Novel Anthracycline Utorubicin for Cancer Therapy

Abstract Novel anticancer compounds and their precision delivery systems are actively developed to create potent and well‐tolerated anticancer therapeutics. Here, we report the synthesis of a novel anthracycline, Utorubicin (UTO), and its preclinical development as an anticancer payload for nanocarriers. Free UTO was significantly more toxic to cultured tumor cell lines than the clinically used anthracycline, doxorubicin. Nanoformulated UTO, encapsulated in polymeric nanovesicles (polymersomes, PS), reduced the viability of cultured malignant cells and this effect was potentiated by functionalization with a tumor‐penetrating peptide (TPP). Systemic peptide‐guided PS showed preferential accumulation in triple‐negative breast tumor xenografts implanted in mice. At the same systemic UTO dose, the highest UTO accumulation in tumor tissue was seen for the TPP‐targeted PS, followed by nontargeted PS, and free doxorubicin. Our study suggests potential applications for UTO in the treatment of malignant diseases and encourages further preclinical and clinical studies on UTO as a nanocarrier payload for precision cancer therapy.


Introduction
Anthracyclines,s uch as doxorubicin (DOX) and daunorubicin, have been among the most widely used cancer therapeutics for more than 30 years. [1]Mechanistically,t he four-ring anthraquinone backbone of anthracyclines intercalates DNAa nd the sugar moiety bound through ag lycosidic linkage interacts with base pairs in the minor groove. [2]These drug interactions interfere with DNAs ynthesis,r epair, and transcription, resulting in inhibition of cell replication and, eventually,cell death. [3,4] omplexation of anthracyclines with DNAalso inhibits the activity of topoisomerase II, an enzyme that manages DNAt angles and supercoils,i mpairing DNA repair. [5]In addition, the quinone moiety in anthracyclines induces the formation of reactive oxygen species. [6,7]Anthracyclines are used to treat many types of cancers,i ncluding leukemia, lymphomas,b reast, gastric,u terine,o varian, bladder,a nd lung cancers.R ecognition that therapeutic application of anthracyclines is limited by side effects,e specially by dose-limiting cardiotoxicity, [1] has led to extensive efforts to develop anthracycline derivatives and their nanoformulations of improved safety profile and broader therapeutic index.For example,s ynthetic 9-aminoanthracycline amrubicin, which was designed for reduced cardiotoxicity, [8,9] has higher antitumor activity than DOX. [10]Anthracyclines have been covalently coupled to nanocarriers (e.g.p olymeric, gold, silver,silica, and graphene oxide nanoparticles) using linkers sensitive to physiological stimuli such as modulation of surrounding pH and redox potential, or enzymatic processing. [5]Unmodified anthracyclines have been encapsulated in polymeric, lipidic and inorganic nanoparticles and in dendrimers. [11]DOXn anoformulated in PEGylated-liposomes (Caelyx TM ), developed in the late 1980s,isclinically approved for the treatment of multiple myeloma, Kaposi sarcoma, and breast and ovarian carcinoma. [1,12]However,compared to the parental free drug, the anticancer efficacyo fl iposomal doxorubicin is not significantly higher [13,14] and, despite improved safety profile,i ts till elicits side effects such as dermal toxicities and mucositis. [15]hetumor accumulation of systemic drug nanocarriers is driven by structural and functional abnormalities in tumorfeeding blood and lymphatic vessels that manifest in ap henomenon known as the enhanced permeability and retention (EPR) effect. [16]However,t he EPR effect is affected by the tumor location, type,a nd stage,a nd it shows wide inter-a nd intratumoral heterogeneity. [17]Affinity targeting with ligands such as peptides and antibodies may allow for more efficient and uniform delivery of payloads to the tumor vessels and parenchyma than is possible by solely relying on EPR effect.[20][21] TheL inTT1 (AKRGARSTA) is aT PP first recruited to p32 overexpressed on the surface of vascular and malignant cells and on tumor-associated macrophages in many solid tumors, [22][23][24] followed by proteolytic processing to expose the C-end Rule (CendR) motif of the peptide (AKRGAR, CendR motif underlined) to enable neuropilin-1( NRP-1) binding, which activates at umor penetration pathway. [25]LinTT1 has been used for targeting therapeutic nanoparticles in the preclinical treatment of breast cancer, [26] peritoneal carcinomatosis, [27] and glioblastoma [28] as well as for delivery of positron emission tomography (PET)-active nanoparticles for early detection of triple-negative breast tumors in mice. [29]ere,i naquest to develop improved anthracyclines suitable for nanocarrier-based delivery,wedeveloped anovel 9-aminoanthracycline prodrug,UTO.W ereport studies on its synthesis,c omparative toxicity profiling on cultured malignant cells,aswell as comparison of uptake and activity of free versus nanoformulated UTO.

Synthesis of UTO
9-Aminoanthracyclines have shown less cardiotoxicity and more controllable toxic effects than other anthracyclines such as DOX, daunorubicin and idarubicin. [8,9] nthracyclines undergo enzymatic reduction of the C-13 that is catalyzed by ac ytoplasmic carbonyl reductase and converts ac arbonyl to ah ydroxyl group. [30]In the case of the 9aminoanthracycline amrubicin, the corresponding metabolite amrubicinol is up to 50 times more potent than the parent drug. [31]Thec ytotoxic effect can be further increased by attaching am ethylene group to the amino and hydroxyl groups (between respective nitrogen and oxygen atoms) of the 1,2-aminoalcohol moiety of the daunosamine. [32]The resulting oxazolidine cycle forms an anthracycline-DNA adduct that inhibits cell replication. [33]ere we developed an ovel 9-aminoanthracycline prodrug, UTO, that contains ah ydroxyl group at the C-13 position and an oxazolidine cycle in C-3',C -4' of the daunosamine moiety (Scheme 1A).Thea ctive drug (compound 6)was obtained following afour-step procedure and its oxazolidine cycle was subsequently protected to form the prodrug UTO (compound 8).Thes ynthesis started from commercially available amrubicinone (aglycone part of amrubicin, 1).Amrubicinone was glycosylated with protected aminosugar L-daunosamine (1,4-di-O-acetyl-N-trifluoroacetyl-b-L-daunosamine,c ompound 2)int he presence of trimethylsilyl trifluoromethanesulfonate to obtain compound 3.
Thecarbonyl group of compound 3 was reduced with sodium triacetoxyborohydride to yield compound 4.Inthis step,other reduction conditions were tested (NaBH 4 ,N aBH 3 CN) with unsuccessful results.T he low yield of compound 4 could be partially due to compound 3 being not fully pure.Compound 4 was deprotected by cleavage of the N-trifluoroacetyl and Oacetyl groups from the L-daunosamine moiety to obtain compound 5.T he oxazolidine cycle was then formed by reacting compound 5 with paraformaldehyde.T he unreacted compound 5 was separated by filtration and the solution was concentrated and triturated with diethyl ether to obtain compound 6 (yield % 48 %) that was used without further purification to prepare the prodrug.Compound 6w as characterized by 1 H-NMR, 13 C-NMR, and 13 C-DEPT studies (Figure S7-S10).
Theo xazolidine cycle of compound 6 is unstable in aqueous media and undergoes fast hydrolysis under acidic conditions.T herefore,i tw as stabilized with ab iocleavable protecting group [34] by the formation of ac arbamate bond through the nitrogen atom in the oxazolidine cycle.The acetyl group from the acetyloxymethyl carbamate protecting group is susceptible to esterases,f ollowed by spontaneous decomposition of the resulting hemiacetal, resulting in exposure of reactive oxazolidine cycle (Scheme 1B), which is able to form an anthracycline-DNAadduct. [33] Carboxylesterases (CE) are widely distributed in the body and also found in tumors. [35]hea bility of CE to hydrolyze ester,a mide,a nd carbamate groups to alcohol and carboxylic acid has been extensively described [36,37] and is used for the activation of various antiviral, anticancer,a nd antibiotic prodrugs.F or example, pentyl PA BC-Doxaz prodrug is hydrolyzed by carboxylesterase 2( CES2) to the active formaldehyde-Dox conjugate. [38]hecleavage occurs via serine hydrolase mechanism followed by three spontaneous steps,e nding in as econd decarboxylation that, similarly to UTOh ydrolysis,r esults in the free oxazolidine cycle.
Thea nalog of amrubicin containing oxazolidine cycle (compound 6)was reacted with 4-nitrophenyl(acetyloxy)methylcarbonate (Scheme 1A)t oo btain compound 7.U nder these conditions,o nly the oxazolidine cycle in the sugar moiety is available for reaction as the nitrogen atom at C9 is highly sterically hindered.Compound 7 was not isolated and the reaction was mixed with acetic acid, and stirred to hydrolyze the unreacted oxazolidine cycle.T he final purified prodrug (UTO, compound 8)was obtained with an 18 %yield (with respect to compound 6)a t> 96 %p urity (Figure S15) and was characterized by NMR (Figure S11-14) and HMRS (Figure S16).Thel ow yield of 8 might be due to partial hydrolysis of the oxazolidine in the sugar before reaction with 4-nitrophenyl-(acetyloxy)-methylcarbonate,yielding the analogue of compound 8 without oxazolidine cycle.T he optimization of the synthesis steps that resulted in low yields (from compound 3 to 4 and from 6 to 8)w ill be as ubject of follow-up studies.

Cytotoxicity of UTO on Cultured Cancer Cell Lines
We first performed comparative studies on the effect of treatment with UTOa nd DOX on the viability of apanel of cultured cell lines:U 937 monoblast-like human histiocytic lymphoma cells,Jurkat E6.1 human T-lymphocyte from acute T-cell leukemia cells,A 549 human lung carcinoma cells,a nd HT-29 human colorectal adenocarcinoma cells.T he cytotoxicity was tested after 30 and 90 min of incubation of the cells with the drugs and follow-up culture for 24 h( U937 and Jurkat E6.1 cells in suspension), or 72 h( adherent A549 and HT-29 cells).In all tested cell lines,U TO had am ore potent effect on reducing cell viability than DOX( Figure 1).The highest difference in the IC 50 between the UTOand DOXwas observed for A549 (with UTO % 16-fold more effective) and the lowest for HT-29 cells (with UTO % 2-fold more effective).

Generation of Peptide-Guided UTO-Polymeric Nanocarriers
Recognition that hydrophobicity of UTO may limit its therapeutic application as af ree drug, prompted us to evaluate it as an anticancer payload of peptide-guided polyethylene glycol-polycaprolactone (PEG-PCL) polymersomes (PS).We first optimized the density of the prototypic CendR peptide RPARPAR (RPAR) on the surface of PS for targeting of NRP-1 + cells.T he density of RPAR on the PS surface was modulated by varying the content of maleimide-PEG-PCL relative to the whole amount of copolymer between 0a nd 20 %.As the peptide conjugation occurs through the formation of athioether bond between the thiol group of the cysteine added to the N-terminus of the RPAR peptide (Cys-RPAR) and the maleimide group of the copolymer, the number of surface maleimide groups determines the peptide density on PS nanoparticles.P Sw ere prepared by film hydration method using aprotocol adapted from ap revious study. [29]PS contained 5% of FAM-labeled-PEG-PCL (FAM-PEG-PCL) for fluorescence-based tracing.Thehydrodynamic diameter of the PS measured by dynamic light scattering (DLS) (105 nm) and the polydispersity index (0.19) were similar for all vesicle preparations independent of peptide coating and density (Figure 2A).Tr ansmission electron microscopy (TEM) showed that FAM-labeled RPAR-PS (RPAR-FAM-PS) appear as homogeneous spherical vesicles (Figure 2B).We next studied the release of UTOf rom polymersomes in PBS during incubation at 37 8 8C.After 48 h, less than 2% of UTOwas released from the PS (Figure S19).
To study the effect of peptide coating on cellular uptake of PS,w ei ncubated FAM-labeled PS with cultured PPC-1 and M21 cells-a well-established system that has been used for CendR peptide based affinity targeting in vitro studies. [40,41]PC-1 human primary prostate cancer cells have elevated expression of NRP-1 (Figure S20).In contrast, M21 human melanoma cells are negative for NRP-1 expression (Figure S20). [39,40] PC-1 and M21 cells were incubated with RPAR-FAM-PS for 1h,w ashed, detached, and analyzed by flow cytometry for the FAMp ositivity.I nP PC-1 cells,a n increase in surface density of RPAR peptide resulted in ap rogressive increase in binding of RPAR-FAM-PS (Figure 2C).RPAR-PS formed using 20 %maleimide-PEG-PCL showed the highest uptake,w ith nearly all cells positive for FAMf luorescence.I nc ontrast, the binding of RPAR-FAM-PS to the M21 cells was low and not affected by the surface density of the RPAR peptide.T he zeta potential of all PS samples was neutral (Figure 2A), suggesting that the differential binding of PS was not due to differences in PS surface charge.I nl ine with flow cytometry studies,c onfocal microscopy demonstrated the presence of the green signal representing RPAR-FAM-PS only in PPC-1 and not in M21 cells (Figure 2D).Theh ighest RPAR-FAM-PS signal was associated with PPC-1 cells incubated with PS assembled using 20 % maleimide-PEG-PCL.Already at 1htime point, the RPAR-FAM-PS was predominantly seen in vesicular structures in the cytoplasm, indicating cellular uptake (Figure 2D,arrows).

Cytotoxicity of UTO-Loaded RPAR-Targeted PS on Cultured Cancer Cells
We next performed acomparative evaluation of the effect of free UTOand DOX, as well as RPAR-guided and control UTO-PS on the viability of PPC-1 and M21 cells.T he cells were treated for 30 min with the compounds,followed by 48 h follow-up incubation in fresh culture medium and MTT viability assay.I nb oth cell lines,f ree UTOw as more toxic than free DOX.(Figure 3C and D).Forn ano-encapsulated UTO, the cytotoxic effect on cultured cells was dependent on targeting peptide functionalization and on the expression of peptide receptor,N RP-1.RPAR-guided UTO-PS (at 2 mM UTO) showed significant potentiation of antiproliferative effect compared to nontargeted UTO-PS at the same drug loading ( % 40 %v s. % 70 %c ell viability;F igure 3C and E).

In Vivo Homing of TPP-PS
We next studied in vivo biodistribution of systemic LinTT1-, RPAR-, and nontargeted PS loaded with nearinfrared DiR dye (LinTT1-DiR-PS,RPA R-DiR-PS,and DiR-PS) as am odel imaging payload that mimics UTO in being hydrophobic and could be used for initial calibration before using the actual drug.TheDiR-loaded PS appeared spherical (Figure S22) with an average hydrodynamic diameter similar to the previous PS formulations (average size:1 16 nm, PDI % 0.15), suggesting that the presence of dye in the PS membrane does not affect the structure of the nanovesicles.Fort umor induction, the mice were orthotopically injected with human MCF10CA1a triple-negative breast cancer (TNBC) cells.M CF10CA1a cells and activated stromal cells in MCF10CA1a tumor lesions overexpress surface p32, [42] thus rendering LinTT1 (a ligand for p32) as uitable peptide to target this tumor.M oreover,f unctionalization of therapeutic nanoparticles with LinTT1 peptide was found to improve their anticancer efficacy in experimental therapy of MCF10CA1a breast tumor. [26]LinTT1-DiR-PS,R PA R-DiR-PS,and DiR-PS were injected intravenously (i.v.) into TNBC mice and live fluorescence imaging was performed at 1, 3, 6, 24, and 48 hpost-injection.Targeting with LinTT1 and RPAR peptides increased tumor homing of DiR-PS (Figure 4).LinTT1-and RPAR-DiR-PS were detected in tumors already at 3h after administration (arrows in Figure 4A), whereas nontargeted DiR-PS became detectable after 24 hp ostinjection.Compared to nontargeted DiR-PS,t he area under the curve (AUC) in tumor lesions at 24 hfor peptide-targeted DiR-PS was elevated by % 42 %( for LinTT1-DiR-PS) and % 25 %( for RPAR-DiR-PS) (Figure 4C).As expected, the LinTT1-, RPAR-, and non-targeted DiR-PS were also observed in the organs of the reticuloendothelial system: the liver (arrowheads in Figure 4A)a nd the spleen.Fortyeight hours after PS injection, tumors and control tissues were excised, sectioned, and the microscopic localization of PS was analyzed.We observed the accumulation of LinTT1-DiR-PS in the extravascular tumor parenchyma (Figure 5A)and their co-localization with the known receptors of LinTT1 peptide, p32, and NRP-1 (Figure 5C).We studied the co-localization of LinTT1-DiR-PS with the CD206 receptor expressed on M2 protumoral macrophages known to promote growth and metastasis of malignant cells in solid tumors. [43]We observed that LinTT1-DiR-PS accumulate in M2 macrophages in the

Angewandte
Chemie Forschungsartikel 17160 www.angewandte.detumor (Figure 5A).Cardiotoxicity is the dose-limiting toxicity of anthracyclines and we studied the accumulation of LinTT1-DiR-PS in the heart of TNBC-bearing mice.A s shown in Figure 5A and B, the myocardial tissue contained as ignificantly lower signal of the LinTT1-DiR-PS than the malignant tissue.

Tumor Accumulation of UTO Loaded in PS
Having established conditions for enhanced tumor accumulation of LinTT1-targeted PS,w en ext studied the biodistribution of UTO loaded in PS.L inTT1-targeted and nontargeted PS loaded with UTO( LinTT1-UTO-PS,U TO-PS), and free DOX were i.v.i njected into mice bearing orthotopic MCF10CA1a tumors.D OX served as as urrogate for free UTO, which could not be used because of its poor water solubility.
As shown in Figure 6A and B, there was as ignificantly higher tumor accumulation of UTO in mice injected with LinTT1-UTO-PS in comparison with other samples.U TO signal co-localized with CD31-positive blood vessels;i n addition, some signal was seen in the perivascular space with LinTT1-UTO-PS,b ut not with UTO-PS,s uggesting that the drug loaded in the LinTT1-PS had extravasated and spread into the tumor tissue (Figure 6A,inset).Importantly,noUTO signal was observed in the heart of the LinTT1-UTO-PS treated mice (Figure 6A).These observations suggest that the encapsulation of UTOi nL inTT1-PS enhances the specific tumor accumulation of the drug.

Discussion
Anthracycline drugs are among the most widely used antitumor drugs.H owever,t heir side effects,s uch as cardiotoxicity,b one marrow depression, and gastrointestinal disturbances,l imit their therapeutic application at doses sufficient to eradicate malignant lesions.H ere we report the synthesis of novel anthracycline,U TO,a nd show that it is more effective in killing cultured cancer cells than the widely used DOX.Furthermore,w es how that UTO, which is too hydrophobic to be used in vivo as such, can be loaded with high efficiency in polymersome nanocarriers (PS), and that UTO-PS functionalized with homing peptides accumulate selectively in cultured tumor cells expressing the receptor for the peptide and, upon systemic administration, in tumor lesions in vivo.
UTOw as designed to achieve improved anticancer efficacyw ith lower side effects and good tolerability.S imilar to amrubicinol, the reduced metabolite of amrubicin (the first synthetic anthracycline clinically approved for the treatment of lung cancer), UTOc ontains an amino group in the 9position of the anthraquinone and ahydroxyl group in the C-13 position.Whereas both UTOand DOX include adaunosamine moiety,i nU TO it contains an oxazolidine cycle that can facilitate the formation of DNAa dducts. [32,33,44] he acetyloxymethyl carbamate group of UTOi sh ydrolyzed by esterases to expose the reactive oxazolidine cycle.T he fourring structure of UTOi ntercalates the DNAa nd the oxazolidine cycle binds via its methylene carbon of guanine residues to block DNA-replication and transcription processes.A mrubicin and amrubicinol, which induce cell growth

Angewandte
Chemie Forschungsartikel 17162 www.angewandte.deinhibition by stabilizing the topoisomerase II-DNAcomplex, are less potent DNAintercalators than DOX. [45]bout 40 %o fa pproved drugs and nearly 90 %o fd rug candidates are poorly soluble in aqueous solutions.W hereas hydrophobicity of free UTOp oses af ormulation challenge and is likely to limit its bioavailability,w eh ypothesized that its nanoformulation can bypass these problems.S ince ap ioneering 1984 study on nanosized polymeric self-assemblies as hydrophobic drug solubilizers, [46] nano-delivery systems have been used to improve solubility of different hydrophobic drugs,s uch as paclitaxel.Abraxane ,apaclitaxel albuminbound NP formulation with ap article size of % 130 nm, was approved by the FDAin2005 for the treatment of metastatic breast cancer. [47]In the current study,weshowed that PS can be used to formulate hydrophobic UTOf or systemic application without the use of toxic solvents such as Cremophor EL.Besides serving as af ormulation aid for hydrophobic drugs,P Sa re well suited for their precision delivery.F irst, PEG-PCL PS are fully biocompatible. [48][51] Indeed, we observed improved homing of systemic TPP-functionalized PS in the breast tumors.W eo bserved ap rominent homing of LinTT1-PS in the M2-differentiated tumor-associated macrophages (M2-TAMs), in particular at the tumor rim.Thed epletion of protumoral M2-TAMs with DOX-loaded nanoparticles has been reported to result in suppression of tumor growth. [52]oreover,d rug-loaded TAMs act as ad rug reservoir that releases the drug over extended periods of time. [53]Therefore, targeting M2-TAMs with UTO-loaded PS functionalized with LinTT1 (or other M2 TAMs pecific peptides [54,55] )m ay become another anticancer therapeutic strategy.
Interestingly,a fter systemic administration of LinTT1-UTO-PS in breast tumor mice,n oU TO signal was observed in the heart.In reported biodistribution studies using amrubicin, DOX, daunorubicin, and their reduced 13-hydroxy metabolites,i twas found that amrubicinol is more selective for tumors than the rest of anthracycline metabolites. [56]Therefore,t he lower cardiotoxicity of amrubicin might be ac onsequence of the lower heart accumulation of its active metabolite.T he intrinsic tumor selectivity of 13hydroxy anthracyclines in combination with the precision nano-delivery of the drug may result in ap otent antitumor activity and low side effects.T he therapeutic activity of the UTO-loaded TPP-PS will be as ubject of follow-up studies.

Conclusion
We synthesized an ew anthracycline,u torubicin (UTO), with higher cell-killing activity than DOX on cultured tumor cell lines.T od eal with the limitations posed by the hydrophobicity of UTO, we developed protocols for UTOe ncapsulation in peptide-guided, biocompatible,and biodegradable polymersome nanoparticles.T he nanoformulated UTOfunctionalized with tumor targeting peptides showed selective internalization and killing of cultured peptide-receptor positive cells and accumulation in tumors upon systemic administration in vivo.

Figure 1 .
Figure 1.Cytotoxicity of UTO on cultured cancer cell lines using MTS assay.Viability of the cancer cells incubated with indicated concentrations of UTO and DOX for 30 (A) or 90 (B) min and chase in drugfree medium.IC 50 (half-maximal inhibitory concentration) was determined using log(inhibitor)v ersus response-variable slope (four parameters) model (GraphPad Prism version 5.03 for Windows, Graph-Pad Software, La Jolla CaliforniaUSA).The standard error of the mean (SEM) of the IC 50 is shown.The cells were incubated with the anthracyclines, washed, and cultured for 24 h(U937 and Jurkat E6.1 cells in suspension) or 72 h(adherentA549 and HT-29 cells) followed by the viability assay.Asc ontrols, 1% water,or1%D MSO (solvents used to dissolveD OX and UTO respectively)v /v in cell culture medium were used.MTS assay results were plotted as acurve of the percentageo fviable cells (taking the viability of the nontreated control cells as 100 %) versus log concentration of tested sample.Error bars represent the mean standard deviation.N = 3.

Figure 2 .
Figure 2. Optimization of PS for peptide-guided delivery.A)Characterizationo fPSover arange of maleimide-PEG-PCL (Mal-PEG-PCL) (0;2;5; 10 and 20 %): DLS graphs and characteristics of size, PDI, and Z-potential of the PS.B) TEM of the FAM-PS.Scale bar = 100 nm.C) Binding of RPAR-FAM-PS to cultured PPC-1 and M21 cells.Attached cells were incubated with PS samples for 1h,w ashed, detached,a nd flow cytometry was used to quantifyt he binding of RPAR-FAM-PS to PPC-1 (NRP-1 + )a nd M21 (NRP1 À )cells.PS were prepared incorporating the indicated percentageo fMal-PEG-PCL.T he graph represents the percentage of labeled cells.Control is the cells without incubation with PS.N = 3. Error bars indicate AE standard error of the mean (AE SEM).D) Fluorescenceconfocal microscopyi maging of attached PPC-1 and M21 cells incubated with RPAR-FAM-PS samples for 1h.T he PS were labeled with FAM (green signal) and the nuclei were stained with DAPI (blue signal).Scale bar = 50 mm.

Figure 3 .
Figure 3.Effect of free and nano-encapsulated UTO on the viability of culturedc ells with different NRP-1 expression status.A) TEM imaging of RPAR-UTO-PSa nd UTO-PS.S cale bar = 100 nm.B) Size distribution, UTO concentration, and encapsulatione fficiencyo fPSsamples.C, D) Percentage of viable PPC-1 (C) and M21 cells (D) after 30 min incubationw ith PS formulationsat0 .02,0.2, 2, and 20 mMo fUTO and 48 h follow-up incubation.E) Percentage of viable PPC-1 and M21 cells after 30 min treatment with PS formulationsa t2mMofU TO and 48 hf ollowup incubation.Pvalues for the viable PPC-1 and M21 cells after 30 min treatment with PS formulationsat2mMofUTO or DOX are shown in Table 1i nt he Supporting Information.Error bars indicate AE SEM;**** p < 0.0001, N = 5.

Figure 4 .
Figure 4. Systemic peptide-targeted PS home to MCF10CA1a breast tumors.A) Live imaging of MCF10CA1a tumor-bearing mice injected with LinTT1-DiR-PS (LinTT1), RPAR-DiR-PS (RPAR), or non-targeted DiR-PS (PS) at indicated post-administration time points.The signal of DiR-PS in the tumor is pointed out with arrows and in the liver with arrowheads.B)Quantitation of accumulation of DiR-labeled PS in the tumor tissue at different post-injection time points.The fluorescence signal was quantified from the IVIS images.For1,3,6,and 24 htime points N = 3; for 48 h RPAR-DiR-PS N = 1, for LinTT1-DiR-PS and DiR-PS N = 2. Error bars indicate + SEM.C) The area under the curve (AUC) in the tumors of LinTT1-DiR-PS, RPAR-DiR-PS, and DiR-PS at 24 hp ost-injection.N = 3, error bars = AE SEM, ** p < 0.01, * p < 0.05, N.S. = not significant.