Long term safety of targeted internalization of cell penetrating peptide crotamine into renal proximal tubular epithelial cells in vivo

Activated proximal tubular epithelial cells (PTECs) play a crucial role in progressive tubulo-interstitial fibrosis in native and transplanted kidneys. Targeting PTECs by non-viral delivery vectors might be useful to influence the expression of important genes and/or proteins in order to slow down renal function loss. However, no clinical therapies that specifically target PTECs are available at present. We earlier showed that a cationic cell penetrating peptide isolated from South American rattlesnake venom, named crotamine, recognizes cell surface heparan sulfate proteoglycans and accumulates in cells. In healthy mice, crotamine accumulates mainly in kidneys after intraperitoneal (ip) injection. Herein we demonstrate for the first time, the overall safety of acute or long-term treatment with daily ip administrated crotamine for kidneys functions. Accumulation of ip injected crotamine in the kidney brush border zone of PTECs, and its presence inside these cells were observed. In addition, significant lower in vitro crotamine binding, uptake and reporter gene transport and expression could be observed in syndecan-1 deficient HK-2 PTECs compared to wild-type cells, indicating that the absence of syndecan-1 impairs crotamine uptake into PTECs. Taken together, our present data show the safety of in vivo long-term treatment with crotamine, and its preferential uptake into PTECs, which are especially rich in HSPGs such as syndecan-1. In addition to the demonstrated in vitro gene delivery mediated by crotamine in HK-2 cells, the potential applicability of crotamine as prototypic non-viral (gene) delivery nanocarrier to modulate PTEC gene and/or protein expression was confirmed.

www.nature.com/scientificreports www.nature.com/scientificreports/ ip administration of crotamine. However, the MS analyses did not allow detecting crotamine in the urine of mice treated daily with low doses of crotamine (1 μg/animal).
Tissue morphology assessment of mice kidneys by histological analysis after classical hematoxylin/eosin (H&E) and periodic acid-Schiff (PAS) staining highlighted the absence of any significant observable alterations in kidney tissues sections of long-term crotamine-treated mice, as representatively shown in Fig. 2. It is also important to note the absence of any remarkable injuries or any other important alterations in other tissues as liver, lung, spleen and heart of mice treated daily with crotamine (1 μg/animal) by ip route, for 21 days, which was exactly the same condition previously adopted by us for the antitumoral therapy with this peptide 34,35 .

Crotamine internalization in renal proximal epithelial cells in vivo.
Immunohistochemical analysis of the same kidneys allowed visualizing native crotamine administered at low doses (1 µg/animal) by ip injections daily, during 21 days. Crotamine was localized into the PTECs, indicating the in vivo uptake of crotamine by these cells (Fig. 3). As a control for antibody specificity, immunohistochemical analysis of the same kidneys with rabbit anti-crotamine, blocked with native crotamine, for 1 h prior to staining procedure, showed a completely negative staining.
Fluorescent analysis of sections of mice kidney collected 2 h after acute single dose ip administration of fluorescently labeled Cy3-crotamine (5 µg/animal), employing In Cell Analyzer for whole tissue section analysis and also confocal microscopy for magnified pictures allowed observing the presence of crotamine localized in the microvilli-covered luminal surface (brush borders) of PTECs in several regions of the kidney (Fig. 4).

Crotamine interaction with and internalization by human PTEC HK-2 cells. First we confirmed by FACS analysis (Supplementary Method S3) that the shRNA strategy reduced the expression of in PTEC HK-2 cells in about 80%
(see Supplementary Fig. S5). The binding of Cy3-crotamine was also demonstrate to be ~3-4 times higher for wild-type human PTEC HK-2 cells compared with Synd-1 KD cells, which support the suggestion of a possible Synd-1 role in crotamine uptake by PTEC HK-2 cells (see Supplementary Fig. S5). In addition, confocal microscopy analysis allowed observing crotamine distributed in the cell cytosol, and accumulated in perinuclear area of HK-2 cells incubated with crotamine for 1 h (Fig. 6A,B), and this crotamine distribution pattern was similar to that in Synd-1 KD cells (Fig. 6C,D), but with a significant reduced intensity (~four fold less) ( Supplementary  Fig. S6).
Delivery of plasmid DNA into PTEC cells. To verify whether crotamine is able to translocate plasmid DNA into PTEC cells, a complex of crotamine and pEGFP-N1 plasmid vector was used to transfect wild-type HK-2 and Synd-1 KD cells. Crotamine-DNA complexes, formed by a peptide:DNA liquid charge ratio of 10:4, diluted in PBS solution was dropped onto semi-confluent cultured cells, and 24 h after, s strong green fluorescent signal, due to the plasmid DNA delivery and green fluorescent reporter protein (GFP) expression, was observed in wild-type HK-2 cells (Fig. 7), demonstrating the efficiency of transfection mediated by crotamine in this specific cell type. As expected, the green fluorescence signal was mainly localized in the cytoplasm of transfected cells, as similarly www.nature.com/scientificreports www.nature.com/scientificreports/ observed for Synd-1 KD HK-2 cells, but with significantly lower signal intensity (about 2.5 fold lower compared to wild-type cells) ( Supplementary Fig. S6), reinforcing the contribution of the presence of HSPG Synd-1 for the internalization of crotamine-DNA complex.

Discussion
In this communication we show that the ip injected CPP crotamine is partly reabsorbed by PTECs, and the full-length form of this polypeptide is partly excreted via urine. In vitro experiments with human PTEC cell line HK-2 suggest Synd-1 as one of the contributors for crotamine uptake into these cells. The possibility of oriented delivery of therapeutic compounds to PTECs would be of great interest for the treatment or intervention in different renal diseases with progressive tubulo-interstitial fibrosis such as diabetic kidney disease 36 .
It is worth to consider that after long-term daily administration of crotamine by ip injection for 21 days, crotamine main localization was noticed inside kidneys PTECs, but with no detectable noxious effect, as suggested by the general analysis of tissues (as liver and heart, among others) and cells morphologies by H&E staining. In addition, general biochemical biomarkers (namely creatinine kinase and uric acid) assessments suggested the healthy functioning of kidneys with excretion of the full-length intact crotamine in the urine of animals receiving high doses of this peptide acutely (single dose of 30 µg/animal, 1.5 h before urine collection), although long-term treated mice receiving low doses of this peptide (1 µg/animal/day) did not present detectable amount www.nature.com/scientificreports www.nature.com/scientificreports/ of crotamine in the urine collected daily along the treatment. On the other hand, single ip administration of fluorescently-labeled Cy3-crotamine (5 µg/animal), 2 h before euthanasia, allowed observing its presence in the brush borders area of PTECs (Fig. 3) and in the urine ( Fig. 1 and Supplementary Figs S1 and S2), suggesting that crotamine is reabsorbed from pro-urine by PTECs. Unfortunately, the non-quantitative character of the MALDI-TOF MS 37 did not allow us to quantify the precise amount of crotamine in the urine herein.
The universally expressed isoform of the syndecan, syndecan-4, is known to bind and mediate the transport of the most frequently utilized cationic CPPs, as penetratin, octaarginine and TAT 38 . Various experiments performed www.nature.com/scientificreports www.nature.com/scientificreports/ here indicate the participation of Synd-1 in the crotamine internalization into HK-2 PTECs. Firstly, compared with wild type HK-2 PTEC, crotamine binding is largely diminished in Synd-1 KD cells ( Supplementary Fig. S5). Secondly, the LC 50 concentration of crotamine is higher for the Synd-1 KD cells (Fig. 5). Thirdly, the crotamine cellular uptake and cytoplasmic/perinuclear staining is also reduced in the Synd-1 KD cells (Figs 6, 7).
The crucial roles of the proteoglycans for crotamine internalization was previously shown by us using Chinese hamster ovary (CHO) knockout cells for proteoglycans, namely CHO-745 18 . However, although we have showed earlier that Synd-1 is the dominant HSPG on PTECs both in vivo and in vitro 30 , it is important to consider that the knockdown strategy in the current report via shRNA strategy did not promote the complete suppression of Synd-1 expression 30 , and strong reduced expression (about 80%) but not a complete suppression of Synd-1 expression was observed (see Supplementary Fig. S5). At this point, one also needs to consider that potentially other proteoglycans with the ability to bind to crotamine 18 might be present and eventually over-expressed as compensation in these human PTEC HK-2 cells, possibly ensuring the internalization of this peptide into PTEC cells. Although the suppression of Synd-1 expression did not completely suppress the internalization of crotamine or of crotamine-DNA complexes in Synd-1 KD, the internalization was significantly diminished in this knockdown (KD) cells, as demonstrated here. Also, lower toxicity for crotamine was also observed in the Synd-1 KD cells compared to HK-2 cell line (Fig. 5). It is worth to mention here that the LC 50 for the wild-type human   Cytotoxicity of about 50% of wild type HK-2 cells was observed with ~18 µM of crotamine, while for Synd-1 KD cells, concentrations of ~25 µM of crotamine were required. Cytotoxicity for 90% of the wild-type HK-2 cells was observed with ~33 µM of crotamine, while a concentration higher than ~40 µM of crotamine would be required for Synd-1 KD cells. LC 50 : concentration of drug required to reach 50% of cytotoxicity (dotted line). LC 90 concentration of drug required to determine 90% of cytotoxicity (dashed line).
www.nature.com/scientificreports www.nature.com/scientificreports/ kidney cells was also significantly higher compared to those determined for tumoral cell lines, including murine melanoma B16F10 cells, for which a LC 50 of about 1 µM was described 19,34,39 . Lower cytotoxicity of crotamine against non-tumoral compared to tumoral cells was also previously described by the group 34,39 , and the higher availability of proteoglycans on tumoral cells surface was suggested to explain this specificity of crotamine for highly proliferating cells 18 . Importantly, crotamine is therefore non-toxic for PTECs at working concentrations, which was always below 5 µM in the present study. Also, LC 50 for cytotoxic effect against human HK-2 PTEC cells far exceeds the concentrations needed for gene transfection.
Although we did not show the in vivo delivery of reporter and/or therapeutic genes in the mice kidney PTECs, the absence of any noticeable damage or noxious effects on targeted cells and tissues using the free purified native crotamine, in addition to the successful delivery and expression of reporter gene in HK-2 cells mediated by crotamine showed in the present work, strongly stimulate us to continue this work, aiming the optimization of the conditions for the effective transfection of PTECs in vivo. We also envision therapeutic possibilities for non-viral crotamine-based PTEC-specific inhibition of Rho kinase in acute renal injury or peri-operatively, in renal transplantation. Beneficial effects of a lysozyme-conjugated Rho kinase inhibitor on acute renal allograft rejection were earlier shown 40 .
The specific targeting by crotamine can be used to reach specificities of PTECs such as IL-22R1 renal expression which is exclusive of this cell type, as this interleukin protects against renal I/R injury by activating STAT3 and AKT ameliorating I/R-induced renal inflammation and tubular cell injury 41 . Mitochondrial targeting ability of crotamine 32 may also motivate the exploration of its potential application in mitochondrial dysfunctions in proximal tubule cells, aiming to protect, for instance, from nephrotoxicity associated with progressive tubulo-interstitial disease 42 .
Different therapeutic strategies targeting PTECs are described in literature. Protein-and peptide-based carrier systems, polymeric carrier systems, folate and antibody fragments are examples of PTECs targeting approaches 43 . www.nature.com/scientificreports www.nature.com/scientificreports/ A variety of therapies aims the drug delivery into the luminal side of the tubular cells, as on the apical membranes of these cells are found different internalizing receptors, which are able to internalize small molecules and macromolecules, such as proteins that are filtered into the urine, obtaining efficient drug uptake. Two well-known endocytic receptors are the megalin and cubilin, both expressed on the apical membrane of proximal tubular cells 43 . Molecules with strong affinity toward megalin are used as targeting delivery mediators such as polymyxin B, a polypeptide antibiotic 44 and aminoglycoside antibiotics 45 , but these compounds show nephrotoxicity risk even in therapeutic doses.
During years, conjugate delivery systems were tested as renal-specific delivery alternatives to reduce drug nephrotoxicity such as 2-glucosamine conjugated with prednisolone 46 , carboxylated polyvinylpyrrolidone carrier 47 , conjugation with streptavidin 48 or low-molecular-weight proteins such as lysozyme 49 , but these strategies still present renal toxicity and cardiovascular side effects. At this point, the use of crotamine as alternative drug carrier shows the advantage of showing no nephrotoxicity at therapeutic doses, even if needed for long-term treatments.

Conclusion
Long-term (21 days) continuous (daily) administration of crotamine (1 µg/animal) by ip route in mice was generally non-toxic to the animals, as no significant change could be observed in weight of body or main organs, or bone size of treated animals. In addition, crotamine is also non-toxic for PTECs or kidneys of treated animals at the working concentrations, as no lesions were observed after histopathological analysis with two different staining and also kidney function biomarkers were not altered after daily treatment with crotamine. In addition, it is worth to mention that the cytotoxic concentrations far exceed the concentrations needed for gene transfection. Unprecedentedly, the accumulation of crotamine in kidney, more precisely, in PTECs was confirmed both for acute and long-term administration of crotamine, and we suggest Synd-1 as one of the contributors for crotamine internalization in PTECs, as it is the most abundant proteoglycan present in this cell type. Assays with Synd-1 deficient cells showed lower internalization of crotamine and also lower reporter gene transfection mediated by crotamine compared to wild-type HK-2 cells, reinforcing this hypothesis. Therefore, we propose here crotamine as a powerful tool for non-viral delivery and transport of therapeutic molecules aiming the modulation of aberrant gene and/or protein expression in PTECs, with no important noxious effects, to eventually slow down progressive tubulo-interstitial fibrosis. www.nature.com/scientificreports www.nature.com/scientificreports/

Materials and Methods
Crotamine and other reagents. Crotamine was purified in the Ribeirão Preto Medical School, (FMRP), São Paulo University (USP), by Prof. Dr. Eduardo Oliveira from commercially available Crotalus durissus terrificus venom, basically employing the protocol previously described 20 . Rabbit anti-crotamine antibody was produced and purified by affinity chromatography by Dr. Eduardo Oliveira. Details about specificity are given in Supplementary Method. S4 and Supplementary Fig. S3. All other reagents, when not specified in the text, were of analytical grade and were mainly purchased from Sigma-Aldrich Inc. (St. Louis, MO, USA). The overall yield and purity of crotamine isolated from rattlesnake crude venom was determined by amino acid analysis after acid hydrolysis as previously described 20 . Purified native crotamine was labeled with the Cy3 fluorescent dye (Thermo Fisher Scientific Inc., Waltham, MA, USA), strictly following the instruction of the manufacturer, as described 20 . After labeling, the excess of fluorescent dye was removed by centrifugal 3 K filter unit device (Centricon, Amicon, Millipore Corp, Billerica, MA, USA). The labeling yield was calculated by absorbance measuring accordingly to manufacturer instructions, as described in details by Hayashi and collaborators 20 . (1) control group, which received daily, 100 μL of saline per animal by intraperitoneal (ip) injection; (2) crotamine-treated experimental group which received during 21 days, the purified native crotamine (1 µg/animal/day by ip injection) freshly diluted in 100 μL of saline at the time of injection, following essentially the same protocol previously employed by us for the antitumor therapy 34,35 . Food and water intake were monitored per group of five mice and the data represent the mean value of 24 h consumption assessment. At the end of the treatment (i.e. after 21 days), for the analysis of crotamine localization in tissues, two animals of the treated group received a single shot fluorescently-labeled Cy3-crotamine (5 µg/animal in 100 μL of saline ip injected), while two control mice received 100 µL of saline by ip route, 2 h before the euthanasia. All animals were then weighed using a semi-analytical balance and after animal euthanasia, tissues were removed and weighed, before they were prepared for subsequent histological analysis.
Blood and urine samples were also collected for analysis at the end of 21 days treatment 24 h and 1.5 h, respectively, after last administration of crotamine. In order to detect crotamine in urine samples, two naive mice received a single ip injection of native purified crotamine or fluorescently-labeled Cy3-crotamine (30 µg/animal) while two control mice received 100 µL of saline by ip route, 1,5 h before the urine collection. The total volume of blood collected from each animal (~0.5 mL) was placed in microtubes containing heparin sodium 100 I.U./mL, and the sample was centrifuged for 10 min at 300 × g at room temperature for plasma fraction (supernatant) collection. The urine collected was centrifuged for 10 min, at 2000 × g, at 4 °C, to remove sediments. Aliquots of plasma and urine were stored in 0.2 mL microtubes at −80 °C until analysis.
Biochemical markers analysis in blood and urine. The levels of creatinine in plasma and urine were measured by alkaline picrate method, as described by Jaffé 50 . Levels of uric acid in plasma and urine were measured by Uricase-PAP method 51 . For this purpose, specific kits, namely uric acid (Ácido Úrico Monoreagente -K139) and creatine (Creatinina Enzimática -K161), both from BIOCLIN (São Paulo, SP, Brazil), were used according to the manufacturer's instructions.

sDs-pAGe electrophoresis and Western blotting in urine.
The presence of crotamine in the urine was determined by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS/PAGE), Western blot analysis and mass spectrometry (MS) analysis (Supplementary Method. S1), which also confirmed the integrity of detected crotamine. For SDS-PAGE, 5 µL of urine samples of mice receiving vehicle (control), native crotamine (1 µg/animal/day) for 21 days, or acute administration of native crotamine (30 µg/animal). For the gel loading control, 100 ng of native purified crotamine were used. All samples were mixed with 2 × sample buffer (2% SDS, 10% glycerol, 0.1% bromophenol blue, 50 mM Tris pH 6.8), before denaturation by heating at 95 °C, and separation by electrophoresis in 15% SDS-PAGE, followed by Coomassie blue or silver staining. Another gel was prepared exactly in the same way, before it was transferred to nitrocellulose membrane (Hybond ECL; GE HealthCare, Little Chalfont, UK) for development by Western blot employing the anti-crotamine antibody (1:200) diluted in blocking buffer (3% bovine serum albumin), essentially as previously described 52 . The employed anti-rabbit IgG conjugated with alkaline phosphatase (Promega, Wiscosin, USA) secondary antibody was diluted 1:10,000 in TBST buffer (150 mM NaCl, 20 mM Tris-HCl pH 7.5 and 0.05% Tween-20), and a solution of 5-bromo-4-chloro-3-indolyl phosphate (BCIP) and nitro blue tetrazolium (NBT) in dilution buffer (0.10 M Tris-HCl pH 9.5, 0.02 M NaCl and 0.005 M MgCl 2 ), strictly following the manufacturer's protocols, was employed for the development of the immunorecognized protein bands. Imaging was captured by digital camera SM-G950x (Samsung, Seoul, South Korea) with the following specifications, ISO: 100, exposure time: 1/60 s, and focal length: 4.20 mm.
Histopathological analysis of kidney. Kidneys were fixed in 4% paraformaldehyde before embedment in paraffin. Tissue sections (5-7 μm thick) were obtained from the paraffin-embedded blocks and they were mounted onto glass microscope slides after stretching at 40 °C. After deparaffinization, sections were stained with www.nature.com/scientificreports www.nature.com/scientificreports/ H&E and PAS for routine light microscopy examinations. Sections of kidneys from mice receiving Cy3-crotamine were deparaffinated and only incubated with DAPI followed by embedment with Citifluor TM AF1 Mounting Medium (EMS Acquisition Corp., Pennsylvania, USA), before fluorescence confocal microscopy analysis.
Immunohistochemistry. For the immunohistochemical analysis, tissue sections (3-4 μm thick) were prepared on a microtome from paraffin-embedded blocks, and they were mounted on superfrost glass slides, and dried overnight at 60 °C. Slides were deparaffinized with xylene/ethanol and rehydrated in PBS. Then, an antigen retrieval protocol was performed with Tris-HCl buffer pH 9.0, at 95 °C for 20 min. All further immunohistochemistry steps were carried out at room temperature (25 °C). Endogenous peroxidase was blocked with 0.03% H 2 O 2 in phosphate buffer solution (PBS) for 30 min. To prevent non-specific binding of the antibodies to the tissues, the sections were incubated with 5% normal goat serum in PBS for 30 min. Rabbit anti-crotamine was used as primary antibody (diluted 1:2000 in 1% BSA/PBS), which was incubated on the tissue sections for 1 h. To control for specificity rabbit anti-crotamine in 1% BSA/PBS was blocked with 25 µg/mL of crotamine for 1 h prior to staining procedure. Goat anti-rabbit IgG conjugated with horseradish peroxidase (HRP; DAKO, Glostrup, Denmark) (1:100 in 1% BSA/PBS) was used as secondary antibody, which was incubated for 30 min, before the 10 min incubation in the dark with Tyramide Signal Amplification (TRITC-labeled tyramide solution, tetramethyl rhodamine system, Perkin Elmer, Waltham, MA, USA), employed to show the localization of the HRP-label by immunofluorescence. For nuclear staining, sections were incubated for 10 min in the dark with 300 nM DAPI (4′,6-diamidino-2-phenylindole), and embedment was done by Citifluor TM AF1 Mounting Medium (EMS Acquisition Corp., Pennsylvania, USA) before fluorescence confocal microscopy analysis.
Cell viability assay. Cell viability after exposure to crotamine was examined using the MTT assay, in which metabolically active mitochondrial dehydrogenase activity converts the tetrazolium salt 3-[4,5-dime thylthiazol-2-yl]-2,5-diphenyltetrazolium bromide to insoluble purple formazan crystals at a rate proportional to cell viability. Cultured Synd-1 deficient and wild type HK-2 cells were plated in 96-wells microtiter plates, at a density of 2,500 cells/well, in 200 μL of culture medium (density was established by absorption curve and growth curve evaluation). After overnight incubation at 37 °C, the cells were incubated with crotamine (at concentrations ranging from 0 to 50 μM), in 200 μL of culture medium for 3 days, at 37 °C. At the end of the incubation, 20 μL of MTT solution (5 mg/mL in PBS) was added to each well. After 4 h, the plates were centrifuged at 900 rpm for 15 min, the culture medium was removed and 100 μL of DMSO 100% was added to each well and thoroughly mixed before the plate was read at 520 nm on a FlexStation3 microplate reader (Molecular Devices, Sunnyvale, CA, USA). Percentage of cytotoxicity was calculated using the following equation 100 × (1 − [optical density at 520 nm with crotamine]/[optical density at 520 nm without crotamine]). The assay was performed twice each of them in triplicate.
Crotamine internalization assay. Analysis of crotamine internalization by PTEC HK-2 cells, cultured on glass cover slips, was performed by incubation of the cells with 5 μM of native non-labeled crotamine, at 37 °C for 1 h. After rinsing with PBS, the cells were fixed with 2% paraformaldehyde in PBS for 20 min, at room temperature. All further immunocytochemistry steps were carried out at room temperature (25 °C). Cover slips were mounted and evaluated for fluorescence signals by fluorescence microscopy.

Crotamine transfection assay. Wild-type human HK-2 cells and Synd-1 deficient HK-2 cells (Synd-1 KD)
were plated on 6 × 35 mm well plates using appropriate cell culture media. One day before transfection the cells were plated on glass slides at a density reaching about 50% confluence in 24 h. The crotamine-DNA complexes were prepared with plasmidial vector pEGFP-N1 (Clontech, Mountain View, CA, USA), containing a gene coding for GFP, essentially as previously described 18,20 . On the transfection day, the peptide-DNA complexes solution was added dropwisely to the cultured cells, and the cells with the peptide-DNA complex were incubated for 48 h. Cells were then fixed and stained with DAPI, before analysis in confocal microscopy.
Confocal microscopy analysis. For immunohistochemical, immunocytochemical and transfection analysis, data acquisition was performed in Leica DM4000B fluorescence microscope (Leica Microsystems) or Leica TCS SP8 confocal microscope (Leica Microsystems), using 40× and 63× objectives, Leica TCS SP8 CARS -Coherent Anti-Stokes Raman Scattering (Leica Microsystems, Wetzlar, Germany) using 63× objective or IN Cell Analyzer 2000 (GE Healthcare, Little Chalfont, UK), using 2× objective. The parameters used were λ EX 545 nm and λ EM at 590-620 nm for Cy3-crotamine or TRITC, λ EX 475-495 nm and λ EM at 520-560 nm for GFP and λ EX 405 nm/λ EM 420-470 nm for DAPI fluorescence for microscopes or LED light source with TRITC filter for IN Cell Analyzer.
statistics. Data were analyzed using Student's t-test comparing two groups (namely control and crotamine-treated groups). Results are presented as mean ± SEM with p < 0.05 considered as statistically significant.