Ratiometric Activatable Cell-Penetrating Peptides Provide Rapid In Vivo Readout of Thrombin Activation**

In real time: thrombin activation in vivo can be imaged in real time with ratiometric activatable cell penetrating peptides (RACPPs). RACPPs are designed to combine 1) dual-emission ratioing, 2) far red to infrared wavelengths for in vivo mammalian imaging, and 3) cleavage-dependent spatial localization. The most advanced RACPP uses norleucine (Nle)-TPRSFL as a linker that increases sensitivity to thrombin by about 90-fold.


General reagents and methods:
All the reagents and solvents were obtained from commercial sources and used without further purification. All the reactions involving Cy5 and Cy7 were carried out under dark or shielded from light. HPLC characterizations and purifications were performed on Agilent 1100 or 1200 with reverse-phase C 18 column (Phenomenex) using acetonitrile and water solvent system with 0.05% TFA as additive. Low resolution electrospray ionization (ESI) mass spectrometry was performed using Agilent HPLC connected to an Agilent LCMS trap XCT. UV absorbance was recorded on Cary 3E (Varian) or UV-2700 (Shimadzu). Fluorescence was recorded in quartz cuvets using a spectrofluorometer (FluoroLog®, Horiba Scientific).
Centrifugation was performed to isolate the pellet that was dried under high vacuum. The peptide was dissolved in dimethyl sulfoxide (DMSO) and purified by high performance liquid chromatography (HPLC) using 5-55% acetonitrile in water and 0.05% TFA. The purified product (1) was dried by lyophilization (mass obtained 3695.7 Da, mass calculated 3696.1 Da).
All the other RACPPs (10, 15, 20 and 25) were synthesized according to general scheme 1 and followed a similar synthetic protocol as that of RACPP DPRSFL (5). Analytical HPLC combined with mass spectrometry was used to confirm the identity of the compounds (Tables S1 and S2).
Purity of the final compounds was assessed by analytical HPLC ( Figure S5) using 5-55% acetonitrile in water gradient over 25 mins at 1 ml/min flow rate. Figure S1. Cleavage of RACPP PPRSFL (5) and RACPP PLGC(Me)AG (20) by 100 nM MMP-9 or thrombin. Digested peptides were separated on 4% agarose gels in 50 mM pentaethylenehexamine-acetate at pH 5.6. Gels were imaged for a) Cy5 fluorescence, b) Cy7 fluorescence upon exciting Cy5, brightened 6-fold, and c) Cy7 fluorescence. Panel d) is an overlay of a) in red, b) in green, and c) in blue. Panel e) is a pseudocolor Cy5/Cy7 emission ratio image of panels a) and b), generated as in Fig. 2 and 3. The two RACPPs were run on separate gels with the electrophoretic polarity shown between them. The thin white separation in the RACPP PLGC(Me)AG gel marks where irrelevant lanes and specular reflections were excised for clarity. Figure S2. Ratiometric fluorescent images of exudates from tail injuries in mice injected with RACPPs (PLGC(Me)AG, 20 or PPRSFL, 5 with hirudin), before and after the addition of exogenous enzyme. Images were taken 5 minutes after injection (top row), 40 minutes (RACPP PLGC(Me)AG ) or 25 minutes post injury (RACPP PPRSFL ) (middle row), or 10 minutes after local addition of either MMP-9 (1 µl of 1.5 µM, bottom left) or thrombin (10 µl of 5µM, bottom right). In each case there was little ratio change until the addition of exogenous enzyme (bottom row). Addition of the appropriate enzyme caused an increase of Cy5/Cy7 emission ratio (arrows). Color scale is identical to Fig. 3. Figure S3. Ratiometric fluorescent images of exudates from tail injuries in mice injected with RACPPs (PPRSFL, 5 or NleTPRSFL, 25 ). Images were taken 1, 5, or 15 minutes post injury. For quantitation average Cy5 and Cy7 fluorescent intensities were acquired from identical ROIs (highlighted rectangles in figure) and divided to determine Cy5/Cy7 ratios. For the top two rows images were scaled from with Cy5/Cy7 from 0.2 to 4.0, bottom row was scaled from 0.2 to 8.0 as in Figure 4.

Probe analysis using SDS polyacrylamide gel electrophoresis
Prior to gel electrophoresis 1 µM of each RACPP were cleaved in 150 mM NaCl, 20 mM Tris pH 7.5, 2 mM CaCl 2 , 1% BSA, 50 nM enzyme at 37°C. Purified thrombin, plasmin, factor Xa, and MMP-9 (activated) were purchased from EMD Chemicals. Samples were mixed with tricine SDS gel loading buffer and heated to 95°C for 5 min before loading in 10-20% tricine gels for electrophoresis. Images were taken on a Maestro multispectral imager (CRI Inc.) with 620 nm excitation and collection for Cy5 (660 to 720 nm) and Cy7 (760 to 830 nm) emission, respectively. Ratiometric images were synthesized by dividing the Cy5 emission by the Cy7 emission. Pseudocolors from blue (ratio minimum) to red (ratio maximum) were assigned using custom designed software.
Probe analysis using PEHA-Acetate agarose gel electrophoresis RACP PLGC(Me)AG and RACPP PPRSFL were cleaved in 150 mM NaCl, 20 mM Tris pH 7.5, 2 mM CaCl 2 or 2 X PBS buffer with 100 nM MMP-9 or thrombin at 37°C respectively. Samples were mixed with PEHA (80 mM pentaethylenehexamine neutralized with acetic acid to make pH 5.6)