C, but note that the filter cut-off was 5 khz). (a) Representative current trace of Trx112 -P -(dc) 3. (b) Representative current trace of Trx112 +P -(dc) 3. (c) Level 3 noise spectra of Trx112 -P -(dc) 3 (blue) and Trx112 +P -(dc) 3 (red). (d) Representative current

7 trace of Trx17 -P -(dc) 3. (e) Representative current trace of Trx17 +P -(dc) 3. (f) Level 3 noise spectra of Trx17 -P -(dc) 3 (blue) and Trx17 +P -(dc) 3 (red). (g) Representative current trace of Trx95 -P -(dc) 3. (h) Representative current trace of Trx95 +P -(dc) 3. (i) Level 3 noise spectra of Trx95 -P -(dc) 3 (blue) and Trx95 +P - (dc) 3 (red). (j) Representative current trace of TrxS17 -P /S112 -P -(dc) 3. (k) Representative current trace of TrxS17 +P /S112 +P -(dc) 3. (l) Level 3 noise spectra of TrxS17 -P /S112 -P -(dc) 3 (blue) and TrxS17 +P /S112 +P -(dc) 3 (red). (m) Representative current trace of Trx17/S112 -P -(dc) 3. (n) Representative current trace of Trx17/S112 +P -(dc) 3. (o) Level 3 noise spectra of Trx17/S112 -P - (dc) 3 (blue) and Trx17/S112 +P -(dc) 3 (red). (p) Representative current trace of TrxS17 -P /112-(dC) 3. (q) Representative current trace of TrxS17 +P /112- (dc) 3. (r) Level 3 noise spectra of TrxS17 -P /112-(dC) 3 (blue) and TrxS17 +P /112-(dC) 3 (red). ll constructs were examined at +14 mv.

8 a b Supplementary Figure 5: Voltage-dependences of the residual current (I RES ) and noise (I n ) values of level 3. (a) Residual current (I RES ) of level 3 as a function of the applied voltage: (Ο) Trx112 -P -(dc) 3; (Ο) Trx112 +P -(dc) 3. (b) Noise (I n ) as a function of voltage. I n is the standard deviation of a Gaussian fit to an all-points histogram of the ionic current in level 3. (Ο) Trx112 -P -(dc) 3; (Ο) Trx112 +P -(dc) 3. Error bars represent the standard deviations between independent experiments, each with a different pore (n = 3).

9 a b c d e f Supplementary Figure 6: Voltage dependences of the co-translocational unfolding of various Trx within the αhl pore. (a) Voltage dependences of the rates of step 2 3: (Δ) Trx17 -P -(dc) 3 ; (Δ) Trx17 +P -(dc) 3 ; (Ο) Trx112 -P -(dc) 3. (b) Voltage dependences of the rates of step 3 4: (Δ) Trx17 -P -(dc) 3 ; (Δ) Trx17 +P -(dc) 3 ; (Ο) Trx112 -P -(dc) 3. (c) Voltage dependences of the rates of step 4 1: (Δ) Trx17 -P - (dc) 3 ; (Δ) Trx17 +P -(dc) 3 ; (Ο) Trx112 -P -(dc) 3. (d) Voltage dependences of step 2 3: ( ) Trx95 -P -(dc) 3 ; (Ο) Trx112 -P -(dc) 3. (e) Voltage dependences of the rates of step 3 4: ( ) Trx95 -P -(dc) 3 ; (Ο) Trx112 -P -(dc) 3. (f) Voltage dependences of the rates of step 4 1: ( ) Trx95 -P -(dc) 3 ; (Ο) Trx112 -P -(dc) 3. Error bars represent the standard deviations between independent experiments, each with a different pore (n = 3).

10 Supplementary Figure 7: 2D plot of residual currents (I RES ) and noise (I n ) values and associated histograms for: (Ο) TrxS17 -P /S112 -P -(dc) 3, (Ο) Trx17/S112 -P - (dc) 3 and (Ο) TrxS17 -P /112-(dC) 3. ll three constructs were examined with the same αhl pore (the cis compartment was perfused before the addition of each Trx variant) at +14 mv. 3 events were analyzed in total. The high conductivity sub-states of level 3 are not included in this figure (see Supplementary Fig. 8 for a zoom out). Each construct was further analyzed in two additional independent experiments, each with a different pore, with similar results.

11 Supplementary Figure 8: Zoom out of Supplementary Fig 7. 2D plot of residual current (I RES ) and noise (I n ) values for: (Ο) TrxS17 -P /S112 -P -(dc) 3, (Ο) Trx17/S112 -P - (dc) 3 and (Ο) TrxS17 -P /112-(dC) 3. ll three constructs were examined with the same WT αhl pore (the cis compartment was perfused before the addition of each Trx variant) at +14 mv. 3 events were analyzed in total. Sub-states of higher conductance are observed at I RES values of approximately 24 of the open pore value (I O ). Each construct was further analyzed in two additional independent experiments, each with a different pore, with similar results.

12 a b c 1234 m/z m/z Blind-EXT, HLX t CRL (3.937) m/z TOF MS ES+ 1.31e d e f double phos more PK 44h CRL (4.359) m/z EXT, BL-HLX t2h4min TOF MS ES+ CRL (4.19) 1.14e m/z TOF MS ES+ 1.2e6 BL-EXT,HLX more PK 44h CRL (4.375) m/z TOF MS ES+ 8.71e Supplementary Figure 9: ESI LC-MS in positive ion mode before and after phosphorylation of constructs TrxS17/S112, Trx17/S112 and TrxS17/112 (deconvoluted spectra). (a) TrxS17 -P /S112 -P (expected 1233). (b) Trx17/S112 -P (expected 12287). (c) TrxS17 -P /112 (expected 12287). (d) TrxS17 +P /S112 +P. (e) Trx17/S112 +P. (f) Trx17/S112 +P. The expected gain after phosphorylation at one site is 8 Da.

13 Supplementary Figure 1: Zoom-out of a 2D plot of residual currents (I RES ) versus noise (I n ) from Fig. 2 (main text). (Ο) TrxS17 +P /S112 +P -(dc) 3, (Ο) Trx17/S112 +P - (dc) 3, (Ο) TrxS17 +P /112-(dC) 3, and (Ο) TrxS17 -P /S112 -P -(dc) 3. ll the measurements were done with the same WT αhl pore (the cis compartment was perfused before the addition of each Trx variant) at +14 mv and involved the measurement of a total of 342 events. few events may be due to carry over because of incomplete perfusion. Sub-states of higher conductance are observed at I RES values of 21 to 22 of the open pore value (I O ). Each construct was further analyzed in two additional independent experiments, each with a different pore, with similar results.

14 17 16 a b I RES () I RES () I n (p) I n (p) c d I RES () I RES () I n (p) I n (p) Supplementary Figure 11: 2D scatter plots of residual currents (I RES ) versus current noise values (I n ) for calibration of the pore used in Fig. 3b. (a) The pore was exposed to TrxS17 +P /112-(dC) 3. (b) Without perfusion of the cis compartment, Trx17/S112 +P -(dc) 3, was then added and a new recording made. Two populations are now apparent. (c) TrxS17 +P /S112 +P -(dc) 3 was then added to the cis compartment without perfusion and a recording made. (d) TrxS17 -P /S112 -P -(dc) 3 was then added, again without perfusion. The calibration involved a total of 162 events, and was performed with the same pore after data were collected for Fig. 3b. similar calibration was performed before the experiment, with the phosphorylated Trx added in a different order, with a similar result. Each construct was further analyzed in two additional independent experiments, each with a different pore, with similar results.

15 References 1. Rodriguez-Larrea, D. & Bayley, H. Multistep protein unfolding during nanopore translocation. Nat. Nanotechnol (213)

16 METHODS αhl nanopores Wild-type (WT) αhl monomers were expressed by Ellina Mikhailova in an E. coli in vitro transcription/translation (IVTT) system and merized to form heptameric pores on rabbit red blood cell membranes. The heptameric pores were purified by sodium dodecyl sulphate (SDS) polyacrylamide gel electrophoresis 15,16. Thioredoxin mutants The thioredoxin (Trx) V5-C19 gene was cloned into the pet 3ab plasmid (TopGene). The Trx mutants were produced by site-directed mutagenesis (QuickChange II XL, Stratagene), and verified by DN sequencing. Protein expression was performed using E. coli BL21(DE3) cells (Novagen) after induction with IPTG in the exponential growth phase. The proteins were purified by size-exclusion chromatography (Superdex 75 1/3 GL, Tricorn, GE Healthcare) in TE buffer (1 mm Tris.HCl, 1 mm EDT, ph 8.3) with 1 mm DTT followed by ion-exchange chromatography (HiTrap Q FF, GE Healthcare) eluted with a gradient of -1 M KCl in TE buffer with 1 mm DTT, ph 8.3. Protein es were confirmed by electrospray ionization liquid chromatography- spectrometry (ESI LC- MS) 8. Catalytic subunit of protein kinase Hexahistidine-tagged catalytic subunit of protein kinase (PK) was purified for sitespecific serine phosphorylation of Trx mutants. The pet15b PK Cat plasmid 17 was transformed into Rosetta(DE3)pLysS cells (Novagen). The cells were grown at 37 C in Luria Broth containing ampicillin (5 µg ml -1 ) to OD 6 =.6 to.8. The cell culture was then induced with IPTG, at a final concentration of.5 mm, and incubated at 18 C for 24 h. Cells were harvested by centrifugation and lysed with BugBuster Master Mix (Novagen) before loading into a gravity-flow Ni-NT Superflow affinity column (Qiagen). fter washing with phosphate buffer (1 mm phosphate, 15 mm NaCl, ph 7.2) the hexahistidine-tagged catalytic subunit was eluted with 5 mm imidazole in phosphate buffer. The of the protein was confirmed by ESI LC-MS. Phosphorylation of thioredoxin mutants Trx mutants were phosphorylated on the serine residue of the RRXS recognition sequence by using the catalytic subunit of PK. The Trx mutants (~.5-1 mg ml -1 ) in 2 mm Tris.HCl buffer, containing 2 mm MgOc, ph 7.4, were incubated with 2 mm DTT,.2 mm adenosine 5 -triphosphate (TP, disodium salt hydrate, Sigma-ldrich), and ~.6 mg ml - 1 PK. The phosphorylation kinetics were followed by ESI LC-MS and isoelectric focusing (IEF) gel electrophoresis. Phosphorylation on TrxS112 -P was complete within 2 h. For TrxS17 -P and TrxS95 -P, additional TP and PK were added to increase the yield of phosphorylation and the incubation time was extended. The phosphorylated proteins were purified by size-exclusion chromatography in TE buffer (1 mm Tris.HCl, 1 mm EDT, ph 8.3) containing 1 mm DTT (Superdex 75 1/3 GL, Tricorn, GE Healthcare).

17 Oligonucleotide-thioredoxin conjugates Oligonucleotide-Trx conjugates were obtained as previously described 8. Briefly, the Trx mutants and 5'-thiol (hexamethylene linker) modified (dc) 3 (Integrated DN Technologies) were separately reduced for 24 h in DTT (1 mm). DTT was removed by buffer exchange (1 mm Tris.HCl, ph 8.) by using PD-1 Desalting Columns (GE Healthcare) and the 5'-thiol (dc) 3 was activated with 2,2'-dipyridyl disulfide (1 mm in acetonitrile), purified with a PD-1 Desalting Column (GE Healthcare) and then reacted with the reduced Trx mutants for 16 h at room temperature (after buffer exchange of the proteins into 1 mm Tris.HCl, ph 1.). The conjugates were purified by ion-exchange chromatography (HiTrap Q FF, GE Healthcare) by using a gradient of -1 M KCl in TE buffer (1 mm Tris.HCl, 1 mm EDT, ph 8.3). Final concentrations were determined from the absorbance at 26 nm by using the calculated molar extinction coefficient of (dc) 3. Single channel recordings and data analysis Electrical recordings were performed with planar lipid bilayers at 21. ± 2. C. bilayer of 1,2-diphytanoyl-sn-glycero-3-phosphatidylcholine (vanti Polar Lipids) was formed across an aperture of 1 µm diameter in a Teflon film (Goodfellow) separating the cis and trans compartments of the recording apparatus (1 ml each). Both compartments were filled with 1 mm HEPES, 2 M KCl, ph 7.4. Gel-purified αhl heptamers (~.2 µl, ~1 ng μl -1 ) were added to the grounded cis compartment. fter the insertion of a single pore, the cis compartment was manually perfused with fresh buffer to prevent further insertions. Trx mutants were added to the cis compartment to a final concentration of.1-.2 µm. Ionic currents produced by an applied potential were measured by using g/gcl electrodes connected to a patch-clamp amplifier (xopatch 2B, xon Instruments). Signals were low-pass-filtered at 5 khz and sampled at 25 khz with a Digidata 144 digitizer (xon Instruments). Data analysis was performed with pclamp software (Molecular Devices). Events were collected by threshold searches. Very short events (<1 ms) and long blockades (>1 s) were excluded. Residual current values (I RES ) and noise levels (I n ) of level 3 were determined by fitting all-points histograms (.2 p bin) to Gaussian curves (I RES = I B /I O X 1, where I B is the residual current and I O the current through the unoccupied pore 16 ; I n = standard deviation of the fit). pproximately 1 individual events for each construct were used for the 2D I RES versus I n plots. Dwell times for levels 1, 2, and 3 were plotted as unbinned cumulative histograms and fitted to single exponentials (Igor Pro 6.12, WaveMetrics) to obtain mean dwell times. Error bars for each construct represent the standard deviation for 3 independent experiments. ssociated References 15. Stoddart, D., Heron,.J., Mikhailova, E., Maglia, G. & Bayley, H. Single-nucleotide discrimination in immobilized DN nucleotides with a biological nanopore. Proc. Natl. cad. Sci. U.S (29) 16. Maglia, G., Heron,.J., Stoddart, D., Japrung, D. & Bayley, H. nalysis of single nucleic acid molecules with protein nanopores. Methods Enzymol (21)

18 17. Narayana, N., Cox, S., Shaltiel, S., Taylor, S.S. & Xuong, N. Crystal structure of a polyhistidine-tagged recombinant catalytic subunit of cmp-dependent protein kinase complexed with the peptide inhibitor PKI(5-24) and adenosine. Biochemistry (1997)