Delivery of mirror image polypeptides into cells† †Electronic supplementary information (ESI) available. See DOI: 10.1039/c4sc02078b Click here for additional data file.

Mirror image peptides have unique stability and immunogenic properties in mammals, making them attractive agents to investigate.


Table of Contents
. Peptides used in this investigation 11 Table S2. Observed molecular masses of expressed protein constructs when analyzed by LC-MS 12 Table S3. List of variants 13 Table S4. Isolated yields of sortagging ligations from SrtA* reaction 14 Table S5. EC50 values of 30-minute protein synthesis inhibition assay 15 FIGURES Figure S1. Immunoblot of media from CHO-K1 cells treated with Sv1-3 16 Figure  All peptides, proteins and semi-synthetic products were analyzed on an Agilent 6520 Accurate-Mass quadrupole time-of-flight (Q-TOF) liquid chromatography-mass spectrometry (LC-MS) system. Solvent A' 0.1% formic acid in H 2 O and solvent B': 0.1% formic acid in MeCN was used. LC-MS used Agilent Zorbax 300SB C 3 column (2.1 x 150 mm, 5 µm) using a linear gradient of 5-65% B' over 15 min at a flow rate of 0.4 mL/min. The observed mass was generated by averaging the major peak in the total ion current (TIC). The charge-state series of the species were deconvoluted using Agilent MassHunter Bioconfirm using maximum entropy setting.

Preparative, semi-preparative, and analytical RP-HPLC
The crude peptides were dissolved in 99:1 or 95:5 A:B. If the peptide was insoluble under these solvent conditions, 6M guanidine·HCl was added to the solution. Peptides were purified by preparative RP-HPLC on Agilent Zorbax SB C 18 column (21.2 x 250 mm, 7 µm) at a flow rate of 10 mL/min at 1-41% B, 5-45% B, or 10-50% B over 80 min. For semi-preparative RP-HPLC, a Agilent Zorbax 300SB C 18 column (9.4 x 250 mm, 5 µm) was used at a flow rate of 5 mL/min over the same gradients. HPLC fractions were spotted with MALDI matrix alpha-cyano-4hydroxycinnamic acid (α-CHCA) in 50:50 A:B and checked for the correct molecular masses. The analytical RP-HPLC Agilent C 18 Zorbax SB column (2.1 x 150 mm, 5 µm) was used to confirm the purity of fractions at a flow rate of 0.5 mL/min over a linear gradient of 1-51% B over 12 min. Analytical HPLC UV absorbance traces were measured at 214 nm.

Synthesis of D-affibody
The synthesis of D-affibody containing G 5 at the N-terminus (63-mer) was performed by native chemical ligation (NCL) of three peptide segments (Scheme S1).

Synthesis of D-affibody-alkyne
D-affibody-alkyne was synthesized using the same strategy with an additional propargylglycine incorporated at the C-terminus of the protein. D-affibody-biotin was obtained by Cu(I)catalyzed azide-alkynyl click reaction to label the D-affibody-alkyne with biotin-azide. For the labeling reaction, 50 µL D-affibody-alkyne (332 µg, 46.8 nmol) was mixed with final concentration of 100 mM Tris pH 8.5, 1 mM biotin-azide (70 nmol), 2 mM CuSO 4 and 200 mM ascorbic acid in a total 70 µL reaction. LC-MS analysis showed 80% yield after 4 hours reaction at room temperature. The reaction mixture was dissolved in 6 M guanidine·HCl, 20 mM Tris-HCl, and 150 mM NaCl, pH 8.5, and was serially diluted from 6 M to 3 M then to 1 M guanidine·HCl using the same buffer without guanidine·HCl. The solution was desalted into 20 mM Tris-HCl, 150 mM NaCl, pH 7.5 using a HiTrap Desalting column. The folded protein was concentrated using a 3 kDa concentrator to give 202 µg, 28.9 nmol (61% yield).

Synthesis of D-GB1
The synthesis of D-GB1 containing G 5 at the N-terminus (61-mer) was performed by NCL of three peptide segments (Scheme S2).

Circular dichroism (CD) spectroscopy of folded proteins
Circular dichroism spectra of D-affibody and D-GB1 were recorded on an Aviv model 202 instrument at 25 °C. A 1 mm path length cell was used. The proteins were prepared by dissolving 0.06 mg in 50 mM sodium sulfate and 5 mM Tris-HCl at pH 8.5. The molar ellipticity (θ in deg cm 2 dmol -1 ) was calculated by [θ] λ = θ obs x 1/(10 lcn), where θ obs = observed ellipticity at λ, l = pathlength (cm), c = concentration of peptide (M), n = # of amino acids. The mirror image form of the L proteins showed the same optical rotation, but with the opposite sign, confirming the mirror image fold of the D-proteins ( Figure S4).

Synthesis of biotinylated p53/MDM2 Inhibitor D-peptide
Biotinylated p53/MDM2 inhibitor D-peptide was synthesized using Fmoc SPPS. The Nterminal glycine residue was protected with Boc. After the peptide was synthesized, the Alloc (allyloxycarbonyl) protecting group was removed as described above. Biotin (0.5 mmol) dissolved in 0.2 M HBTU (0.95 eq) was activated with 1.4 eq DIEA then coupled for 25 min at RT. The peptide was cleaved and purified and described above.

Construction of plasmids for recombinant proteins
The gene for 25-109 MDM2 was purchased from Addgene (pGEX-4T MDM2 WT, 16237). [5] The pET SUMO-25-109 MDM2 was prepared using the Champion pET SUMO protein expression system (Invitrogen, CA). AccuPrime Taq DNA polymerase (Invitrogen, CA) was used to PCR amplify the DNA using the following primers: The PCR product was purified by the QIAquick PCR purification kit (Qiagen, Netherlands), then cloned into pET SUMO by an overnight ligation at 15 °C with 3 ng PCR product, 50 ng pET SUMO vector, and 1 µl T4 DNA ligase. 2 µl of the ligation product was transformed into One Shot Mach1-T1 competent cells and plated on 30 µg/mL kanamycin plates and incubated overnight at 37 °C. Colonies were grown in LB media containing 30 µg/mL kanamycin. The plasmid DNA was isolated using the Qiaprep spin miniprep kit (Qiagen, Netherlands).

Protein expression and purification
SUMO-25-109 MDM2 was expressed in Rosetta (DE3)  Approximately 40 g of cell pellet was resuspended in 100 ml of 50 mM Tris-HCl, 150 mM NaCl, pH 7.5 buffer containing 200 mg lysozyme, 4 mg Roche DNAase I, and 2 tablet of Roche protease inhibitor cocktail then sonicated for three times for 20 seconds. After sonication, the suspension was centrifuged at 17,000 rpm for 40 minutes. The lysate was loaded onto three 5 ml HisTrap FF crude Ni-NTA column (GE Healthcare, UK) and washed with 100 mL of 20 mM Tris-HCl pH 8.5, 150 mM NaCl, at pH 8.5 and 100 mL of 40 mM imidazole in 20 mM Tris-HCl pH 8.5, 500 mM NaCl. The protein was eluted from the column using 500 mM imidazole in 20 mM Tris-HCl pH 8.5, 500 mM NaCl. The eluted protein was buffer exchanged into 20 mM Tris-HCl pH 8.5, 150 mM NaCl using a HiPrep 26/10 Desalting column (GE Healthcare, UK). WT PA and PA [F427H] was overexpressed in the periplasm of E. coli BL21 (DE3) cells and purified by anion exchange chromatography. [6] One-pot sortagging reaction using Staphylococcus aureus SrtA* Enzyme-mediated ligation using Sortase A was utilized to ligate the peptide and protein cargo to LF N -DTA-LPSTGG (for SDv's) or LF N -LPSTGG (for Sv's), depending on the assay. We used Staphylococcus aureus SrtA 59-206 that was evolved (P94S/D160N/K196T) by Chen et al. (SrtA*). [7] N-terminal small ubiquitin-like modifier (SUMO) was first cleaved from the protein substrate, His 6 -SUMO-LF N -DTA-LPSTGG-His 5 (or His 6 -SUMO-LF N -LPSTGG-His 6 ), using 1 µg SUMO protease per mg of protein substrate at RT for 30 minutes to give the native Nterminus of LF N -DTA-LPSTGG-His 5 (or LF N -LPSTGG-His 6 ).
The ligation reactions were performed on Ni-NTA agarose beads using SrtA*. Ni-NTA beads were washed three times with SrtA* buffer (10 mM CaCl 2 , 50 mM Tris-HCl, 150 mM NaCl, pH 7.5). In one pot, 50 µM LF N -DTA-LPSTGG-His 5 (or LF N -LPSTGG-His 6 ), 5 µM SrtA*, and 100 to 500 µM G 5 -cargo were incubated with Ni-NTA beads in SrtA* buffer for 30 min at RT nutating. After 30 min, the Ni-NTA beads containing any unreacted LF N -DTA-LPSTGG-His 5 , SrtA*-His 6 , His 6 -SUMO, and GG-His 6 were spun down at 4 °C to minimize the formation of the hydrolyzed LF N -DTA-LPST (or LF N -LPST) side product. The supernatant containing the sortagged LF N -DTA variant (SDv) or (sortagged LF N; Sv) was collected. The beads were washed three times with 1 mL of 20 mM Tris-HCl, 150 mM NaCl, pH 7.5. Three buffer exchanges were performed into 20 mM Tris-HCl, 150 mM NaCl, pH 7.5 to remove the excess G 5 -peptide. Size exclusion chromatography into 20 mM Tris-HCl, 150 mM NaCl, pH 7.5 was used to remove the excess G 5 -protein. The purity of the ligated product was analyzed by LC-MS.

Translocation of SDv's and protein inhibition assay in CHO-K1 cells
The CHO-K1 cells were maintained in F-12K media supplemented with 10% (v/v) fetal bovine serum at 37 °C and 5% CO 2 . The cells were plated at 3.0 x 10 4 per well in a 96-well plate one day prior to the assay. The SDv's were prepared in ten-fold serial dilutions in 50 µL, and added 50 µL of F-12K media containing 20 nM PA. The 100 µL samples were added to CHO-K1 cells and incubated for 30 minutes at 37 °C and 5% CO 2 . (SDv4 and SDv4-biotin was incubated for 2 hours.) After incubation with SDv's, the cells were washed three times with PBS and then 100 µL leucine-free F-12K medium supplemented with 1 µCi/mL 3 H-leucine (Perkin Elmer, MA) was added and incubated for 1 h at 37 °C 5% CO 2 . The cells were washed three times with PBS and suspended in 150 µL of scintillation fluid. 3 H-Leu incorporation into cellular proteins was measured to determine the inhibition of protein synthesis by LF N -DTA. The scintillation counts from cells treated with only PA were used as control value for normalization. Each experiment was done in triplicate.
The data were plotted using Origin8 using Sigmoidal Boltzmann Fit using equation and x 0 represents the logEC 50 values.

Cytosolic protein extraction and whole cell lysate preparation for western blot
CHO-K1 cells were plated in 12-well plate. The cells were treated with 250 nM Sv1, 2, 4, and 5 in the presence of 40 nM PA overnight at 37 °C and 5% CO 2 . After treatment, the cells were washed with PBS and lifted with 0.25% trypsin-EDTA for 5 minutes to remove surface bound proteins then washed twice with PBS. For cytosolic protein extraction, 1 x 10 6 cells were resuspended in 100 µL of 50 µg mL -1 digitonin in 75 mM NaCl, 1 mM NaH 2 PO 4 , 8 mM Na 2 HPO 4 , 250 mM sucrose supplemented with Roche protease inhibitor cocktail for 10 min on ice, and centrifuged for 5 minutes at 13,000 rpm. For whole cell lysate, cells were lysed in IP lysis buffer (25 mM Tris, 150 mM NaCl, 1% v/v NP-40, pH 7.5) supplemented with Roche protease inhibitor cocktail on ice for 30 minutes and centrifuged for 10 minutes at 13,000 rpm. Both supernatants were collected for blotting and analysis.

Western Blot of Sv1, 2, 4, and 5 translocated into CHO-K1 cells
Lysates were run on an SDS-PAGE gel then transferred onto a nitrocellulose membrane soaked in 48 mM Tris, 39 mM glycine, 0.0375% SDS, 20% methanol and using a TE 70 Semi-Dry Transfer Unit (GE). After transferring, the membrane was blocked at room temperature for 2 hours with LI-COR blocking buffer. The membrane was then incubated with anti-LF or anti-Erk1/2, and anti-Rab5 in TBST overnight at 4 ºC. The membrane was washed and incubated with the appropriate secondary antibodies in TBST for 1 h then imaged by the LI-COR Odyssey infrared imaging system.

Trypsin digestion of L-and D-Affibody
Digestion of 3 µg L-and D-affibody with 3 ng trypsin at 37 °C in 100 µL were monitored up to 23 hours. The digestion was monitored by LC-MS.

Pull down of Sv4-biotin in U-87 MG cells
U-87 MG cells were grown in T-75 flasks at a density of 5x10 6 per flask. Cells were treated with 100 nM Sv3-biotin in the presence of 20 nM PA or PA [F427H] for 24 hours at 37 °C and 5% CO 2 . After treatment, the cells were lifted with 0.25% trypsin-EDTA, washed three times, and lysed with 500 µL IP buffer for 10 min on ice. As a negative control, untreated cells. Lysates were incubated with 20 µL (42 µg) streptavidin-agarose beads (Sigma Aldrich) for 4 hours at 4 °C. The beads were washed three times with lysis buffer and once with lysis buffer without NP-40. The bound proteins were eluted in 1X Laemmli sample buffer (Bio-Rad), boiled for 10 min, and then analyzed by SDS-PAGE gel. Standard immunoblotting was performed using anti-MDM2 antibody and streptavidin-680.

Translocation of Sv4 and Sv4-biotin in U-87 MG or K562 cells
U-87 MG cells were grown in T-25 flasks at a density of 1x10 6 per flask. K562 cells were treated in 24 well dishes at a density of 8.0 x 10 5 per well. Cells were treated with 150 nM of Sv1, Sv4, or Sv4-biotin in the presence of 20 nM PA or 20 nM PA [F427H] for 24 hours at 37 °C and 5% CO 2 . As a positive control, cells were treated with 1 µM nutlin-3 for 24 h. After treatment, cells were lifted by trypsin and lysed with IP lysis buffer. The isolated lysates were subjected SDS-PAGE separation and transferred to a nitrocellulose blot as previously described. The membrane was immunoblotted with anti-LF, anti-MDM2, anti-p21, anti-p53, and anti-GAPDH overnight in TBST. The membrane was washed and incubated with the appropriate secondary antibody in TBST for 1 h and then imaged.
The western blot images were analyzed and quantified using the Image Studio program (LI-COR). Each band was normalized to the GAPDH loading control. The band intensity of each protein was set to one in the PA only control. Percent increase in amount of MDM2, p53, and p21 present in each lane compared to the PA only control was plotted in Figure S6.

Binding interaction between SUMO-25-109 MDM2 and 4 or Sv4
A competition binding assay was performed using a bilayer interferometry system (Octet RED96, ForteBio, CA) to determine the binding affinity of 4 (peptide) and Sv4 to SUMO-25-109 MDM2. Super streptavidin (SSA) sensors were soaked in 1X kinetics buffer (1X PBS containing 0.02% Tween-20, 0.1% BSA, pH 7.4) for 10 minutes at 30 °C. For immobilization, the wild-type 15-29 p53 peptide was synthesized containing a biotin on the N-terminus (biotin-15-29 p53). The biotin-15-29 p53 peptide (1 µM in 1X kinetics buffer) was immobilized on the SSA sensor surface for 5 minutes at 30 °C and 1000 rpm. The surface was washed for 5 minutes with 1X kinetics buffer to establish a baseline. Two-fold dilutions of SUMO-25-109 MDM2 in 1X kinetics buffer were analyzed for binding over 5 minutes at 30 °C and 1000 rpm followed by dissociation with 1X kinetics buffer. A calibration curve was generated using GraphPad Prism 6 software using non-linear regression analysis (one-site, specific binding fit). The K d was found to be 578 ± 170 nM.
For the competition assay, various concentrations of 4 (peptide) or Sv4 were incubated with 50 nM SUMO-25-109 MDM2 at room temperature for 30 minutes. Meanwhile, SSA sensors were soaked in 1X kinetics buffer for 10 minutes at 30 °C. The biotin-15-29 p53 peptide (1 µM) was immobilized on the SSA sensor surface then washed to establish a baseline, as previously described for the calibration curve. After immobilization, the association and dissociation of SUMO-25-109 MDM2 pre-incubated with 4 or Sv4 samples were analyzed at 30 °C and 1000 rpm and using 1X kinetics buffer. Based on the binding (nm) values, the concentration of unbound MDM2 was interpolated for each sample using the calibration curve.
Non-linear regression analysis was performed using GraphPad Prism 6 software to determine the K d value based on the equation: K d = [peptide][MDM2]/[complex]. [8] We used the following equation to generate fitted curves: = ; where y is [MDM2] in nM, x is [inhibitor] in nM, K d is the dissociation constant, and b is y max .

Sequence
Observed ( Figure S1. Immunoblot of media from CHO-K1 cells treated with Sv1-3 ( Figure 2c). Serumcontaining F12K with 250 nM wild-type LF N or Sv1-3 and 40 nM PA was removed after overnight treatment with CHO-K1 cells. The stability of the LF N variants in media was analyzed by immunostaining with anti-LF.
Sv-LF N 1 2 4 ng 2 3 anti-LF Figure S2. Linear relationship between Sv4 band intensity and the amount of protein loaded (ng): y = 566000x + 107000, R 2 = 0.993). The signal intensity of each band was compared to 0.1 -10 ng of pure Sv4 protein loaded on the gel and immunoblotted with anti-LF. immobilized to super streptavidin sensors. b) A calibration curve was generated using GraphPad Prism 6 software using non-linear regression analysis (one-site, specific binding fit). The K d was found to be 578 ± 170 nM. . [8] We used the following equation to generate fitted curves: = ; where y is [MDM2] in nM, x is [inhibitor] in nM, K d is the dissociation constant, and b is y max . The K d values calculated for 4 and Sv4 were 1.0 ± 0.7 nM and 12.3 ± 4.3 nM, respectively.  . For A and B, the right panel corresponds to the mass spectra of the major peak area and for C and D, the right panel corresponds to the deconvoluted mass of the major peak area.